Compare commits

..

5 Commits

Author SHA1 Message Date
abeatrix e9bef4253a add tests 2026-02-04 12:33:11 +08:00
abeatrix 2a1c3fd9ba update fetch to work with both bun and node fetch 2026-02-04 12:10:02 +08:00
abeatrix 5ac991902b fix scripts 2026-02-04 12:05:02 +08:00
abeatrix c40c7c098c update bun lock 2026-02-04 12:02:27 +08:00
abeatrix b62c7874f6 chore: migrate from npm to bun as package manager
Replace all npm commands with bun equivalents across the entire codebase including:
- CI/CD workflows (GitHub Actions)
- Setup scripts and hooks
- Documentation and guidelines
- Test scripts and coverage checks
- Package.json scripts
- Pull request templates

This migration improves installation speed and reduces dependency resolution time while maintaining compatibility with existing workflows.
2026-02-04 11:27:56 +08:00
647 changed files with 32390 additions and 66687 deletions
-5
View File
@@ -1,5 +0,0 @@
---
"cline": patch
---
docs: remove stale Claude 5 mention from auto compact model support list
+5
View File
@@ -0,0 +1,5 @@
---
"cline": patch
---
fix(cli): prevent hang when spawned without TTY
+5
View File
@@ -0,0 +1,5 @@
---
"claude-dev": patch
---
Fix decimal input crash in OpenAI Compatible price fields (#8129)
+5
View File
@@ -0,0 +1,5 @@
---
"cline": patch
---
Supports rendering markdown table in chat view.
+5
View File
@@ -0,0 +1,5 @@
---
"claude-dev": patch
---
fix: build complete handlers when upadting the api config
+5
View File
@@ -0,0 +1,5 @@
---
"claude-dev": patch
---
Updating script documentation and removing unnecessary continue on error
-5
View File
@@ -1,5 +0,0 @@
---
"cline": patch
---
Remove the expired MiniMax M2.5 free promo, restore GLM 5 free promo in What's New, and keep Sonnet 4.6 free messaging at 24 hours.
+5
View File
@@ -0,0 +1,5 @@
---
"claude-dev": patch
---
Fixed missing provider from list
+5
View File
@@ -0,0 +1,5 @@
---
"claude-dev": patch
---
feat(skills): Make skills always enabled and remove feature toggle setting
+5
View File
@@ -0,0 +1,5 @@
---
"claude-dev": patch
---
Fixed Favorite Icon / Star from getting clipped in the task history view
+2 -2
View File
@@ -41,11 +41,11 @@ fi
# Install project dependencies
echo "Installing dependencies..."
npm run install:all
bun run install:all
# Generate gRPC/protobuf types (required for TypeScript)
echo "Generating proto types..."
npm run protos
bun run protos
echo ""
echo "Session setup complete!"
+5 -20
View File
@@ -147,29 +147,14 @@ When filling out the template:
### Create PR with gh CLI
**Use a temporary file for the PR body** to avoid shell escaping issues, newline problems, and other command-line flakiness:
1. Write the PR body to a temporary file:
```
/tmp/pr-body.md
```
2. Create the PR using the file:
```bash
gh pr create --title "PR_TITLE" --body-file /tmp/pr-body.md --base main
```
3. Clean up the temporary file:
```bash
rm /tmp/pr-body.md
```
For draft PRs:
```bash
gh pr create --title "PR_TITLE" --body-file /tmp/pr-body.md --base main --draft
gh pr create --title "PR_TITLE" --body "PR_BODY" --base main
```
**Why use a file?** Passing complex markdown with newlines, special characters, and checkboxes directly via `--body` is error-prone. The `--body-file` flag handles all content reliably.
Alternatively, create as draft if the user wants review before marking ready:
```bash
gh pr create --title "PR_TITLE" --body "PR_BODY" --base main --draft
```
## Post-Creation
-11
View File
@@ -147,17 +147,6 @@ Required steps:
Common mistake: Adding only the return value without the `context.globalState.get()` call. This compiles but the value is always `undefined` on load.
Settings plumbing gotcha: if a key is user-toggleable from settings, wire both controller update paths:
- `src/core/controller/state/updateSettings.ts` for webview `updateSetting(...)`
- `src/core/controller/state/updateSettingsCli.ts` for CLI/ACP settings updates
Missing one path causes a toggle to appear to change in one surface while the backend state stays unchanged.
Webview toggle gotcha: settings changes must also round-trip back in state payloads.
- Add the field to `UpdateSettingsRequest` in `proto/cline/state.proto` (for webview update requests), then run `npm run protos`
- Include the key in `Controller.getStateToPostToWebview()` (`src/core/controller/index.ts`)
- Ensure `ExtensionState` and webview defaults include the key (`src/shared/ExtensionMessage.ts`, `webview-ui/src/context/ExtensionStateContext.tsx`)
If this round-trip wiring is missing, the backend value can update but the toggle in webview appears stuck or reverts.
## StateManager Cache vs Direct globalState Access
StateManager uses an in-memory cache populated during `StateManager.initialize(context)` in `common.ts`. For most state, use `controller.stateManager.setGlobalState()`/`getGlobalStateKey()`.
+1 -1
View File
@@ -42,7 +42,7 @@ Here, we use the common `StringRequest` and `KeyValuePair` types.
After editing a `.proto` file, regenerate the TypeScript code. From the project root, run:
```bash
npm run protos
bun run protos
```
This command compiles all `.proto` files and outputs the generated code to `src/generated/` and `src/shared/`. Do not edit these generated files manually.
+1 -1
View File
@@ -98,7 +98,7 @@ On the main branch, create a commit that updates:
Each changeset file in `.changeset/` corresponds to a PR. Read them to identify which ones belong to the commits you're hotfixing, then delete those files.
**Skip running `npm run install:all`** - the automation handles outdated lockfiles.
**Skip running `bun run install:all`** - the automation handles outdated lockfiles.
Commit with message format: `v{VERSION} Release Notes (hotfix)`
-49
View File
@@ -1,49 +0,0 @@
# THIS IS AUTOGENERATED. DO NOT EDIT MANUALLY
version = 1
name = "cline"
[setup]
script = '''
if [ ! -d "node_modules" ]; then
MAIN_WORKTREE="$(git worktree list | head -n1 | awk '{print $1}')"
ln -s "$MAIN_WORKTREE/node_modules" node_modules
ln -s "$MAIN_WORKTREE/webview-ui/node_modules" webview-ui/node_modules
fi
'''
[[actions]]
name = "VS Code"
icon = "run"
command = "chmod +x ./scripts/run-extension-host.sh && ./scripts/run-extension-host.sh production"
[[actions]]
name = "CLI"
icon = "run"
command = '''
npm run cli:build
npm run cli:run
'''
[[actions]]
name = "npm install"
icon = "tool"
command = '''
rm node_modules
rm webview-ui/node_modules
npm run install:all
'''
[[actions]]
name = "pull main"
icon = "tool"
command = '''
git fetch origin main
if ! git merge-base --is-ancestor main origin/main; then
echo "Local main has commits not on origin/main. Aborting..."
exit 1
fi
git update-ref refs/heads/main refs/remotes/origin/main
echo "main updated to $(git rev-parse --short main)"
'''
+3 -2
View File
@@ -1,2 +1,3 @@
/.github/ @saoudrizwan @arafatkatze @maxpaulus43 @candieduniverse
/README.md @saoudrizwan @juanpflores
/docs/
/.github/ @saoudrizwan @garoth @sjf
/README.md @saoudrizwan @nickbaumann98
+2 -2
View File
@@ -59,8 +59,8 @@ We're not looking for exhaustive documentation - just evidence that you've thoug
<!-- Put an 'x' in all boxes that apply -->
- [ ] Changes are limited to a single feature, bugfix or chore (split larger changes into separate PRs)
- [ ] Tests are passing (`npm test`) and code is formatted and linted (`npm run format && npm run lint`)
- [ ] I have created a changeset using `npm run changeset` (required for user-facing changes)
- [ ] Tests are passing (`bun test`) and code is formatted and linted (`bun run format && bun run lint`)
- [ ] I have created a changeset using `bun run changeset` (required for user-facing changes)
- [ ] I have reviewed [contributor guidelines](https://github.com/cline/cline/blob/main/CONTRIBUTING.md)
### Screenshots
+3 -3
View File
@@ -62,9 +62,9 @@ class TestCoverage(unittest.TestCase):
# Use xvfb-run on Linux
if sys.platform.startswith('linux'):
cmd = f"cd {root_dir} && xvfb-run -a npm run test:coverage > {cls.extension_coverage_file} 2>&1"
cmd = f"cd {root_dir} && xvfb-run -a bun run test:coverage > {cls.extension_coverage_file} 2>&1"
else:
cmd = f"cd {root_dir} && npm run test:coverage > {cls.extension_coverage_file} 2>&1"
cmd = f"cd {root_dir} && bun run test:coverage > {cls.extension_coverage_file} 2>&1"
log("Running extension tests...")
log(f"Command: {cmd}")
@@ -73,7 +73,7 @@ class TestCoverage(unittest.TestCase):
# Run webview tests with coverage
log("Running webview tests...")
cmd = f"cd {webview_dir} && npm run test:coverage > {cls.webview_coverage_file} 2>&1"
cmd = f"cd {webview_dir} && bun run test:coverage > {cls.webview_coverage_file} 2>&1"
log(f"Command: {cmd}")
result = subprocess.run(cmd, shell=True, check=False, capture_output=True, text=True)
log(f"Webview tests exit code: {result.returncode}")
+173
View File
@@ -0,0 +1,173 @@
name: Claude Issue Triage
on:
issues:
types: [opened]
# Manual trigger for backfilling existing issues. Run from terminal:
# gh workflow run claude-issue-triage.yml -f issue_number=1234
# Or batch process:
# gh issue list --state open --limit 10 --json number --jq '.[].number' | while read num; do
# gh workflow run claude-issue-triage.yml -f issue_number=$num
# sleep 60
# done
workflow_dispatch:
inputs:
issue_number:
description: 'Issue number to triage'
required: true
type: string
jobs:
claude-issue-triage:
runs-on: ubuntu-latest
timeout-minutes: 120
# SECURITY: These permissions are intentionally restrictive.
# - contents: read -> Claude can read the codebase but CANNOT write/push any code
# - issues: write -> Claude can comment and add labels (the only write access needed)
# - pull-requests: read -> Claude can view PR context but CANNOT create PRs
# This ensures that even if a malicious user attempts prompt injection via issue content,
# Claude cannot modify repository code, create branches, or open PRs.
permissions:
contents: read
issues: write
pull-requests: read
steps:
- name: Checkout repository
uses: actions/checkout@v4
with:
fetch-depth: 0
- name: Run Issue Response & Triage
id: triage
uses: anthropics/claude-code-action@v1
with:
anthropic_api_key: ${{ secrets.ANTHROPIC_API_KEY }}
github_token: ${{ secrets.GITHUB_TOKEN }}
allowed_non_write_users: "*"
# Allow all tools - security is enforced by GitHub permissions above (contents: read, issues: write)
claude_args: --model claude-opus-4-5-20251101 --allowedTools "Bash,Read,Write,Edit,Glob,Grep,WebFetch,WebSearch"
prompt: |
You're a GitHub issue first responder for the open source Cline repository.
**Issue:** #${{ github.event.issue.number || inputs.issue_number }}
**Title:** ${{ github.event.issue.title || 'See issue details below' }}
**Author:** @${{ github.event.issue.user.login || 'See issue details below' }}
## Your job
Investigate this issue thoroughly, then post a single helpful comment that helps the user and gives maintainers the context they need.
## Investigation
Start by reading the full issue:
gh issue view ${{ github.event.issue.number || inputs.issue_number }}
### Search for duplicates and related issues
Search thoroughly for existing issues that match this one:
gh issue list --search "<keywords from the issue>" --state all --limit 30
gh issue list --search "<error messages>" --state all --limit 20
gh issue list --search "<affected feature/component>" --state all --limit 20
For each relevant issue you find, read it including its comments:
gh issue view <number> --comments
You're looking for:
- **Duplicates**: Issues describing the same problem. Link to them and explain why you think they're duplicates. If closed, check how they were resolved - the solution might apply here.
- **Related issues**: Similar problems or context that could help. Pull useful information from their comments (workarounds others found, debugging steps that helped, maintainer explanations). Link to them and explain the connection.
If there are closed issues with solutions, surface those solutions prominently - this might immediately solve the user's problem.
### Analyze recent changes (ALWAYS DO THIS)
Many issues are regressions from recent releases. **Always** check what changed recently:
gh release list --limit 10
gh pr list --state merged --limit 50 --json number,title,mergedAt,author,body
Look for PRs merged in the last few weeks that might correlate with the issue. If you find a likely connection:
gh pr view <number>
gh pr diff <number>
git log --since="1 month ago" --oneline -- <relevant paths>
git show <commit>
**Always include your findings in your comment:**
- If you find a regression, call it out explicitly: which PR/commit likely caused it, who authored it, what changed, and suggest a fix direction if you can see one.
- If you don't find anything related, still mention it: "I analyzed recent PRs and releases but didn't find any changes that seem related to this issue."
### Search the codebase
Find the relevant code:
- Use grep/find to locate code related to the issue
- Key areas: `src/api/` (providers/models), `src/core/prompts/` (tools/prompts), platform-specific code for VS Code vs JetBrains
### Find documentation
Cline docs are at **https://docs.cline.bot/** and built with Mintlify from the `docs/` directory.
The URL structure maps directly to the file structure:
- `docs/getting-started/selecting-your-model.mdx` → https://docs.cline.bot/getting-started/selecting-your-model
- `docs/troubleshooting.mdx` → https://docs.cline.bot/troubleshooting
- Headings become anchors: `## Which Model` → `#which-model`
Search the `docs/` directory to find relevant documentation, then construct URLs to link users to:
```bash
ls docs/
grep -r "keyword" docs/ --include="*.mdx" -l
```
### Identify subject matter experts
For issues that clearly need engineering attention:
git log --since="6 months ago" --format="%an" -- <relevant paths> | sort | uniq -c | sort -rn | head -5
Cross-reference with GitHub usernames. Include in your response (@mention, do NOT assign):
| SME | Reason |
|-----|--------|
| @username1 | Authored PR #X which modified this area |
| @username2 | Primary contributor to affected file |
## Weak model detection
Many issues are caused by users running small or non-frontier models that don't tool-call reliably. Signs include:
- Model failing to use tools correctly
- Nonsensical or malformed responses
- User is running a small/local model or older model version
If this looks like a weak model issue, kindly suggest they try reproducing with Claude Sonnet and report back if it persists. Link to https://docs.cline.bot/getting-started/selecting-your-model if helpful. Still label and triage normally.
## Your comment
Write a single comment as a helpful community member. Be conversational, not robotic. Include what's relevant:
- **Helpful response** - Answer their question, suggest a fix, provide a workaround. If you found solutions in related closed issues, surface those prominently.
- **Duplicates and related issues** - Link to any you found and explain why they're duplicates/related. Summarize useful context from their comments.
- **Regression analysis** - If this looks like a regression, explain what change likely caused it, link to the PR/commit, and tag the author.
- **Clarifying questions** - If you need more info, ask specific questions. Don't ask for things already provided.
- **SME table** - Include the table above if this needs engineering attention. Don't tag people for questions with obvious answers or weak-model issues.
- **Context for maintainers** - Relevant code paths, what you found. Keep it concise.
- **Docs links** - If there's relevant documentation, link to it naturally in your response as a recommendation (e.g., "For more details, check out [the Ollama setup guide](url)"). Do NOT add a "Sources" section at the end - integrate doc links into your response where they're helpful.
- **Possible Duplicates section** - ALWAYS include a "Possible Duplicates" section at the end of your comment listing issues that might be duplicates so maintainers can quickly close if appropriate. If none found, say "No obvious duplicates found."
## Labels
First, retrieve all available labels and read their descriptions to understand what each is for:
gh label list --json name,description --limit 100
Then apply the appropriate labels based on your analysis. Only use labels from the list above—do not create new labels.
gh issue edit ${{ github.event.issue.number || inputs.issue_number }} --add-label "label1,label2"
If your regression analysis found a likely culprit (a recent PR/commit that probably caused this issue), add the "Regression" label:
gh issue edit ${{ github.event.issue.number || inputs.issue_number }} --add-label "Regression"
IMPORTANT: After posting your comment, add the "Bot Responded" label to indicate this issue has received an automated response:
gh issue edit ${{ github.event.issue.number || inputs.issue_number }} --add-label "Bot Responded"
## Remember
- **This is a one-time automated response** - you will NOT see their reply or respond again. Never say things like "I can help you", "let me know", "once I have that info", or "I can give you more targeted help" - you won't be there to follow up. If you ask clarifying questions, frame them for the maintainers who will follow up, e.g., "If you can share X, that would help the maintainers diagnose this."
- Don't be formulaic. Respond to what the issue actually needs.
- Surface solutions from past issues - often the fastest path to helping.
- Connecting regressions to specific changes is extremely valuable.
- Link issues with #number so they're clickable.
+284
View File
@@ -0,0 +1,284 @@
name: Claude PR Review
on:
pull_request:
types: [opened, ready_for_review]
# Manual trigger for backfilling existing PRs. Run from terminal:
# gh workflow run claude-pr-review.yml -f pr_number=1234
# Or batch process open PRs:
# gh pr list --state open --limit 10 --json number --jq '.[].number' | while read num; do
# gh workflow run claude-pr-review.yml -f pr_number=$num
# sleep 60
# done
workflow_dispatch:
inputs:
pr_number:
description: 'PR number to review'
required: true
type: string
jobs:
claude-pr-review:
# Runs on PR opened/ready_for_review (skips drafts) or manual trigger for backfilling
if: |
(github.event_name == 'pull_request' && github.event.pull_request.draft == false) ||
github.event_name == 'workflow_dispatch'
runs-on: ubuntu-latest
timeout-minutes: 120
# SECURITY: These permissions are intentionally restrictive.
# - contents: read -> Claude can read the codebase but CANNOT write/push any code
# - pull-requests: write -> Claude can post reviews and inline suggestions
# - issues: read -> Claude can search for related issues
# NOTE: Even with pull-requests: write, Claude CANNOT merge PRs because branch protection
# requires 1 approval from a Code Owner. The GITHUB_TOKEN cannot bypass this.
permissions:
contents: read
pull-requests: write
issues: read
steps:
- name: Checkout repository
uses: actions/checkout@v4
with:
fetch-depth: 0
- name: Print HEAD commit
run: |
echo "HEAD is at: $(git rev-parse HEAD)"
echo "Short: $(git rev-parse --short HEAD)"
git log -1 --format="Commit: %H%nAuthor: %an <%ae>%nDate: %ad%nMessage: %s"
- name: Get PR number
id: pr
run: |
if [ "${{ github.event_name }}" == "workflow_dispatch" ]; then
echo "number=${{ inputs.pr_number }}" >> $GITHUB_OUTPUT
else
echo "number=${{ github.event.pull_request.number }}" >> $GITHUB_OUTPUT
fi
- name: Run PR Review
id: review
uses: anthropics/claude-code-action@v1
with:
anthropic_api_key: ${{ secrets.ANTHROPIC_API_KEY }}
github_token: ${{ secrets.GITHUB_TOKEN }}
allowed_non_write_users: "*"
claude_args: --model claude-opus-4-5-20251101 --allowedTools "Bash,Read,Write,Edit,Glob,Grep,WebFetch,WebSearch"
prompt: |
You're a GitHub PR reviewer for the open source Cline repository. Your goal is to give the PR author helpful feedback and give maintainers the context they need to review efficiently.
PR: #${{ steps.pr.outputs.number }}
## Gather context
```bash
# Get full PR details
gh pr view ${{ steps.pr.outputs.number }} --json number,title,body,author,createdAt,updatedAt,isDraft,labels,commits,files,additions,deletions,changedFiles,baseRefName,headRefName,mergeable,reviewDecision
# Get the diff
gh pr diff ${{ steps.pr.outputs.number }}
# Check CI status
gh pr checks ${{ steps.pr.outputs.number }}
# Get existing review comments (to understand context and your previous feedback)
gh api repos/${{ github.repository }}/pulls/${{ steps.pr.outputs.number }}/comments --jq '.[] | {user: .user.login, body: .body, path: .path, created_at: .created_at}'
# Get conversation comments
gh pr view ${{ steps.pr.outputs.number }} --comments
```
If this is a re-review (workflow_dispatch event):
Read your previous comments carefully. Understand what you asked for before.
Check if new commits or comments address your previous feedback.
## Check contributing guidelines
Flag (but don't block) if:
- Missing changeset - For user-facing changes, check if there's a `.changeset/` file:
```bash
gh pr diff ${{ steps.pr.outputs.number }} --name-only | grep '.changeset/' || echo "No changeset found"
```
If missing, ask them to run `npm run changeset`
- Missing tests - New features should have tests
## Find related issues and PRs
Search thoroughly for context that might help with the review:
```bash
# Find related issues for context
gh issue list --search "<keywords from the PR>" --state all --limit 30
gh issue list --search "<error messages or feature names>" --state all --limit 20
# Find similar PRs for reference
gh pr list --search "<keywords>" --state all --limit 30
```
For each relevant issue or PR you find, read it including comments:
```bash
gh issue view <number> --comments
gh pr view <number> --comments
```
Look for:
- Open issues this PR might fix that weren't linked in the description
- Similar PRs that went through review - what feedback did they get? What patterns did they follow?
- Context from maintainer discussions that could inform your review
## Find subject matter experts
For files changed in this PR, find who knows the code best:
```bash
# Get files changed
gh pr diff ${{ steps.pr.outputs.number }} --name-only
# For each relevant path, find contributors
git log --since="6 months ago" --format="%an" -- <path> | sort | uniq -c | sort -rn | head -5
```
Cross-reference git authors with GitHub usernames. Include an SME table in your response:
| SME | Reason |
|-----|--------|
| @username1 | Authored PR #X which modified this area |
| @username2 | Primary contributor to affected file (15 commits in 6 months) |
| @username3 | Reviewed similar PR #Y with extensive feedback |
## Deep code review
This is the most important part. Don't just look for syntax issues - understand what the PR is trying to achieve and whether the implementation is the right approach.
Step 1: Understand the intent
Read the PR description and understand what the author is trying to accomplish. What problem are they solving? What feature are they adding?
Step 2: Form your own opinion first
Before analyzing their code, think about how YOU would implement this feature or fix. What files would you touch? What patterns would you follow? What edge cases would you handle?
Step 3: Compare approaches
Now look at their implementation. How does it compare to what you would have done?
- Is their approach better in some ways? Note what they did well.
- Is their approach missing something? Be specific about what and why.
- Are there edge cases they haven't considered?
- Does it follow the patterns established in similar parts of the codebase?
Step 4: Look at the bigger picture
- What other files or systems does this change interact with?
- Could this break anything else?
- Is there additional work needed beyond this PR to complete the feature?
- Does this fit well with the overall architecture?
Step 5: Find reference implementations
Look for similar changes in the codebase:
```bash
git log --oneline --all --grep="<relevant keywords>" | head -20
git log --oneline -- <similar files> | head -20
```
If this is adding a new API provider, look at how other providers are implemented.
If this is adding a new feature, look at how similar features were added.
Note where their implementation aligns with or diverges from established patterns.
Step 6: Standard code review checks
- DRY: Is there duplicated code that could be extracted?
- Error handling: Are errors handled appropriately?
- Security: Any injection risks, credential exposure, unsafe dependencies?
- Performance: Any obvious inefficiencies, memory leaks, N+1 patterns?
- Types: Is TypeScript used correctly? Any unsafe type assertions?
- Naming: Are variables and functions named clearly?
- Comments: Is complex logic explained? Are there outdated comments?
## Inline code suggestions
For specific code improvements, use GitHub's suggestion syntax via `gh api`.
This creates suggestions the author can commit with one click.
Single-line suggestion:
```bash
gh api repos/${{ github.repository }}/pulls/${{ steps.pr.outputs.number }}/reviews \
-X POST \
-f commit_id="$(gh pr view ${{ steps.pr.outputs.number }} --json headRefOid -q .headRefOid)" \
-f event="COMMENT" \
-f body="" \
-F comments='[
{
"path": "src/example.ts",
"line": 42,
"body": "Consider simplifying:\n\n```suggestion\nconst result = items.filter(Boolean);\n```"
}
]'
```
Multi-line suggestion (replacing lines 40-45):
```bash
gh api repos/${{ github.repository }}/pulls/${{ steps.pr.outputs.number }}/reviews \
-X POST \
-f commit_id="$(gh pr view ${{ steps.pr.outputs.number }} --json headRefOid -q .headRefOid)" \
-f event="COMMENT" \
-f body="" \
-F comments='[
{
"path": "src/example.ts",
"start_line": 40,
"line": 45,
"body": "This can be simplified:\n\n```suggestion\nconst simplified = doThing();\n```"
}
]'
```
Use inline suggestions for concrete improvements. Use regular comments for questions or broader feedback.
## Post your review
After your investigation, post a single helpful comment that helps the author and gives maintainers context.
Start by noting the commit hash you reviewed:
```bash
git rev-parse --short HEAD
```
Include this at the top of your comment: "Reviewed at commit: <short hash>"
Then thank them for their contribution. Be conversational, not robotic.
Include what's relevant:
- In-depth explanation of what the PR does - Be comprehensive. A maintainer should be able to read this section and fully understand the author's intent, why they made the changes, how they implemented it, and what files/systems are affected. Don't just summarize - explain.
- Related issues/PRs you found that provide useful context (link to them)
- Your review findings (issues to address, suggestions, etc.)
- Clear next steps for the author
Include a "For Maintainers" section with:
- Anything else useful to help the maintainer resolve this PR
- Related issues/PRs with context on why they're relevant
- Open issues this PR might fix that weren't linked in the description
- Your recommendation: merge as-is, needs changes, needs discussion, close, etc.
- SME table - who should review this and why
For the SME table:
| SME | Reason |
|-----|--------|
| @username | Primary contributor to affected files |
## Update labels
Add appropriate labels based on your analysis:
```bash
gh label list --json name,description --limit 100
gh pr edit ${{ steps.pr.outputs.number }} --add-label "label1,label2"
```
When done, add the reviewed label:
```bash
gh pr edit ${{ steps.pr.outputs.number }} --add-label "Bot Reviewed"
```
## Remember
- This is a one-time automated response - you will NOT see their reply or respond again. Never say things like "let me know if you have questions", "I can help you with", or "feel free to ask" - you won't be there to follow up. Frame any questions for the maintainers who will follow up.
- Be helpful and welcoming - Many contributors are new to the project
- Be specific - Point to exact lines and suggest fixes, don't give vague feedback
- Think deeply - Don't just surface-level review, understand the intent and evaluate the approach
- Use inline suggestions - Make it easy for authors to accept changes
- You're a first-pass reviewer - A human maintainer will do final approval
@@ -1,70 +0,0 @@
name: Smoke Tests
on:
push:
branches: [main]
paths:
- 'src/core/**'
- 'src/shared/**'
- 'proto/**'
- 'evals/**'
- '.github/workflows/cline-evals-regression.yml'
pull_request:
paths:
- 'src/core/**'
- 'src/shared/**'
- 'proto/**'
- 'evals/**'
- '.github/workflows/cline-evals-regression.yml'
workflow_dispatch:
permissions:
contents: read
concurrency:
group: smoke-tests-${{ github.ref }}
cancel-in-progress: true
jobs:
smoke-tests:
name: Smoke Tests
runs-on: ubuntu-latest
timeout-minutes: 15
steps:
- uses: actions/checkout@v4
- uses: actions/setup-node@v4
with:
node-version: '22'
cache: 'npm'
- name: Install dependencies
run: npm ci
- name: Build and install CLI
run: |
npm run protos
cd cli && npm install && npm run build && npm link
echo "$(npm config get prefix)/bin" >> $GITHUB_PATH
- name: Verify CLI
run: cline --version
- name: Run smoke tests
env:
CLINE_API_KEY: ${{ secrets.CLINE_API_KEY }}
run: |
cline auth -p cline -k "$CLINE_API_KEY" -m "anthropic/claude-sonnet-4.5"
npx tsx evals/smoke-tests/run-smoke-tests.ts --trials 1 --parallel
- name: Generate summary
if: always()
run: cat evals/smoke-tests/results/latest/summary.md >> $GITHUB_STEP_SUMMARY
- name: Upload results
uses: actions/upload-artifact@v4
if: always()
with:
name: smoke-test-results-${{ github.run_id }}
path: evals/smoke-tests/results/latest/
retention-days: 30
+325
View File
@@ -0,0 +1,325 @@
name: Cline PR Code Review
on:
pull_request:
types:
[opened, ready_for_review]
# Manual trigger for backfilling existing PRs. Run from terminal:
# gh workflow run cline-pr-review.yml -f pr_number=1234
# Or batch process open PRs:
# gh pr list --state open --limit 10 --json number --jq '.[].number' | while read num; do
# gh workflow run cline-pr-review.yml -f pr_number=$num
# sleep 60
# done
workflow_dispatch:
inputs:
pr_number:
description: "PR number to review"
required: true
type: string
concurrency:
group: pr-review-${{ github.event.pull_request.number || inputs.pr_number }}
cancel-in-progress: true
jobs:
cline-pr-review:
# Runs on PR opened/ready_for_review (skips drafts) or manual trigger for backfilling
if: |
(github.event_name == 'pull_request' && github.event.pull_request.draft == false) ||
github.event_name == 'workflow_dispatch'
runs-on: ubuntu-latest
timeout-minutes: 60
# SECURITY: These permissions are intentionally restrictive.
# - contents: read -> cline can read the codebase but CANNOT write/push any code
# - pull-requests: write -> cline can post reviews and inline suggestions
# - issues: read -> cline can search for related issues
# NOTE: Even with pull-requests: write, cline CANNOT merge PRs because branch protection
# requires 1 approval from a Code Owner. The GITHUB_TOKEN cannot bypass this.
permissions:
contents: read
pull-requests: write
issues: read
steps:
- name: Checkout repository
uses: actions/checkout@v4
with:
fetch-depth: 0
- name: Print HEAD commit
run: |
echo "HEAD is at: $(git rev-parse HEAD)"
echo "Short: $(git rev-parse --short HEAD)"
git log -1 --format="Commit: %H%nAuthor: %an <%ae>%nDate: %ad%nMessage: %s"
- name: Setup Node.js
uses: actions/setup-node@v4
with:
node-version: 22
cache: "npm"
- name: Install and Verify Cline CLI
run: |
npm install -g cline
cline version # verify installation
- name: Configure Cline with Anthropic
run: |
npx cline auth --provider anthropic \
--apikey "${{ secrets.ANTHROPIC_API_KEY }}" \
--modelid claude-opus-4-5-20251101
- name: Get PR number
id: pr
run: |
if [ "${{ github.event_name }}" == "workflow_dispatch" ]; then
echo "number=${{ inputs.pr_number }}" >> $GITHUB_OUTPUT
else
echo "number=${{ github.event.pull_request.number }}" >> $GITHUB_OUTPUT
fi
- name: Review PR with Cline
env:
PR_NUMBER: ${{ steps.pr.outputs.number }}
GITHUB_REPO: ${{ github.repository }}
GH_TOKEN: ${{ github.token }}
CLINE_COMMAND_PERMISSIONS: |
{
"allow": [
"gh pr diff *",
"gh pr view *",
"gh pr checks *",
"gh pr list *",
"gh label list *",
"gh issue list *",
"gh issue view *",
"git log *",
"gh pr comment ${{ steps.pr.outputs.number }} *",
"gh pr edit ${{ steps.pr.outputs.number }} *",
"gh api repos/${{ github.repository }}/pulls/${{ steps.pr.outputs.number }}/comments *",
"gh api repos/${{ github.repository }}/pulls/${{ steps.pr.outputs.number }}/reviews *"
]
}
run: |
npx cline --yolo 'You'\''re a GitHub PR reviewer for the open source Cline repository. Your goal is to give the PR author helpful feedback and give maintainers the context they need to review efficiently.
PR: #'"${PR_NUMBER}"'
## Gather context
```bash
# Get full PR details
gh pr view '"${PR_NUMBER}"' --json number,title,body,author,createdAt,updatedAt,isDraft,labels,commits,files,additions,deletions,changedFiles,baseRefName,headRefName,mergeable,reviewDecision
# Get the diff
gh pr diff '"${PR_NUMBER}"'
# Check CI status
gh pr checks '"${PR_NUMBER}"'
# Get existing review comments (to understand context and your previous feedback)
gh api repos/'"${GITHUB_REPO}"'/pulls/'"${PR_NUMBER}"'/comments --jq '\''.[] | {user: .user.login, body: .body, path: .path, created_at: .created_at}'\''
# Get conversation comments
gh pr view '"${PR_NUMBER}"' --comments
```
If this is a re-review (workflow_dispatch event):
Read your previous comments carefully. Understand what you asked for before.
Check if new commits or comments address your previous feedback.
## Check contributing guidelines
Flag (but don'\''t block) if:
- Missing changeset - For user-facing changes, check if there'\''s a `.changeset/` file:
```bash
gh pr diff '"${PR_NUMBER}"' --name-only | grep '\''.changeset/'\'' || echo '\''No changeset found'\''
```
If missing, ask them to run `npm run changeset`
- Missing tests - New features should have tests
## Find related issues and PRs
Search thoroughly for context that might help with the review:
```bash
# Find related issues for context
gh issue list --search '\''<keywords from the PR>'\'' --state all --limit 30
gh issue list --search '\''<error messages or feature names>'\'' --state all --limit 20
# Find similar PRs for reference
gh pr list --search '\''<keywords>'\'' --state all --limit 30
```
For each relevant issue or PR you find, read it including comments:
```bash
gh issue view <number> --comments
gh pr view <number> --comments
```
Look for:
- Open issues this PR might fix that weren'\''t linked in the description
- Similar PRs that went through review - what feedback did they get? What patterns did they follow?
- Context from maintainer discussions that could inform your review
## Find subject matter experts
For files changed in this PR, find who knows the code best:
```bash
# Get files changed
gh pr diff '"${PR_NUMBER}"' --name-only
# For each relevant path, find contributors
git log --since='\''6 months ago'\'' --format='\''%an'\'' -- <path> | sort | uniq -c | sort -rn | head -5
```
Cross-reference git authors with GitHub usernames. Include an SME table in your response:
| SME | Reason |
|-----|--------|
| @username1 | Authored PR #X which modified this area |
| @username2 | Primary contributor to affected file (15 commits in 6 months) |
| @username3 | Reviewed similar PR #Y with extensive feedback |
## Bash command usage
Don'\''t use operators like `|`, `&&`, or `;` - run each command separately and analyze the output.
When referencing command outputs, quote them properly to avoid formatting issues.
## Deep code review
This is the most important part. Don'\''t just look for syntax issues - understand what the PR is trying to achieve and whether the implementation is the right approach.
Step 1: Understand the intent
Read the PR description and understand what the author is trying to accomplish. What problem are they solving? What feature are they adding?
Step 2: Form your own opinion first
Before analyzing their code, think about how YOU would implement this feature or fix. What files would you touch? What patterns would you follow? What edge cases would you handle?
Step 3: Compare approaches
Now look at their implementation. How does it compare to what you would have done?
- Is their approach better in some ways? Note what they did well.
- Is their approach missing something? Be specific about what and why.
- Are there edge cases they haven'\''t considered?
- Does it follow the patterns established in similar parts of the codebase?
Step 4: Look at the bigger picture
- What other files or systems does this change interact with?
- Could this break anything else?
- Is there additional work needed beyond this PR to complete the feature?
- Does this fit well with the overall architecture?
Step 5: Find reference implementations
Look for similar changes in the codebase:
```bash
git log --oneline --all --grep='\''<relevant keywords>'\'' | head -20
git log --oneline -- <similar files> | head -20
```
If this is adding a new API provider, look at how other providers are implemented.
If this is adding a new feature, look at how similar features were added.
Note where their implementation aligns with or diverges from established patterns.
Step 6: Standard code review checks
- DRY: Is there duplicated code that could be extracted?
- Error handling: Are errors handled appropriately?
- Security: Any injection risks, credential exposure, unsafe dependencies?
- Performance: Any obvious inefficiencies, memory leaks, N+1 patterns?
- Types: Is TypeScript used correctly? Any unsafe type assertions?
- Naming: Are variables and functions named clearly?
- Comments: Is complex logic explained? Are there outdated comments?
## Inline code suggestions
For specific code improvements, use GitHub'\''s suggestion syntax via `gh api`.
This creates suggestions the author can commit with one click.
Single-line suggestion:
```bash
gh api repos/'"${GITHUB_REPO}"'/pulls/'"${PR_NUMBER}"'/reviews \
-X POST \
-f commit_id="$(gh pr view '"${PR_NUMBER}"' --json headRefOid -q .headRefOid)" \
-f event='\''COMMENT'\'' \
-f body='\'''\'' \
-F comments='\''[
{
"path": "src/example.ts",
"line": 42,
"body": "Consider simplifying:\n\n```suggestion\nconst result = items.filter(Boolean);\n```"
}
]'\''
```
Multi-line suggestion (replacing lines 40-45):
```bash
gh api repos/'"${GITHUB_REPO}"'/pulls/'"${PR_NUMBER}"'/reviews \
-X POST \
-f commit_id="$(gh pr view '"${PR_NUMBER}"' --json headRefOid -q .headRefOid)" \
-f event='\''COMMENT'\'' \
-f body='\'''\'' \
-F comments='\''[
{
"path": "src/example.ts",
"start_line": 40,
"line": 45,
"body": "This can be simplified:\n\n```suggestion\nconst simplified = doThing();\n```"
}
]'\''
```
Use inline suggestions for concrete improvements. Use regular comments for questions or broader feedback.
## Post your review
After your investigation, post a single helpful comment that helps the author and gives maintainers context.
Start by noting the commit hash you reviewed:
```bash
git rev-parse --short HEAD
```
Include this at the top of your comment: "Reviewed at commit: <short hash>"
Then thank them for their contribution. Be conversational, not robotic.
Include what'\''s relevant:
- In-depth explanation of what the PR does - Be comprehensive. A maintainer should be able to read this section and fully understand the author'\''s intent, why they made the changes, how they implemented it, and what files/systems are affected. Don'\''t just summarize - explain.
- Related issues/PRs you found that provide useful context (link to them)
- Your review findings (issues to address, suggestions, etc.)
- Clear next steps for the author
Include a '\''For Maintainers'\'' section with:
- Anything else useful to help the maintainer resolve this PR
- Related issues/PRs with context on why they'\''re relevant
- Open issues this PR might fix that weren'\''t linked in the description
- Your recommendation: merge as-is, needs changes, needs discussion, close, etc.
- SME table - who should review this and why
For the SME table:
| SME | Reason |
|-----|--------|
| @username | Primary contributor to affected files |
## Update labels
Add appropriate labels based on your analysis:
```bash
gh label list --json name,description --limit 100
gh pr edit '"${PR_NUMBER}"' --add-label '\''label1,label2'\''
```
When done, add the reviewed label:
```bash
gh pr edit '"${PR_NUMBER}"' --add-label '\''Bot Reviewed'\''
```
## Remember
- This is a one-time automated response - you will NOT see their reply or respond again. Never say things like '\''let me know if you have questions'\'', '\''I can help you with'\'', or '\''feel free to ask'\'' - you won'\''t be there to follow up. Frame any questions for the maintainers who will follow up.
- Be helpful and welcoming - Many contributors are new to the project
- Be specific - Point to exact lines and suggest fixes, don'\''t give vague feedback
- Think deeply - Don'\''t just surface-level review, understand the intent and evaluate the approach
- Use inline suggestions - Make it easy for authors to accept changes
- You'\''re a first-pass reviewer - A human maintainer will do final approval'
+9 -28
View File
@@ -35,26 +35,10 @@ jobs:
contents: read
steps:
- uses: actions/checkout@v4
- name: Setup Node.js environment
uses: actions/setup-node@v4
- name: Setup Bun
uses: oven-sh/setup-bun@v2
with:
node-version: 22
# Cache root dependencies - only reuse if package-lock.json exactly matches
- name: Cache root dependencies
uses: actions/cache@v4
id: root-cache
with:
path: node_modules
key: ${{ runner.os }}-npm-${{ hashFiles('package-lock.json') }}
# Cache webview-ui dependencies - only reuse if package-lock.json exactly matches
- name: Cache webview-ui dependencies
uses: actions/cache@v4
id: webview-cache
with:
path: webview-ui/node_modules
key: ${{ runner.os }}-npm-webview-${{ hashFiles('webview-ui/package-lock.json') }}
bun-version: latest
# Cache VS Code installation
- name: Cache VS Code
@@ -75,20 +59,17 @@ jobs:
~/.cache/ms-playwright
~/Library/Caches/ms-playwright
~/AppData/Local/ms-playwright
key: playwright-browsers-${{ runner.os }}-${{ hashFiles('package-lock.json') }}
key: playwright-browsers-${{ runner.os }}-${{ hashFiles('bun.lockb') }}
restore-keys: |
playwright-browsers-${{ runner.os }}-
- name: Install root dependencies
run: npm ci
- name: Install dependencies
run: bun install --frozen-lockfile
env:
GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
- name: Install webview-ui dependencies
run: cd webview-ui && npm ci
- name: Install vsce
run: npm install -g @vscode/vsce
run: bun add -g @vscode/vsce
- name: Install xvfb on Linux
if: matrix.runner == 'ubuntu'
@@ -97,11 +78,11 @@ jobs:
# Run optimized E2E tests (eliminates redundant builds)
- name: Run E2E tests - Linux
if: matrix.runner == 'ubuntu'
run: xvfb-run -a npm run test:e2e:optimal
run: xvfb-run -a bun run test:e2e:optimal
- name: Run E2E tests - Non-Linux
if: matrix.runner != 'ubuntu'
run: npm run test:e2e:optimal
run: bun run test:e2e:optimal
- uses: actions/upload-artifact@v4
if: ${{ failure() }}
+15 -12
View File
@@ -1,7 +1,7 @@
name: Publish NPM Release
on:
workflow_call:
workflow_dispatch:
inputs:
confirm_publish:
description: 'Type "publish" to confirm you want to publish to NPM'
@@ -9,8 +9,7 @@ on:
type: string
permissions:
contents: write # Required for pushing tags
id-token: write # Required for npm trusted publishing (OIDC)
contents: read
checks: write # Required by test workflow
pull-requests: write # Required by test workflow
@@ -21,7 +20,7 @@ jobs:
publish-npm-release:
needs: test
name: Publish Cline CLI to NPM
if: github.repository == 'cline/cline' && github.ref == 'refs/heads/main' && inputs.confirm_publish == 'publish'
if: github.repository == 'cline/cline' && github.ref == 'refs/heads/main' && github.event.inputs.confirm_publish == 'publish'
runs-on: ubuntu-latest
steps:
@@ -31,10 +30,19 @@ jobs:
- name: Setup Node.js
uses: actions/setup-node@v4
with:
node-version: "24.x"
node-version: "20.x"
registry-url: "https://registry.npmjs.org"
# Cache root dependencies - only reuse if package-lock.json exactly matches
- name: Cache root dependencies
uses: actions/cache@v4
id: root-cache
with:
path: node_modules
key: ${{ runner.os }}-npm-${{ hashFiles('package-lock.json') }}
- name: Install root dependencies and CLI dependencies
if: steps.check_commits.outputs.skip != 'true'
run: npm ci --include=optional # this will also install cli deps because "cli" in included in root package.json workspaces field
- name: Generate Protos
@@ -73,18 +81,13 @@ jobs:
cat dist-standalone/package.json | grep version
- name: Publish to NPM with latest tag
env:
NODE_AUTH_TOKEN: ${{ secrets.NPM_RELEASE_TOKEN }}
run: |
echo "Publishing version ${{ steps.version.outputs.version }} to NPM with tag 'latest'..."
cd dist-standalone
npm publish --tag latest --access public
- name: Tag release
run: |
git config user.name "github-actions[bot]"
git config user.email "github-actions[bot]@users.noreply.github.com"
git tag "v${{ steps.version.outputs.version }}-cli"
git push origin "v${{ steps.version.outputs.version }}-cli"
- name: Summary
run: |
echo "✅ Successfully published cline@${{ steps.version.outputs.version }} to NPM with tag 'latest'"
+15 -15
View File
@@ -1,17 +1,12 @@
name: Publish NPM Nightly
on:
workflow_call:
inputs:
force_publish:
description: "Force publish even if there are no commits in the last 24 hours"
required: false
type: boolean
default: false
schedule:
- cron: "0 12 * * *" # 4 AM PST (UTC-8) = 12 UTC
workflow_dispatch:
permissions:
contents: read
id-token: write # Required for npm trusted publishing (OIDC)
checks: write # Required by test workflow
pull-requests: write # Required by test workflow
@@ -32,12 +27,6 @@ jobs:
- name: Check for recent commits
id: check_commits
run: |
if [ "${{ inputs.force_publish }}" = "true" ]; then
echo "force_publish enabled, proceeding with publish"
echo "skip=false" >> $GITHUB_OUTPUT
exit 0
fi
if [ $(git rev-list --count HEAD --since="24 hours ago") -eq 0 ]; then
echo "No commits in last 24 hours, skipping publish"
echo "skip=true" >> $GITHUB_OUTPUT
@@ -50,9 +39,18 @@ jobs:
if: steps.check_commits.outputs.skip != 'true'
uses: actions/setup-node@v4
with:
node-version: "24.x"
node-version: "20.x"
registry-url: "https://registry.npmjs.org"
# Cache root dependencies - only reuse if package-lock.json exactly matches
- name: Cache root dependencies
if: steps.check_commits.outputs.skip != 'true'
uses: actions/cache@v4
id: root-cache
with:
path: node_modules
key: ${{ runner.os }}-npm-${{ hashFiles('package-lock.json') }}
- name: Install root dependencies and CLI dependencies
if: steps.check_commits.outputs.skip != 'true'
run: npm ci --include=optional # this will also install cli deps because "cli" in included in root package.json workspaces field
@@ -120,6 +118,8 @@ jobs:
- name: Publish to NPM with nightly tag
if: steps.check_commits.outputs.skip != 'true'
env:
NODE_AUTH_TOKEN: ${{ secrets.NPM_RELEASE_TOKEN }}
run: |
echo "Publishing version ${{ steps.version.outputs.version }} to NPM with tag 'nightly'..."
cd dist-standalone
-215
View File
@@ -1,215 +0,0 @@
# Build and Pack CLI
#
# Builds a CLI tarball from any branch/commit and publishes it as a GitHub Release.
# Requires write access to the repository (maintainers/collaborators only).
#
# Security: Split into two jobs to isolate untrusted build code from write tokens.
# The build job runs arbitrary ref code with zero permissions. The release job
# only runs trusted GitHub Actions with write scope.
#
# Usage (helper script, auto-detects current branch):
# ./scripts/build-cli-artifact.sh
# ./scripts/build-cli-artifact.sh feature/my-changes
# ./scripts/build-cli-artifact.sh feature/my-changes 1234 # comments on PR
#
# Usage (gh CLI directly):
# gh workflow run pack-cli.yml -f ref=main
# gh workflow run pack-cli.yml -f ref=abc123 -f pr_number=1234
#
# Install the built CLI (no auth required):
# npm install -g https://github.com/cline/cline/releases/download/cli-build-<sha>/cline-<ver>.tgz
#
# Find releases:
# gh release list --limit 10
name: Build and Pack CLI
permissions:
contents: read
on:
workflow_dispatch:
inputs:
ref:
description: 'Branch, tag, or commit SHA to build (leave empty for default branch)'
required: false
type: string
pr_number:
description: 'PR number to comment on with install instructions (optional)'
required: false
type: number
jobs:
# ── Build job: runs untrusted ref code with ZERO permissions ──
build:
name: Build CLI
runs-on: ubuntu-latest
permissions: {}
outputs:
commit_sha: ${{ steps.commit.outputs.sha }}
tarball: ${{ steps.pack.outputs.tarball }}
steps:
- name: Checkout code
uses: actions/checkout@v4
with:
ref: ${{ inputs.ref || github.ref }}
persist-credentials: false
- name: Get commit SHA
id: commit
run: |
COMMIT_SHA=$(git rev-parse --short HEAD)
echo "sha=$COMMIT_SHA" >> $GITHUB_OUTPUT
echo "Building from commit: $COMMIT_SHA"
- name: Setup Node.js
uses: actions/setup-node@v4
with:
node-version: "20.x"
- name: Install dependencies
run: npm ci --include=optional
- name: Generate Protos
run: npm run protos
- name: Build standalone package
run: node scripts/package-npm.mjs
- name: Create Tarball
id: pack
run: |
cd dist-standalone
TARBALL=$(npm pack)
echo "tarball=$TARBALL" >> $GITHUB_OUTPUT
echo "Created tarball: $TARBALL"
- name: Upload artifact
uses: actions/upload-artifact@v4
with:
name: cli-tarball
path: dist-standalone/*.tgz
# ── Release job: only trusted Actions code, with write permissions ──
release:
name: Release CLI
needs: build
runs-on: ubuntu-latest
permissions:
contents: write
pull-requests: write
issues: write
steps:
- name: Download artifact
uses: actions/download-artifact@v4
with:
name: cli-tarball
path: dist-standalone
- name: Create GitHub Release
id: create_release
uses: actions/github-script@v7
with:
script: |
const fs = require('fs');
const path = require('path');
const commit = '${{ needs.build.outputs.commit_sha }}';
const tarball = '${{ needs.build.outputs.tarball }}';
// Delete existing release/tag if re-running for the same commit
const tagName = `cli-build-${commit}`;
try {
const existing = await github.rest.repos.getReleaseByTag({
owner: context.repo.owner,
repo: context.repo.repo,
tag: tagName
});
await github.rest.repos.deleteRelease({
owner: context.repo.owner,
repo: context.repo.repo,
release_id: existing.data.id
});
await github.rest.git.deleteRef({
owner: context.repo.owner,
repo: context.repo.repo,
ref: `tags/${tagName}`
});
core.info(`Deleted existing release for ${tagName}`);
} catch (e) {
// Release doesn't exist yet, that's fine
}
// Create a release
const release = await github.rest.repos.createRelease({
owner: context.repo.owner,
repo: context.repo.repo,
tag_name: tagName,
name: `CLI Build (${commit})`,
body: `Automated CLI build from commit ${commit}\n\nInstall with:\n\`\`\`bash\nnpm install -g https://github.com/${context.repo.owner}/${context.repo.repo}/releases/download/${tagName}/${tarball}\n\`\`\``,
draft: false,
prerelease: true
});
// Upload the tarball as a release asset
const tarballPath = path.join('dist-standalone', tarball);
const tarballData = fs.readFileSync(tarballPath);
await github.rest.repos.uploadReleaseAsset({
owner: context.repo.owner,
repo: context.repo.repo,
release_id: release.data.id,
name: tarball,
data: tarballData
});
const downloadUrl = `https://github.com/${context.repo.owner}/${context.repo.repo}/releases/download/${tagName}/${tarball}`;
core.setOutput('release_url', release.data.html_url);
core.setOutput('download_url', downloadUrl);
- name: Comment on PR with download instructions
if: inputs.pr_number != ''
uses: actions/github-script@v7
with:
script: |
const commit = '${{ needs.build.outputs.commit_sha }}';
const releaseUrl = '${{ steps.create_release.outputs.release_url }}';
const downloadUrl = '${{ steps.create_release.outputs.download_url }}';
const prNumber = ${{ inputs.pr_number || 0 }};
if (!prNumber) return;
const comment = `## 📦 CLI Build Ready
A CLI build has been created for commit \`${commit}\`.
### Install Directly from URL (No Authentication Required!)
\`\`\`bash
npm install -g ${downloadUrl}
\`\`\`
### Alternative: Download and Install
\`\`\`bash
curl -L ${downloadUrl} -o cline.tgz
npm install -g ./cline.tgz
\`\`\`
📦 [View Release](${releaseUrl})
`;
await github.rest.issues.createComment({
issue_number: prNumber,
owner: context.repo.owner,
repo: context.repo.repo,
body: comment
});
- name: Summary
run: |
echo "✅ CLI build complete!"
echo ""
echo "📦 Release: ${{ steps.create_release.outputs.release_url }}"
echo "🔗 Download URL: ${{ steps.create_release.outputs.download_url }}"
echo ""
echo "Install from anywhere (no authentication required):"
echo " npm install -g ${{ steps.create_release.outputs.download_url }}"
@@ -1,53 +0,0 @@
name: Publish CLI (Trusted)
on:
schedule:
- cron: "0 12 * * *" # 4 AM PST (UTC-8) = 12 UTC
workflow_dispatch:
inputs:
publish_target:
description: "Which publish flow to run"
required: true
default: "main"
type: choice
options:
- main
- nightly
confirm_publish:
description: 'Required when publish_target=main. Type "publish" to confirm release publish.'
required: false
type: string
force_nightly_publish:
description: "Force nightly publish even with no commits in last 24h"
required: false
type: boolean
default: false
permissions:
id-token: write # Required for npm trusted publishing (OIDC)
contents: write # Required because npm-main creates/pushes git tags
checks: write # Required by nested reusable test workflow
pull-requests: write # Required by nested reusable test workflow
jobs:
publish-main:
if: |
github.repository == 'cline/cline' && (
github.event_name == 'workflow_dispatch' &&
github.event.inputs.publish_target == 'main' &&
github.event.inputs.confirm_publish == 'publish' &&
!endsWith(github.actor, '[bot]')
)
uses: ./.github/workflows/npm-main.yaml
with:
confirm_publish: ${{ github.event.inputs.confirm_publish }}
publish-nightly:
if: |
github.repository == 'cline/cline' && (
github.event_name == 'schedule' ||
(github.event_name == 'workflow_dispatch' && github.event.inputs.publish_target == 'nightly')
)
uses: ./.github/workflows/npm-nightly.yaml
with:
force_publish: ${{ github.event_name == 'workflow_dispatch' && github.event.inputs.force_nightly_publish == 'true' }}
+7 -10
View File
@@ -33,19 +33,16 @@ jobs:
fi
echo "Found recent commits, proceeding with build"
- name: Setup Node.js
uses: actions/setup-node@v4
- name: Setup Bun
uses: oven-sh/setup-bun@v2
with:
node-version: "lts/*"
bun-version: latest
- name: Install root dependencies
run: npm ci --include=optional
- name: Install webview-ui dependencies
run: cd webview-ui && npm ci --include=optional
- name: Install dependencies
run: bun install --frozen-lockfile
- name: Install Publishing Tools
run: npm install -g @vscode/vsce ovsx
run: bun add -g @vscode/vsce ovsx
- name: Publish Extension as Pre-release
env:
@@ -61,4 +58,4 @@ jobs:
OTEL_EXPORTER_OTLP_PROTOCOL: ${{ secrets.OTEL_EXPORTER_OTLP_PROTOCOL }}
OTEL_EXPORTER_OTLP_ENDPOINT: ${{ secrets.OTEL_EXPORTER_OTLP_ENDPOINT }}
OTEL_EXPORTER_OTLP_HEADERS: ${{ secrets.OTEL_EXPORTER_OTLP_HEADERS }}
run: npm run publish:marketplace:nightly
run: bun run publish:marketplace:nightly
+8 -11
View File
@@ -39,19 +39,16 @@ jobs:
fetch-depth: 0
fetch-tags: true
- name: Setup Node.js
uses: actions/setup-node@v4
- name: Setup Bun
uses: oven-sh/setup-bun@v2
with:
node-version: "lts/*"
bun-version: latest
- name: Install root dependencies
run: npm install --include=optional
- name: Install webview-ui dependencies
run: cd webview-ui && npm install --include=optional
- name: Install dependencies
run: bun install --frozen-lockfile
- name: Install Publishing Tools
run: npm install -g @vscode/vsce ovsx
run: bun add -g @vscode/vsce ovsx
- name: Get Version
id: get_version
@@ -93,10 +90,10 @@ jobs:
vsce package --allow-package-secrets sendgrid --out "cline-${{ steps.get_version.outputs.version }}.vsix"
if [ "${{ github.event.inputs.release-type }}" = "pre-release" ]; then
npm run publish:marketplace:prerelease
bun run publish:marketplace:prerelease
echo "Successfully published pre-release version ${{ steps.get_version.outputs.version }} to VS Code Marketplace and Open VSX Registry"
else
npm run publish:marketplace
bun run publish:marketplace
echo "Successfully published release version ${{ steps.get_version.outputs.version }} to VS Code Marketplace and Open VSX Registry"
fi
+27 -54
View File
@@ -24,25 +24,18 @@ jobs:
- name: Checkout code
uses: actions/checkout@v4
- name: Setup Node.js environment
uses: actions/setup-node@v4
- name: Setup Bun
uses: oven-sh/setup-bun@v2
with:
node-version: 22
cache: 'npm'
cache-dependency-path: |
package-lock.json
webview-ui/package-lock.json
bun-version: latest
- name: Install root dependencies
run: npm ci
- name: Install dependencies
run: bun install --frozen-lockfile
env:
GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
- name: Install webview-ui dependencies
run: cd webview-ui && npm ci
- name: Run Quality Checks (Parallel)
run: npm run ci:check-all
run: bun run ci:check-all
test:
needs: quality-checks
@@ -59,66 +52,54 @@ jobs:
- name: Checkout code
uses: actions/checkout@v4
- name: Setup Node.js environment
uses: actions/setup-node@v4
- name: Setup Bun
uses: oven-sh/setup-bun@v2
with:
node-version: 22
cache: 'npm'
cache-dependency-path: |
package-lock.json
webview-ui/package-lock.json
bun-version: latest
- name: Install root dependencies
run: npm ci
- name: Install dependencies
run: bun install --frozen-lockfile
env:
GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
- name: Install webview-ui dependencies
run: cd webview-ui && npm ci
- name: Set up NPM on Windows
if: runner.os == 'Windows'
run: |
npm config set script-shell "C:\\Program Files\\Git\\bin\\bash.exe"
# Build the extension and tests (without redundant checks)
- name: Build Tests and Extension
id: build_step
run: npm run ci:build
run: bun run ci:build
- name: Unit Tests with coverage - Linux
id: unit_tests_linux
if: ${{ !cancelled() && steps.build_step.outcome == 'success' && runner.os == 'Linux' }}
run: |
npx nyc --nycrc-path .nycrc.unit.json --reporter=lcov npm run test:unit
bunx nyc --nycrc-path .nycrc.unit.json --reporter=lcov bun run test:unit
- name: Unit Tests - Non-Linux
id: unit_tests_non_linux
if: ${{ !cancelled() && steps.build_step.outcome == 'success' && runner.os != 'Linux' }}
run: |
npm run test:unit
bun run test:unit
- name: Extension Integration Tests - Linux
id: integration_tests_linux
if: ${{ !cancelled() && steps.build_step.outcome == 'success' && runner.os == 'Linux' }}
run: xvfb-run -a npm run test:coverage
run: xvfb-run -a bun run test:coverage
- name: Extension Integration Tests - Non-Linux
id: integration_tests_non_linux
if: ${{ !cancelled() && steps.build_step.outcome == 'success' && runner.os != 'Linux' }}
run: npm run test:integration
run: bun run test:integration
- name: Webview Tests with Coverage
id: webview_tests
if: ${{ !cancelled() && steps.build_step.outcome == 'success' }}
run: |
cd webview-ui
npm run test:coverage
bun run test:coverage
- name: CLI Tests
id: cli_tests
if: ${{ !cancelled() && steps.build_step.outcome == 'success' }}
run: cd cli && npm run test:run
run: cd cli && bun run test:run
- name: Save Coverage Reports
uses: actions/upload-artifact@v4
@@ -137,36 +118,28 @@ jobs:
- name: Checkout code
uses: actions/checkout@v4
- name: Setup Node.js environment
uses: actions/setup-node@v4
- name: Setup Bun
uses: oven-sh/setup-bun@v2
with:
node-version: 22
cache: 'npm'
cache-dependency-path: |
package-lock.json
webview-ui/package-lock.json
testing-platform/package-lock.json
bun-version: latest
- name: Install root dependencies
run: npm ci
- name: Install dependencies
run: bun install --frozen-lockfile
env:
GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
- name: Install webview-ui dependencies
run: cd webview-ui && npm ci
- name: Download ripgrep binaries
run: npm run download-ripgrep
run: bun run download-ripgrep
- name: Compile Standalone
run: npm run compile-standalone
run: bun run compile-standalone
- name: Install testing platform dependencies
run: cd testing-platform && npm ci
run: cd testing-platform && bun install
- name: Running testing platform integration spec tests
timeout-minutes: 7
run: npm run test:tp-orchestrator -- tests/specs/ --count=1 --coverage
run: bun run test:tp-orchestrator -- tests/specs/ --count=1 --coverage
- name: Save Coverage Reports
uses: actions/upload-artifact@v4
+3 -3
View File
@@ -10,7 +10,10 @@ tmp
.idea
.husky/_/
# Package manager lock files (we use Bun, so bun.lockb is NOT ignored)
pnpm-lock.yaml
package-lock.json
yarn.lock
.clineignore
.venv
@@ -48,6 +51,3 @@ test-results
.secrets
*.tsbuildinfo
# Smoke test results (generated)
evals/smoke-tests/results/
-3
View File
@@ -1,3 +0,0 @@
[submodule "evals/cline-bench"]
path = evals/cline-bench
url = https://github.com/cline/cline-bench.git
+91 -50
View File
@@ -5,8 +5,12 @@
"tasks": [
{
"label": "compile-standalone",
"type": "npm",
"script": "compile-standalone",
"type": "shell",
"command": "bun",
"args": [
"run",
"compile-standalone"
],
"group": "build",
"problemMatcher": [],
"presentation": {
@@ -14,9 +18,13 @@
}
},
{
"label": "npm: protos",
"type": "npm",
"script": "protos",
"label": "bun:protos",
"type": "shell",
"command": "bun",
"args": [
"run",
"protos"
],
"problemMatcher": [],
"isBackground": false,
"presentation": {
@@ -31,11 +39,11 @@
{
"label": "watch",
"dependsOn": [
"npm: protos",
"npm: build:webview",
"npm: dev:webview",
"npm: watch:tsc",
"npm: watch:esbuild"
"bun:protos",
"bun:build:webview",
"bun:dev:webview",
"bun:watch:tsc",
"bun:watch:esbuild"
],
"presentation": {
"reveal": "always"
@@ -48,11 +56,11 @@
{
"label": "watch:test",
"dependsOn": [
"npm: protos",
"npm: build:webview:test",
"npm: dev:webview",
"npm: watch:tsc",
"npm: watch:esbuild:test"
"bun:protos",
"bun:build:webview:test",
"bun:dev:webview",
"bun:watch:tsc",
"bun:watch:esbuild:test"
],
"presentation": {
"reveal": "always"
@@ -60,14 +68,18 @@
"group": "build"
},
{
"type": "npm",
"script": "build:webview",
"type": "shell",
"command": "bun",
"args": [
"run",
"build:webview"
],
"group": "build",
"problemMatcher": [],
"isBackground": true,
"label": "npm: build:webview",
"label": "bun:build:webview",
"dependsOn": [
"npm: protos"
"bun:protos"
],
"presentation": {
"group": "watch",
@@ -80,14 +92,18 @@
}
},
{
"type": "npm",
"script": "build:webview:test",
"type": "shell",
"command": "bun",
"args": [
"run",
"build:webview:test"
],
"group": "build",
"problemMatcher": [],
"isBackground": true,
"label": "npm: build:webview:test",
"label": "bun:build:webview:test",
"dependsOn": [
"npm: protos"
"bun:protos"
],
"presentation": {
"group": "watch",
@@ -101,8 +117,12 @@
}
},
{
"type": "npm",
"script": "dev:webview",
"type": "shell",
"command": "bun",
"args": [
"run",
"dev:webview"
],
"group": "build",
"problemMatcher": [
{
@@ -122,9 +142,9 @@
}
],
"isBackground": true,
"label": "npm: dev:webview",
"label": "bun:dev:webview",
"dependsOn": [
"npm: protos"
"bun:protos"
],
"presentation": {
"group": "watch",
@@ -137,8 +157,12 @@
}
},
{
"type": "npm",
"script": "watch:esbuild",
"type": "shell",
"command": "bun",
"args": [
"run",
"watch:esbuild"
],
"group": "build",
"problemMatcher": {
"pattern": [
@@ -160,9 +184,9 @@
}
},
"isBackground": true,
"label": "npm: watch:esbuild",
"label": "bun:watch:esbuild",
"dependsOn": [
"npm: protos"
"bun:protos"
],
"presentation": {
"group": "watch",
@@ -175,8 +199,12 @@
}
},
{
"type": "npm",
"script": "watch:esbuild:test",
"type": "shell",
"command": "bun",
"args": [
"run",
"watch:esbuild:test"
],
"group": "build",
"problemMatcher": {
"pattern": [
@@ -198,9 +226,9 @@
}
},
"isBackground": true,
"label": "npm: watch:esbuild:test",
"label": "bun:watch:esbuild:test",
"dependsOn": [
"npm: protos"
"bun:protos"
],
"presentation": {
"group": "watch",
@@ -214,14 +242,18 @@
}
},
{
"type": "npm",
"script": "watch:tsc",
"type": "shell",
"command": "bun",
"args": [
"run",
"watch:tsc"
],
"group": "build",
"problemMatcher": "$tsc-watch",
"isBackground": true,
"label": "npm: watch:tsc",
"label": "bun:watch:tsc",
"dependsOn": [
"npm: protos"
"bun:protos"
],
"presentation": {
"group": "watch",
@@ -229,12 +261,17 @@
}
},
{
"type": "npm",
"script": "watch-tests",
"type": "shell",
"command": "bun",
"args": [
"run",
"watch-tests"
],
"label": "bun:watch-tests",
"problemMatcher": "$tsc-watch",
"isBackground": true,
"dependsOn": [
"npm: protos"
"bun:protos"
],
"presentation": {
"reveal": "always",
@@ -245,9 +282,9 @@
{
"label": "tasks: watch-tests",
"dependsOn": [
"npm: protos",
"npm: watch",
"npm: watch-tests"
"bun:protos",
"bun:watch",
"bun:watch-tests"
],
"problemMatcher": []
},
@@ -265,15 +302,19 @@
"command": "rm -rf ${workspaceFolder}/dist/tmp/user && mkdir -p ${workspaceFolder}/dist/tmp/user"
},
{
"type": "npm",
"script": "storybook",
"type": "shell",
"command": "bun",
"args": [
"run",
"storybook"
],
"group": "build",
"problemMatcher": [],
"isBackground": false,
"label": "npm: storybook",
"label": "bun:storybook",
"dependsOn": [
"npm: protos",
"npm: build:webview"
"bun:protos",
"bun:build:webview"
],
"presentation": {
"reveal": "always"
-86
View File
@@ -1,91 +1,5 @@
# Changelog
## [3.63.0]
### Added
- added zai GLM 5 Free promo
### Fixed
- Restore reasoning trace visibility in chat and improve the thinking row UX so reasoning is visible, then collapsible after completion.
## [3.62.0]
### Fixed
- Banners now display immediately when opening the extension instead of requiring user interaction first
- Resolved 17 security vulnerabilities including high-severity DoS issues in dependencies (body-parser, axios, qs, tar, and others)
## [3.61.0]
- UI/UX fixes with minimax model family
## [3.60.0]
- Fixes for Minimax model family
## [3.59.0]
- Added Minimax 2.5 Free Promo
- Fixed Response chaining for OpenAI's Responses API
## [3.58.0]
### Added
- Subagent: replace legacy subagents with the native `use_subagents` tool
- Bundle `endpoints.json` support so packaged distributions can ship required endpoints out-of-the-box
- Amazon Bedrock: support parallel tool calling
- New "double-check completion" experimental feature to verify work before marking tasks complete
- CLI: new task controls/flags including custom `--thinking` token budget and `--max-consecutive-mistakes` for yolo runs
- Remote config: new UI/options (including connection/test buttons) and support for syncing deletion of remotely configured MCP servers
- Vertex / Claude Code: add 1M context model options for Claude Opus 4.6
- ZAI/GLM: add GLM-5
### Fixed
- CLI: handle stdin redirection correctly in CI/headless environments
- CLI: preserve OAuth callback paths during auth redirects
- VS Code Web: generate auth callback URLs via `vscode.env.asExternalUri` (OAuth callback reliability)
- Terminal: surface command exit codes in results and improve long-running `execute_command` timeout behavior
- UI: add loading indicator and fix `api_req_started` rendering
- Task streaming: prevent duplicate streamed text rows after completion
- API: preserve selected Vercel model when model metadata is missing
- Telemetry: route PostHog networking through proxy-aware shared fetch and ensure telemetry flushes on shutdown
- CI: increase Windows E2E test timeout to reduce flakiness
### Changed
- Settings/model UX: move "reasoning effort" into model configuration and expose it in settings
- CLI provider selection: limit provider list to those remotely configured
- UI: consolidate ViewHeader component/styling across views
- Tools: add auto-approval support for `attempt_completion` commands
- Remotely configured MCP server schema now supports custom headers
## [3.57.1]
### Fixed
- Fixed Opus 4.6 for bedrock provider
## [3.57.0]
### Added
- Cline CLI 2.0 now available. Install with `npm install -g cline`
- Anthopic Opus 4.6
- Minimax-2.1 and Kimi-k2.5 now available for free for a limited time promo
- Codex-5.3 through ChatGPT subscription
### Fixed
- Fix read file tool to support reading large files
- Fix decimal input crash in OpenAI Compatible price fields (#8129)
- Fix build complete handlers when updating the api config
- Fixed missing provider from list
- Fixed Favorite Icon / Star from getting clipped in the task history view
### Changed
- Make skills always enabled and remove feature toggle setting
## [3.56.0]
### Added
+38 -22
View File
@@ -34,23 +34,38 @@ We also welcome contributions to our [documentation](https://github.com/cline/cl
### Local Development Instructions
1. Clone the repository _(Requires [git-lfs](https://git-lfs.com/))_:
> **Note**: This project uses [Bun](https://bun.sh) as the package manager.
1. Install Bun (if you haven't already):
```bash
# macOS/Linux
curl -fsSL https://bun.sh/install | bash
# Windows
powershell -c "irm bun.sh/install.ps1 | iex"
```
2. Clone the repository _(Requires [git-lfs](https://git-lfs.com/))_:
```bash
git clone https://github.com/cline/cline.git
```
2. Open the project in VSCode:
3. Open the project in VSCode:
```bash
code cline
```
3. Install the necessary dependencies for the extension and webview-gui:
4. Install dependencies (installs for all workspaces):
```bash
npm run install:all
bun install
```
4. Generate Protocol Buffer files (required before first build):
5. Generate Protocol Buffer files (required before first build):
```bash
npm run protos
bun run protos
```
5. Launch by pressing `F5` (or `Run`->`Start Debugging`) to open a new VSCode window with the extension loaded. (You may need to install the [esbuild problem matchers extension](https://marketplace.visualstudio.com/items?itemName=connor4312.esbuild-problem-matchers) if you run into issues building the project.)
6. Launch by pressing `F5` (or `Run`->`Start Debugging`) to open a new VSCode window with the extension loaded. (You may need to install the [esbuild problem matchers extension](https://marketplace.visualstudio.com/items?itemName=connor4312.esbuild-problem-matchers) if you run into issues building the project.)
@@ -59,7 +74,7 @@ We also welcome contributions to our [documentation](https://github.com/cline/cl
1. Before creating a PR, generate a changeset entry:
```bash
npm run changeset
bun run changeset
```
This will prompt you for:
- Type of change (major, minor, patch)
@@ -75,9 +90,10 @@ We also welcome contributions to our [documentation](https://github.com/cline/cl
- Changesetbot will create a comment showing the version impact
- When merged to main, changesetbot will create a Version Packages PR
- When the Version Packages PR is merged, a new release will be published
4. Testing
- Run `npm run test` to run tests locally.
- Before submitting PR, run `npm run format:fix` to format your code
- Run `bun run test` to run tests locally
- Before submitting PR, run `bun run format:fix` to format your code
### Extension
@@ -88,12 +104,12 @@ We also welcome contributions to our [documentation](https://github.com/cline/cl
- If you dismissed the prompts, you can install them manually from the Extensions panel
2. **Local Development**
- Run `npm run install:all` to install dependencies
- Run `npm run protos` to generate Protocol Buffer files (required before first build)
- Run `npm run test` to run tests locally
- Run `bun run install:all` to install dependencies
- Run `bun run protos` to generate Protocol Buffer files (required before first build)
- Run `bun run test` to run tests locally
- Run → Start Debugging or `>Debug: Select and Start Debugging` and wait for a new VS Code instance to open
- **Terminal Workflow**: Use `npm run dev` (generates protos + runs watch mode) or `npm run watch` (if protos already generated)
- Before submitting PR, run `npm run format:fix` to format your code
- **Terminal Workflow**: Use `bun run dev` (generates protos + runs watch mode) or `bun run watch` (if protos already generated)
- Before submitting PR, run `bun run format:fix` to format your code
3. **Linux-specific Setup**
VS Code extension tests on Linux require the following system libraries:
@@ -149,8 +165,8 @@ Anyone can contribute code to Cline, but we ask that you follow these guidelines
2. **Code Quality**
- Run `npm run lint` to check code style
- Run `npm run format` to automatically format code
- Run `bun run lint` to check code style
- Run `bun run format` to automatically format code
- All PRs must pass CI checks which include both linting and formatting
- Address any warnings or errors from linter before submitting
- Follow TypeScript best practices and maintain type safety
@@ -158,7 +174,7 @@ Anyone can contribute code to Cline, but we ask that you follow these guidelines
3. **Testing**
- Add tests for new features
- Run `npm test` to ensure all tests pass
- Run `bun test` to ensure all tests pass
- Update existing tests if your changes affect them
- Include both unit tests and integration tests where appropriate
@@ -168,9 +184,9 @@ Anyone can contribute code to Cline, but we ask that you follow these guidelines
- **Running E2E tests:**
```bash
npm run test:e2e # Build and run all E2E tests
npm run e2e # Run tests without rebuilding
npm run test:e2e -- --debug # Run with interactive debugger
bun run test:e2e # Build and run all E2E tests
bun run e2e # Run tests without rebuilding
bun run test:e2e -- --debug # Run with interactive debugger
```
- **Writing E2E tests:**
@@ -194,7 +210,7 @@ Anyone can contribute code to Cline, but we ask that you follow these guidelines
4. **Version Management with Changesets**
- Create a changeset for any user-facing changes using `npm run changeset`
- Create a changeset for any user-facing changes using `bun run changeset`
- Choose the appropriate version bump:
- `major` for breaking changes (1.0.0 → 2.0.0)
- `minor` for new features (1.0.0 → 1.1.0)
+69 -76
View File
@@ -1,5 +1,5 @@
{
"$schema": "./node_modules/@biomejs/biome/configuration_schema.json",
"$schema": "https://biomejs.dev/schemas/2.1.4/schema.json",
"vcs": {
"enabled": true,
"clientKind": "git",
@@ -28,19 +28,19 @@
"rules": {
"recommended": true,
"correctness": {
"useExhaustiveDependencies": "info",
"useExhaustiveDependencies": "off",
"noUndeclaredVariables": "off",
"noEmptyPattern": "info",
"noEmptyPattern": "off",
"useJsxKeyInIterable": "off",
"noInnerDeclarations": "off",
"useHookAtTopLevel": "info",
"useYield": "info",
"useHookAtTopLevel": "off",
"useYield": "off",
"noConstructorReturn": "off",
"noInvalidPositionAtImportRule": "off",
"noSwitchDeclarations": "off",
"noUnusedImports": "error"
},
"a11y": "info",
"a11y": "off",
"style": {
"useNodejsImportProtocol": "off",
"useImportType": "off",
@@ -51,36 +51,35 @@
"noParameterAssign": "off",
"useAsConstAssertion": "off",
"useDefaultParameterLast": "off",
"noNonNullAssertion": "info",
"noNonNullAssertion": "off",
"useEnumInitializers": "off",
"useSelfClosingElements": "info",
"useSelfClosingElements": "off",
"useSingleVarDeclarator": "off",
"useNumberNamespace": "info",
"noInferrableTypes": "info",
"useTemplate": "info",
"noUselessElse": "info"
"useNumberNamespace": "off",
"noInferrableTypes": "off",
"useTemplate": "off",
"noUselessElse": "off"
},
"suspicious": {
"noDoubleEquals": "warn",
"noImplicitAnyLet": "info",
"noThenProperty": "off",
"noAsyncPromiseExecutor": "info",
"noAsyncPromiseExecutor": "off",
"noImportAssign": "off",
"noExplicitAny": "info",
"noControlCharactersInRegex": "warn",
"noExplicitAny": "off",
"noControlCharactersInRegex": "off",
"noShadowRestrictedNames": "off",
"noArrayIndexKey": "info",
"noAssignInExpressions": "info",
"useIterableCallbackReturn": "info"
"noAssignInExpressions": "info"
},
"complexity": {
"noUselessConstructor": "info",
"useOptionalChain": "info",
"noBannedTypes": "warn",
"useLiteralKeys": "info",
"noUselessCatch": "info",
"noUselessSwitchCase": "info",
"noStaticOnlyClass": "info"
"noUselessConstructor": "off",
"useOptionalChain": "off",
"noBannedTypes": "off",
"useLiteralKeys": "off",
"noUselessCatch": "off",
"noUselessSwitchCase": "off",
"noStaticOnlyClass": "off"
},
"security": {
"noDangerouslySetInnerHtml": "info"
@@ -95,11 +94,6 @@
"lineEnding": "lf",
"formatWithErrors": true
},
"css": {
"parser": {
"tailwindDirectives": true
}
},
"javascript": {
"formatter": {
"semicolons": "asNeeded",
@@ -118,21 +112,21 @@
}
},
"files": {
"ignoreUnknown": true,
"includes": [
"**",
// explicitly force files to be ignored by the scanner with !!
"!!**/dist",
"!!**/dist-*",
"!!**/out",
"!!**/evals",
"!!**/playwright",
"!!**/test-results",
"!!**/node_modules",
"!!**/webview-ui/build",
"!!**/generated",
"!!**/proto",
"!!**/tests/specs"
"!**/dist",
"!**/dist-*",
"!**/out",
"!**/evals",
"!**/playwright",
"!**/test-results",
"!**/node_modules",
"!**/webview-ui/build",
"!**/generated",
"!**/proto",
"!**/tests/specs",
"!**/*.lock*",
"!**/*-lock.json"
]
},
"plugins": [
@@ -142,15 +136,14 @@
{
"includes": [
"**",
"!!**/dist",
"!!**/hosts/vscode/**",
"!!**/test/**",
"!!**/*.test.ts",
"!!src/dev/**",
"!!src/extension.ts",
"!!src/integrations/git/commit-message-generator.ts",
"!!src/integrations/terminal/**",
"!!src/core/controller/ui/openWalkthrough.ts"
"!**/hosts/vscode/**",
"!**/test/**",
"!**/*.test.ts",
"!src/dev/**",
"!src/extension.ts",
"!src/integrations/git/commit-message-generator.ts",
"!src/integrations/terminal/**",
"!src/core/controller/ui/openWalkthrough.ts"
],
"plugins": [
"src/dev/grit/vscode-api.grit"
@@ -163,37 +156,37 @@
],
"includes": [
"**",
"!!**/esbuild.*",
"!!**/*.mts",
"!!**/webview-ui/**",
"!!**/evals/**",
"!!**/standalone/**",
"!!**/cli/**",
"!!**/e2e/**",
"!!**/test/**",
"!!**/__tests__/**",
"!!**/*.test.ts",
"!!**/*.stories.ts",
"!!src/dev/**",
"!!**/*.mjs",
"!!**/*.js",
"!!**/scripts/**",
"!!**/*.tsx",
"!!**/testing-platform/**",
"!**/esbuild.*",
"!**/*.mts",
"!**/webview-ui/**",
"!**/evals/**",
"!**/standalone/**",
"!**/cli/**",
"!**/e2e/**",
"!**/test/**",
"!**/__tests__/**",
"!**/*.test.ts",
"!**/*.stories.ts",
"!src/dev/**",
"!**/*.mjs",
"!**/*.js",
"!**/scripts/**",
"!**/*.tsx",
"!**/testing-platform/**",
// ACP mode must redirect console to stderr - this is intentional
"!!cli/src/acp/index.ts"
"!cli/src/acp/index.ts"
]
},
{
"includes": [
"**",
"!!src/core/storage/state-migrations.ts",
"!!src/core/storage/FileContextTracker.ts",
"!!src/core/context/context-tracking/FileContextTracker.ts",
"!!src/common.ts",
"!!src/services/logging/distinctId.ts",
"!!src/core/storage/utils/state-helpers.ts",
"!!src/extension.ts"
"!src/core/storage/state-migrations.ts",
"!src/core/storage/FileContextTracker.ts",
"!src/core/context/context-tracking/FileContextTracker.ts",
"!src/common.ts",
"!src/services/logging/distinctId.ts",
"!src/core/storage/utils/state-helpers.ts",
"!src/extension.ts"
],
"plugins": [
"src/dev/grit/use-cache-service.grit"
+6045
View File
File diff suppressed because it is too large Load Diff
+18
View File
@@ -0,0 +1,18 @@
# Bun configuration for Cline monorepo
[install]
# Use npm for package resolution compatibility
registry = "https://registry.npmjs.org/"
# Cache configuration
[install.cache]
# Enable caching for faster installs
disable = false
[install.scopes]
# Configure scoped registries if needed
# "@myorg" = "https://registry.example.com/"
[test]
# Test configuration
preload = []
-95
View File
@@ -1,95 +0,0 @@
# cline
## [2.4.0]
### Changes
- Adding Anthropic Sonnet 4.6
## [2.3.0]
### Patch Changes
- Allows users to enter custom aws region when selecting bedrock as a provider in CLI
- Keep reasoning rows visible when low-stakes tool groups start immediately after reasoning.
- Restore reasoning trace visibility in chat and improve the thinking row UX so streamed reasoning is visible, then collapsible after completion.
### Fixed
- Banners now display immediately when opening the extension instead of requiring user interaction first
- Resolved 17 security vulnerabilities including high-severity DoS issues in dependencies (body-parser, axios, qs, tar, and others)
## [2.2.2]
- Allows users to enter custom aws region when selecting bedrock as a provider
- Prevent Parent Container Scrolling In Dropdowns
## [2.2.1]
- Added Minimax 2.5 Free Promo
- Fixed Response chaining for OpenAI's Responses API
## [2.2.0]
### Added
- Subagent: replace legacy subagents with the native `use_subagents` tool
- Bundle `endpoints.json` support so packaged distributions can ship required endpoints out-of-the-box
- Amazon Bedrock: support parallel tool calling
- New "double-check completion" experimental feature to verify work before marking tasks complete
- CLI: new task controls/flags including custom `--thinking` token budget and `--max-consecutive-mistakes` for yolo runs
- Remote config: new UI/options (including connection/test buttons) and support for syncing deletion of remotely configured MCP servers
- Vertex / Claude Code: add 1M context model options for Claude Opus 4.6
- ZAI/GLM: add GLM-5
### Fixed
- CLI: handle stdin redirection correctly in CI/headless environments
- CLI: preserve OAuth callback paths during auth redirects
- VS Code Web: generate auth callback URLs via `vscode.env.asExternalUri` (OAuth callback reliability)
- Terminal: surface command exit codes in results and improve long-running `execute_command` timeout behavior
- UI: add loading indicator and fix `api_req_started` rendering
- Task streaming: prevent duplicate streamed text rows after completion
- API: preserve selected Vercel model when model metadata is missing
- Telemetry: route PostHog networking through proxy-aware shared fetch and ensure telemetry flushes on shutdown
- CI: increase Windows E2E test timeout to reduce flakiness
### Changed
- Settings/model UX: move "reasoning effort" into model configuration and expose it in settings
- CLI provider selection: limit provider list to those remotely configured
- UI: consolidate ViewHeader component/styling across views
- Tools: add auto-approval support for `attempt_completion` commands
- Remotely configured MCP server schema now supports custom headers
## [2.1.0]
### Minor Changes
- 42ce100: Add Generate API Key on Hicap Provider selection
### Patch Changes
- 195294f: Add support for bundled endpoints.json in enterprise distributions. Extensions can now include a pre-configured endpoints.json file that automatically switches Cline to self-hosted mode. Includes packaging scripts for VSIX, NPM, and JetBrains plugins.
- a1f2601: Replace the LiteLLM model list with a selector
- 739d75a: Add Claude Code provider support for Claude Opus 4.6 and Sonnet 4.5 1M variants via both full model names and aliases (`opus[1m]`, `sonnet[1m]`), and align the `opus` alias with Opus 4.6.
- 8440380: Add GitHub Actions workflow to build CLI from any commit for testing
- b1a8db2: fix(cli): prevent hang when spawned without TTY
- 7c87017: Add Claude Opus 4.6 model support
- d116ac5: Supports rendering markdown table in chat view.
- 6d8fb85: Fix CLI crashing in CI environments and with stdin redirection (e.g., `cline "prompt" < /dev/null`). Now checks both stdin and stdout TTY status before using Ink, and only errors on empty stdin when no prompt is provided.
- 70a9904: Fix JetBrains sign-in regression by adding fallback for openExternal RPC
- f440f3a: fix: use vscode.env.openExternal for auth in remote environments
Fixes OAuth authentication in VS Code Server and remote environments by routing browser URL opening through VS Code's native openExternal API instead of the npm 'open' package.
- 70a9904: fix: use vscode.env.asExternalUri for auth callback URLs only in VS Code Web
Fixes OAuth callback redirect in VS Code Web (`code serve-web`, Codespaces) by using `vscode.env.asExternalUri()` to resolve the callback URI. This is gated behind a `vscode.env.uiKind === UIKind.Web` check so regular desktop VS Code continues to use the `vscode://` URI directly. The `getCallbackUrl` API now accepts a `path` parameter so the full callback URI (including route) is resolved correctly, and callers pass their path directly instead of appending after.
- 5308ded: Updating script documentation and removing unnecessary continue on error
- b514f18: Prevent duplicate streamed text rows when a partial text update arrives after the same text was already finalized.
- 26391c9: Fix Bedrock model id
- d19a877: Unify ViewHeader Styles Across All Views
- 5dcaa8c: Add Vertex Claude Opus 4.6 1M model option and global endpoint support, and pass the 1M beta header for Vertex Claude requests.
+25 -25
View File
@@ -13,7 +13,7 @@ The official CLI for Cline. Run Cline tasks directly from the terminal with the
## Prerequisites
- Node.js 20.x or later
- npm or yarn
- bun or npm or yarn
- The parent Cline project dependencies installed
## Installation
@@ -22,13 +22,13 @@ From the repository root:
```bash
# Install all dependencies first
npm run install:all
bun install
# Ensure protos are generated
npm run protos
bun run protos
# Build and link the CLI globally
npm run cli:link
bun run cli:link
```
## Usage
@@ -192,10 +192,10 @@ These options are available for the default command (running a task directly):
```bash
# 1. Install all dependencies (root, webview-ui, cli)
npm run install:all
bun install
# 2. Build and link globally so you can run `cline` from anywhere
npm run cli:link
bun run cli:link
# 3. Test it
cline --help
@@ -207,29 +207,29 @@ Run these from the repository root:
| Script | Description |
|--------|-------------|
| `npm run install:all` | Install deps for root, webview-ui, and cli |
| `npm run cli:build` | Generate protos and build CLI |
| `npm run cli:build:production` | Production build (minified) |
| `npm run cli:link` | Build and `npm link` so you can run `cline` from anywhere |
| `npm run cli:unlink` | Remove the global `cline` symlink |
| `npm run cli:dev` | Link + watch mode for development |
| `npm run cli:watch` | Watch mode only (no initial build) |
| `npm run cli:test` | Run CLI tests |
| `bun install` | Install deps for root, webview-ui, and cli |
| `bun run cli:build` | Generate protos and build CLI |
| `bun run cli:build:production` | Production build (minified) |
| `bun run cli:link` | Build and `bun link` so you can run `cline` from anywhere |
| `bun run cli:unlink` | Remove the global `cline` symlink |
| `bun run cli:dev` | Link + watch mode for development |
| `bun run cli:watch` | Watch mode only (no initial build) |
| `bun run cli:test` | Run CLI tests |
### Development Workflow
1. Run `npm run cli:dev` - this links the CLI globally and starts watch mode
1. Run `bun run cli:dev` - this links the CLI globally and starts watch mode
2. Make changes to files in `cli/src/`
3. The build automatically rebuilds on save
4. Test your changes by running `cline` in another terminal
5. When done, run `npm run cli:unlink` to clean up
5. When done, run `bun run cli:unlink` to clean up
### Proto Generation
The CLI uses proto-generated types for message passing (same as the VS Code extension). If you modify any `.proto` files, run:
```bash
npm run protos
bun run protos
```
This generates TypeScript types in `src/generated/` that both the CLI and extension use.
@@ -238,12 +238,12 @@ This generates TypeScript types in `src/generated/` that both the CLI and extens
#### 1. Publish to npm
```bash
npm publish
bun publish
```
#### 2. Update the Homebrew formula
```bash
npm run update-brew-formula
bun run update-brew-formula
```
#### 3. Test the formula locally
@@ -335,13 +335,13 @@ If you encounter build errors:
```bash
# Make sure all deps are installed
npm run install:all
bun install
# Regenerate proto types
npm run protos
bun run protos
# Then rebuild
npm run cli:build
bun run cli:build
```
### "command not found: cline"
@@ -349,16 +349,16 @@ npm run cli:build
The CLI isn't linked globally. Run:
```bash
npm run cli:link
bun run cli:link
```
### Changes Not Reflected
If your code changes aren't showing up:
1. Make sure watch mode is running (`npm run cli:dev`)
1. Make sure watch mode is running (`bun run cli:dev`)
2. Check for TypeScript errors in the watch output
3. Try unlinking and relinking: `npm run cli:unlink && npm run cli:link`
3. Try unlinking and relinking: `bun run cli:unlink && bun run cli:link`
### Import Errors from Core
+2 -1
View File
@@ -45,7 +45,7 @@ cline
### Use any API and Model
Cline supports API providers like ChatGPT, Anthropic, OpenAI, Google Gemini, AWS Bedrock, Azure, GCP Vertex, Cerebras, Groq, and Moonshot. You can also configure any OpenAI compatible API, or use a local model through LM Studio/Ollama. If you're using a Cline Account, you'll always have access to the newest models as soon as they're available.
Cline supports API providers like ChatGPT, Anthropic, OpenAI, Google Gemini, AWS Bedrock, Azure, GCP Vertex, Cerebras and Groq. You can also configure any OpenAI compatible API, or use a local model through LM Studio/Ollama. If you're using a Cline Account, you'll always have access to the newest models as soon as they're available.
<!-- Transparent pixel to create line break after floating image -->
@@ -79,3 +79,4 @@ Get the same Cline experience with enterprise-grade controls: SSO (SAML/OIDC), g
## License
[Apache 2.0 © 2026 Cline Bot Inc.](./LICENSE)
+15 -43
View File
@@ -88,10 +88,6 @@ directory
\f[B]\-\-thinking\f[R] : Enable extended thinking (1024 token budget)
.PP
\f[B]\-\-json\f[R] : Output messages as JSON instead of styled text
.PP
\f[B]\-T\f[R], \f[B]\-\-taskId\f[R] \f[I]id\f[R] : Resume an existing
task by ID.
The prompt argument becomes an optional follow\-up message.
.SS history (alias: h)
List task history with pagination.
.PP
@@ -125,13 +121,13 @@ authentication wizard, or use quick setup flags.
Options:
.PP
\f[B]\-p\f[R], \f[B]\-\-provider\f[R] \f[I]id\f[R] : Provider ID for
quick setup (e.g., openai\-native, anthropic, openrouter, moonshot)
quick setup (e.g., openai\-native, anthropic, openrouter)
.PP
\f[B]\-k\f[R], \f[B]\-\-apikey\f[R] \f[I]key\f[R] : API key for the
provider
.PP
\f[B]\-m\f[R], \f[B]\-\-modelid\f[R] \f[I]id\f[R] : Model ID to
configure (e.g., gpt\-4o, claude\-sonnet\-4\-5\-20250929, kimi\-k2.5)
configure (e.g., gpt\-4o, claude\-sonnet\-4\-5\-20250929)
.PP
\f[B]\-b\f[R], \f[B]\-\-baseurl\f[R] \f[I]url\f[R] : Base URL (optional,
for OpenAI\-compatible providers)
@@ -183,10 +179,6 @@ the task
.PP
\f[B]\-\-json\f[R] : Output messages as JSON instead of styled text.
Forces plain text mode.
.PP
\f[B]\-T\f[R], \f[B]\-\-taskId\f[R] \f[I]id\f[R] : Resume an existing
task by ID instead of starting a new one.
The prompt becomes an optional follow\-up message.
.SH JSON OUTPUT FORMAT
When using \f[B]\-\-json\f[R], each message is output as a JSON object
with these fields:
@@ -242,9 +234,6 @@ cline \-m claude\-sonnet\-4\-5\-20250929 \(dqRefactor this function\(dq
\f[I]# Quick auth setup with model\f[R]
cline auth \-p anthropic \-k sk\-ant\-xxxxx \-m claude\-sonnet\-4\-5\-20250929
\f[I]# Quick auth setup for Moonshot\f[R]
cline auth \-p moonshot \-k sk\-xxxxx \-m kimi\-k2.5
.EE
.SS Including Images
.IP
@@ -285,21 +274,6 @@ cline history
\f[I]# Show more tasks with pagination\f[R]
cline history \-n 20 \-p 2
.EE
.SS Resuming Tasks
.IP
.EX
\f[I]# Resume a task by ID (get IDs from cline history)\f[R]
cline \-T abc123def
\f[I]# Resume a task with a follow\-up message\f[R]
cline \-T abc123def \(dqNow add unit tests for the changes\(dq
\f[I]# Resume in plan mode to review before continuing\f[R]
cline \-T abc123def \-p \(dqWhat\(aqs left to do?\(dq
\f[I]# Resume with yolo mode for automated continuation\f[R]
cline \-T abc123def \-y \(dqContinue with the implementation\(dq
.EE
.SS Authentication
.IP
.EX
@@ -312,9 +286,6 @@ cline auth \-p anthropic \-k sk\-ant\-api\-xxxxx
\f[I]# Quick setup for OpenAI\f[R]
cline auth \-p openai\-native \-k sk\-xxxxx \-m gpt\-4o
\f[I]# Quick setup for Moonshot\f[R]
cline auth \-p moonshot \-k sk\-xxxxx \-m kimi\-k2.5
\f[I]# OpenAI\-compatible provider with custom base URL\f[R]
cline auth \-p openai \-k your\-api\-key \-b https://api.example.com/v1
.EE
@@ -377,19 +348,20 @@ export CLINE_COMMAND_PERMISSIONS=\(aq{\(dqallow\(dq: [\(dqnpm *\(dq, \(dqgit *\(
\f[I]# Allow file operations with redirects\f[R]
export CLINE_COMMAND_PERMISSIONS=\(aq{\(dqallow\(dq: [\(dqcat *\(dq, \(dqecho *\(dq], \(dqallowRedirects\(dq: true}\(aq
.EE
.SH CONFIGURATION FILES
.IP
.EX
\(ti/.cline/
├── data/ # Default configuration directory
│ ├── globalState.json # Global settings and state
│ ├── secrets.json # API keys and secrets (stored securely)
│ ├── workspace/ # Workspace\-specific state
│ └── tasks/ # Task history and conversation data
└── log/ # Log files for debugging
.EE
.SH FILES
\f[B]\(ti/.cline/data/\f[R] : Default configuration directory
containing:
.PP
View logs with \f[CR]cline dev log\f[R].
\f[B]globalState.json\f[R] : Global settings and state
.PP
\f[B]secrets.json\f[R] : API keys and secrets (stored securely)
.PP
\f[B]workspace/\f[R] : Workspace\-specific state
.PP
\f[B]tasks/\f[R] : Task history and conversation data
.PP
\f[B]\(ti/.cline/log/\f[R] : Log files for debugging.
View with \f[CR]cline dev log\f[R].
.SH BUGS
Report bugs at: \c
.UR https://github.com/cline/cline/issues
+4 -24
View File
@@ -70,8 +70,6 @@ Run a new task with a prompt.
**\--json** : Output messages as JSON instead of styled text
**-T**, **\--taskId** *id* : Resume an existing task by ID. The prompt argument becomes an optional follow-up message.
## history (alias: h)
List task history with pagination.
@@ -118,7 +116,7 @@ Authenticate a provider and configure the model.
Check for updates and install if available.
**cline update** [*options*] : Check npm for newer versions. Options:
**cline update** [*options*] : Check bun for newer versions. Options:
**-v**, **\--verbose** : Show verbose output
@@ -156,8 +154,6 @@ When running **cline** with just a prompt (no subcommand), these options are ava
**\--json** : Output messages as JSON instead of styled text. Forces plain text mode.
**-T**, **\--taskId** *id* : Resume an existing task by ID instead of starting a new one. The prompt becomes an optional follow-up message.
# JSON OUTPUT FORMAT
When using **\--json**, each message is output as a JSON object with these fields:
@@ -255,22 +251,6 @@ cline history
cline history -n 20 -p 2
```
## Resuming Tasks
```bash
# Resume a task by ID (get IDs from cline history)
cline -T abc123def
# Resume a task with a follow-up message
cline -T abc123def "Now add unit tests for the changes"
# Resume in plan mode to review before continuing
cline -T abc123def -p "What's left to do?"
# Resume with yolo mode for automated continuation
cline -T abc123def -y "Continue with the implementation"
```
## Authentication
```bash
@@ -313,11 +293,11 @@ Format: `{"allow": ["pattern1", "pattern2"], "deny": ["pattern3"], "allowRedirec
**Examples:**
```bash
# Allow only npm and git commands.
export CLINE_COMMAND_PERMISSIONS='{"allow": ["npm *", "git *"]}'
# Allow only bun and git commands.
export CLINE_COMMAND_PERMISSIONS='{"allow": ["bun *", "git *"]}'
# Allow development commands but deny dangerous ones. Deny not strictly required here since allow is set.
export CLINE_COMMAND_PERMISSIONS='{"allow": ["npm *", "git *", "node *"], "deny": ["rm -rf *", "sudo *"]}'
export CLINE_COMMAND_PERMISSIONS='{"allow": ["bun *", "git *", "node *"], "deny": ["rm -rf *", "sudo *"]}'
# Allow file operations with redirects
export CLINE_COMMAND_PERMISSIONS='{"allow": ["cat *", "echo *"], "allowRedirects": true}'
+10 -10
View File
@@ -1,6 +1,6 @@
{
"name": "cline",
"version": "2.4.0",
"version": "2.0.3",
"description": "Autonomous coding agent CLI - capable of creating/editing files, running commands, using the browser, and more",
"main": "dist/cli.mjs",
"bin": {
@@ -21,16 +21,16 @@
"node": ">=20.0.0"
},
"scripts": {
"package:brew": "npx tsx ./scripts/update-brew-formula.mts",
"package": "npm pack --pack-destination ./dist",
"build": "npm run typecheck && npx tsx esbuild.mts",
"build:production": "npm run typecheck && npx tsx esbuild.mts --production",
"watch": "npx tsx esbuild.mts --watch",
"dev": "IS_DEV=true && npm run link && npm run watch ; npm run unlink",
"package:brew": "bunx tsx ./scripts/update-brew-formula.mts",
"package": "bun pm pack --pack-destination ./dist",
"build": "bun run typecheck && bunx tsx esbuild.mts",
"build:production": "bun run typecheck && bunx tsx esbuild.mts --production",
"watch": "bunx tsx esbuild.mts --watch",
"dev": "IS_DEV=true && bun run link && bun run watch ; bun run unlink",
"clean": "rimraf dist",
"typecheck": "npx tsc --noEmit",
"link": "npm run build && npm link",
"unlink": "npm unlink -g cline",
"typecheck": "bunx tsc --noEmit",
"link": "bun run build && bun link",
"unlink": "bun unlink",
"test": "vitest",
"test:run": "vitest run"
},
@@ -172,16 +172,6 @@ class ACPEnvServiceClient implements EnvServiceClientInterface {
Logger.debug("[ACPEnvServiceClient] shutdown called (stub)")
return proto.cline.Empty.create()
}
async openExternal(request: proto.cline.StringRequest): Promise<proto.cline.Empty> {
const url = request.value || ""
if (url) {
Logger.debug(`[ACPEnvServiceClient] openExternal: ${url}`)
const { openUrlInBrowser } = await import("../utils/browser")
await openUrlInBrowser(url)
}
return proto.cline.Empty.create()
}
}
/**
+28 -10
View File
@@ -12,7 +12,11 @@
import type * as acp from "@agentclientprotocol/sdk"
import type { TerminalHandle } from "@agentclientprotocol/sdk"
import { DEFAULT_TERMINAL_OUTPUT_LINE_LIMIT, PROCESS_HOT_TIMEOUT_NORMAL } from "@integrations/terminal/constants"
import {
DEFAULT_SUBAGENT_TERMINAL_OUTPUT_LINE_LIMIT,
DEFAULT_TERMINAL_OUTPUT_LINE_LIMIT,
PROCESS_HOT_TIMEOUT_NORMAL,
} from "@integrations/terminal/constants"
import type {
ITerminal,
ITerminalManager,
@@ -138,12 +142,12 @@ export interface ManagedTerminal {
* Wraps ACP terminal operations and emits events compatible with ITerminalProcess.
*/
class AcpTerminalProcess extends EventEmitter<TerminalProcessEvents> implements ITerminalProcess {
isHot = false
waitForShellIntegration = false
isHot: boolean = false
waitForShellIntegration: boolean = false
private _unretrievedOutput = ""
private _continued = false
private _completed = false
private _unretrievedOutput: string = ""
private _continued: boolean = false
private _completed: boolean = false
private _hotTimeout: NodeJS.Timeout | null = null
private _exitWaitTimeout: NodeJS.Timeout | null = null
private readonly manager: AcpTerminalManager
@@ -393,7 +397,7 @@ export class AcpTerminalManager implements ITerminalManager {
private readonly numericIdToStringId: Map<number, string> = new Map()
/** Next numeric ID to assign */
private nextNumericId = 1
private nextNumericId: number = 1
/** Active processes indexed by numeric terminal ID */
private readonly processes: Map<number, AcpTerminalProcess> = new Map()
@@ -402,8 +406,9 @@ export class AcpTerminalManager implements ITerminalManager {
private readonly terminalInfos: Map<number, TerminalInfo> = new Map()
// Configuration options for ITerminalManager
private terminalReuseEnabled = true
private terminalReuseEnabled: boolean = true
private terminalOutputLineLimit: number = DEFAULT_TERMINAL_OUTPUT_LINE_LIMIT
private subagentTerminalOutputLineLimit: number = DEFAULT_SUBAGENT_TERMINAL_OUTPUT_LINE_LIMIT
/**
* Creates a new AcpTerminalManager.
@@ -662,6 +667,14 @@ export class AcpTerminalManager implements ITerminalManager {
this.terminalOutputLineLimit = limit
}
/**
* Set the maximum number of output lines for subagent commands.
* @param limit Maximum number of lines
*/
setSubagentTerminalOutputLineLimit(limit: number): void {
this.subagentTerminalOutputLineLimit = limit
}
/**
* Set the default terminal profile.
* @param profile The profile identifier
@@ -674,10 +687,15 @@ export class AcpTerminalManager implements ITerminalManager {
* Process output lines, potentially truncating if over limit.
* @param outputLines Array of output lines
* @param overrideLimit Optional limit override
* @param isSubagentCommand Whether this is a subagent command
* @returns Processed output string
*/
processOutput(outputLines: string[], overrideLimit?: number): string {
const limit = overrideLimit !== undefined ? overrideLimit : this.terminalOutputLineLimit
processOutput(outputLines: string[], overrideLimit?: number, isSubagentCommand?: boolean): string {
const limit = isSubagentCommand
? overrideLimit !== undefined
? overrideLimit
: this.subagentTerminalOutputLineLimit
: this.terminalOutputLineLimit
if (outputLines.length > limit) {
const halfLimit = Math.floor(limit / 2)
+4 -7
View File
@@ -28,8 +28,6 @@ import {
groqModels,
mistralDefaultModelId,
mistralModels,
moonshotDefaultModelId,
moonshotModels,
openAiCodexDefaultModelId,
openAiNativeDefaultModelId,
openAiNativeModels,
@@ -74,7 +72,6 @@ const providerModels: Record<string, { models: Record<string, unknown>; defaultI
bedrock: { models: bedrockModels, defaultId: bedrockDefaultModelId },
deepseek: { models: deepSeekModels, defaultId: deepSeekDefaultModelId },
mistral: { models: mistralModels, defaultId: mistralDefaultModelId },
moonshot: { models: moonshotModels, defaultId: moonshotDefaultModelId },
groq: { models: groqModels, defaultId: groqDefaultModelId },
xai: { models: xaiModels, defaultId: xaiDefaultModelId },
}
@@ -176,7 +173,7 @@ export class ClineAgent implements acp.Agent {
async initialize(params: acp.InitializeRequest, connection?: acp.AgentSideConnection): Promise<acp.InitializeResponse> {
this.clientCapabilities = params.clientCapabilities
this.initializeHostProvider(this.clientCapabilities, connection)
await ClineEndpoint.initialize(this.ctx.EXTENSION_DIR)
await ClineEndpoint.initialize()
await StateManager.initialize(this.ctx.extensionContext)
return {
@@ -249,8 +246,8 @@ export class ClineAgent implements acp.Agent {
},
hostBridgeClientProvider,
(message: string) => Logger.info(message),
async (path: string) => {
return AuthHandler.getInstance().getCallbackUrl(path)
async () => {
return AuthHandler.getInstance().getCallbackUrl()
},
async () => "", // get binary location not needed in ACP mode
this.ctx.EXTENSION_DIR,
@@ -976,7 +973,7 @@ export class ClineAgent implements acp.Agent {
// Get the callback URL first to ensure the server is ready
let callbackUrl: string
try {
callbackUrl = await authHandler.getCallbackUrl("/auth")
callbackUrl = await authHandler.getCallbackUrl()
Logger.debug("[ClineAgent] Callback URL ready:", callbackUrl)
} catch (error) {
Logger.error("[ClineAgent] Failed to get callback URL:", error)
-4
View File
@@ -312,10 +312,6 @@ function translateSayMessage(
// API request finished - no specific update needed
break
case "subagent_usage":
// Hidden aggregate metrics event used for task-level accounting.
break
case "task":
// Task started - don't echo the user's prompt back to them
// The ACP client already knows what they typed
+8 -14
View File
@@ -34,7 +34,7 @@ type AsciiMotionCliProps = {
autoPlay?: boolean;
loop?: boolean;
onReady?: (api: PlaybackAPI) => void;
onInteraction?: () => void; // Called when user scrolls, clicks, or drags
onScroll?: () => void; // Called when user scrolls (scroll wheel)
};
const FRAMES: FrameData[] = [
@@ -333364,7 +333364,7 @@ const FRAME_BOTTOM_RIGHT = 128;
export const AsciiMotionCli: React.FC<AsciiMotionCliProps> = ({
hasDarkBackground = true,
onInteraction,
onScroll,
}) => {
const [frameIndex, setFrameIndex] = useState(0);
const [targetFrame, setTargetFrame] = useState(0);
@@ -333390,13 +333390,13 @@ export const AsciiMotionCli: React.FC<AsciiMotionCliProps> = ({
// Stop animation on terminal resize to prevent visual glitches
useEffect(() => {
const handleResize = () => {
onInteraction?.();
onScroll?.();
};
process.stdout.on("resize", handleResize);
return () => {
process.stdout.off("resize", handleResize);
};
}, [onInteraction]);
}, [onScroll]);
// Mouse tracking - gracefully handle environments without tty support
useEffect(() => {
@@ -333417,19 +333417,13 @@ export const AsciiMotionCli: React.FC<AsciiMotionCliProps> = ({
const handleData = (data: Buffer) => {
const str = data.toString();
// Parse mouse events: \x1b[<button;x;yM (M=press, m=release)
// Parse mouse events: \x1b[<button;x;yM
const mouseMatch = str.match(/\x1b\[<(\d+);(\d+);(\d+)([Mm])/);
if (mouseMatch) {
const button = parseInt(mouseMatch[1], 10);
const isPress = mouseMatch[4] === "M";
// Button 64/65 = scroll up/down
// Button 0-2 = left/middle/right click (on press)
// Button 32-34 = drag with left/middle/right button held
const isScroll = button === 64 || button === 65;
const isClick = isPress && button >= 0 && button <= 2;
const isDrag = button >= 32 && button <= 34;
if (isScroll || isClick || isDrag) {
onInteraction?.();
// Button 64 = scroll up, 65 = scroll down
if (button === 64 || button === 65) {
onScroll?.();
}
// Throttle cursor updates to ~20fps to reduce re-renders
const now = Date.now();
+24 -29
View File
@@ -6,21 +6,19 @@
import { Box, Text, useApp, useInput } from "ink"
import Spinner from "ink-spinner"
import React, { useCallback, useEffect, useMemo, useState } from "react"
import { refreshOcaModels } from "@/core/controller/models/refreshOcaModels"
import { StateManager } from "@/core/storage/StateManager"
import { openAiCodexOAuthManager } from "@/integrations/openai-codex/oauth"
import { AuthService } from "@/services/auth/AuthService"
import { openAiCodexDefaultModelId, openRouterDefaultModelId } from "@/shared/api"
import { StringRequest } from "@/shared/proto/cline/common"
import { liteLlmDefaultModelId, openAiCodexDefaultModelId, openRouterDefaultModelId } from "@/shared/api"
import { openExternal } from "@/utils/env"
import { COLORS } from "../constants/colors"
import { getAllFeaturedModels } from "../constants/featured-models"
import { useStdinContext } from "../context/StdinContext"
import { useOcaAuth } from "../hooks/useOcaAuth"
import { useScrollableList } from "../hooks/useScrollableList"
import { type DetectedSources, detectImportSources, type ImportSource } from "../utils/import-configs"
import { isMouseEscapeSequence } from "../utils/input"
import { applyBedrockConfig, applyProviderConfig } from "../utils/provider-config"
import { useValidProviders } from "../utils/providers"
import { ApiKeyInput } from "./ApiKeyInput"
import { StaticRobotFrame } from "./AsciiMotionCli"
import { type BedrockConfig, BedrockSetup } from "./BedrockSetup"
@@ -32,8 +30,7 @@ import {
} from "./FeaturedModelPicker"
import { ImportView } from "./ImportView"
import { getDefaultModelId, hasModelPicker, ModelPicker } from "./ModelPicker"
import { OcaEmployeeCheck } from "./OcaEmployeeCheck"
import { getProviderLabel } from "./ProviderPicker"
import { CLI_EXCLUDED_PROVIDERS, getProviderLabel, getProviderOrder } from "./ProviderPicker"
type AuthStep =
| "menu"
@@ -45,13 +42,15 @@ type AuthStep =
| "success"
| "error"
| "cline_auth"
| "oca_employee_check"
| "oca_auth"
| "cline_model"
| "openai_codex_auth"
| "bedrock"
| "import"
// Featured models loaded from shared constants
const featuredModels = getAllFeaturedModels()
interface AuthViewProps {
controller: any
onComplete?: () => void
@@ -150,9 +149,6 @@ const TextInput: React.FC<{
export const AuthView: React.FC<AuthViewProps> = ({ controller, onComplete, onError, onNavigateToWelcome }) => {
const { exit } = useApp()
const providers = useValidProviders()
const [step, setStep] = useState<AuthStep>("menu")
const [selectedProvider, setSelectedProvider] = useState<string>(
StateManager.get().getApiConfiguration().actModeApiProvider ||
@@ -163,6 +159,7 @@ export const AuthView: React.FC<AuthViewProps> = ({ controller, onComplete, onEr
const [modelId, setModelId] = useState("")
const [baseUrl, setBaseUrl] = useState("")
const [errorMessage, setErrorMessage] = useState("")
const [authStatus, setAuthStatus] = useState<string>("")
const [providerSearch, setProviderSearch] = useState("")
const [providerIndex, setProviderIndex] = useState(0)
const [clineModelIndex, setClineModelIndex] = useState(0)
@@ -173,14 +170,11 @@ export const AuthView: React.FC<AuthViewProps> = ({ controller, onComplete, onEr
// OCA auth hook - enabled when step is oca_auth
const handleOcaAuthSuccess = useCallback(async () => {
await applyProviderConfig({ providerId: "oca", controller })
// Fetch OCA models from the API - this sets actModeOcaModelId/planModeOcaModelId in state
await refreshOcaModels(controller, StringRequest.create({ value: "" }))
const stateManager = StateManager.get()
stateManager.setGlobalState("welcomeViewCompleted", true)
await stateManager.flushPendingState()
setSelectedProvider("oca")
const actModelId = stateManager.getGlobalSettingsKey("actModeOcaModelId") || ""
setModelId(actModelId)
setModelId(liteLlmDefaultModelId)
setStep("success")
}, [controller])
@@ -196,6 +190,11 @@ export const AuthView: React.FC<AuthViewProps> = ({ controller, onComplete, onEr
onError: handleOcaAuthError,
})
// Use providers.json order, filtered to exclude CLI-incompatible providers
const sortedProviders = useMemo(() => {
return getProviderOrder().filter((p) => !CLI_EXCLUDED_PROVIDERS.has(p))
}, [])
// Main menu items - conditionally include import options
const mainMenuItems: SelectItem[] = useMemo(() => {
const items: SelectItem[] = [{ label: "Sign in with Cline", value: "cline_auth" }]
@@ -221,13 +220,15 @@ export const AuthView: React.FC<AuthViewProps> = ({ controller, onComplete, onEr
const providerItems: SelectItem[] = useMemo(() => {
const search = providerSearch.toLowerCase()
const filtered = providerSearch
? providers.filter((p) => p.toLowerCase().includes(search) || getProviderLabel(p).toLowerCase().includes(search))
: providers
? sortedProviders.filter(
(p) => p.toLowerCase().includes(search) || getProviderLabel(p).toLowerCase().includes(search),
)
: sortedProviders
return filtered.map((p: string) => ({
label: getProviderLabel(p),
value: p,
}))
}, [providers, providerSearch])
}, [sortedProviders, providerSearch])
// Use shared scrollable list hook for provider windowing
const TOTAL_PROVIDER_ROWS = 8
@@ -322,6 +323,7 @@ export const AuthView: React.FC<AuthViewProps> = ({ controller, onComplete, onEr
const startClineAuth = useCallback(async () => {
try {
setStep("cline_auth")
setAuthStatus("Starting authentication...")
await AuthService.getInstance(controller).createAuthRequest()
} catch (error) {
setErrorMessage(error instanceof Error ? error.message : String(error))
@@ -331,6 +333,7 @@ export const AuthView: React.FC<AuthViewProps> = ({ controller, onComplete, onEr
const startOcaAuth = useCallback(() => {
setStep("oca_auth")
setAuthStatus("Starting authentication...")
initiateOcaAuth()
}, [initiateOcaAuth])
@@ -361,8 +364,7 @@ export const AuthView: React.FC<AuthViewProps> = ({ controller, onComplete, onEr
(value: string) => {
setSelectedProvider(value)
if (value === "oca") {
// Show employee check screen before starting auth
setStep("oca_employee_check")
startOcaAuth()
} else if (value === "openai-codex") {
setStep("openai_codex_auth")
startOpenAiCodexAuth()
@@ -538,11 +540,8 @@ export const AuthView: React.FC<AuthViewProps> = ({ controller, onComplete, onEr
setBaseUrl("")
setStep("modelid")
break
case "oca_employee_check":
setStep("provider")
break
case "oca_auth":
setStep("oca_employee_check")
setStep("provider")
break
case "cline_auth":
setStep("menu")
@@ -646,7 +645,7 @@ export const AuthView: React.FC<AuthViewProps> = ({ controller, onComplete, onEr
<Box flexDirection="column">
<Text color="white">Model ID</Text>
<Text> </Text>
<Text color="gray">e.g., claude-sonnet-4-6, gpt-4o</Text>
<Text color="gray">e.g., claude-sonnet-4-20250514, gpt-4o</Text>
<Text> </Text>
<TextInput onChange={setModelId} onSubmit={handleModelIdSubmit} placeholder="model-id" value={modelId} />
<Text> </Text>
@@ -682,9 +681,6 @@ export const AuthView: React.FC<AuthViewProps> = ({ controller, onComplete, onEr
</Box>
)
case "oca_employee_check":
return <OcaEmployeeCheck isActive={step === "oca_employee_check"} onCancel={goBack} onSignIn={startOcaAuth} />
case "oca_auth":
case "cline_auth":
return (
@@ -770,7 +766,6 @@ export const AuthView: React.FC<AuthViewProps> = ({ controller, onComplete, onEr
const [menuIndex, setMenuIndex] = useState(0)
// Steps that allow going back with escape (apikey handled by ApiKeyInput component)
// OcaEmployeeCheck handles its own escape key, so oca_employee_check is not in this list
const canGoBack = [
"provider",
"modelid",
@@ -869,7 +864,7 @@ export const AuthView: React.FC<AuthViewProps> = ({ controller, onComplete, onEr
{index === menuIndex ? " " : " "}
{item.label}
</Text>
{item.value === "cline_auth" && <Text color="yellow"> (try Opus 4.6!)</Text>}
{item.value === "cline_auth" && <Text color="yellow"> (try Kimi K2.5 free!)</Text>}
</Text>
</Box>
))}
+9 -19
View File
@@ -114,11 +114,8 @@ export const BedrockSetup: React.FC<BedrockSetupProps> = ({ isActive, onComplete
// Filtered regions
const filteredRegions = useMemo(() => {
const search = regionSearch.toLowerCase().trim()
if (!search) {
return AWS_REGIONS
}
return AWS_REGIONS.filter((r) => r.toLowerCase().includes(search))
const search = regionSearch.toLowerCase()
return search ? AWS_REGIONS.filter((r) => r.includes(search)) : AWS_REGIONS
}, [regionSearch])
const {
@@ -173,18 +170,10 @@ export const BedrockSetup: React.FC<BedrockSetupProps> = ({ isActive, onComplete
}
}, [step, authMethod, onCancel])
const getSelectedRegion = useCallback(() => {
if (filteredRegions.length > 0 && regionIndex >= 0 && regionIndex < filteredRegions.length) {
return filteredRegions[regionIndex]
}
// If no matches, use the search term as custom region
return regionSearch.trim() || "us-east-1"
}, [filteredRegions, regionIndex, regionSearch])
const finish = useCallback(() => {
const config: BedrockConfig = {
awsAuthentication: authMethod === "default" ? "credentials" : authMethod,
awsRegion: getSelectedRegion(),
awsRegion: filteredRegions[regionIndex] || "us-east-1",
awsUseCrossRegionInference: crossRegion,
}
if (authMethod === "profile") {
@@ -195,7 +184,7 @@ export const BedrockSetup: React.FC<BedrockSetupProps> = ({ isActive, onComplete
if (sessionToken) config.awsSessionToken = sessionToken
}
onComplete(config)
}, [authMethod, profileName, accessKey, secretKey, sessionToken, getSelectedRegion, crossRegion, onComplete])
}, [authMethod, profileName, accessKey, secretKey, sessionToken, filteredRegions, regionIndex, crossRegion, onComplete])
// Handle input for auth_method, region, and options steps
useInput(
@@ -215,11 +204,11 @@ export const BedrockSetup: React.FC<BedrockSetupProps> = ({ isActive, onComplete
} else if (step === "region") {
if (key.escape) {
goBack()
} else if (key.upArrow && filteredRegions.length > 0) {
} else if (key.upArrow) {
setRegionIndex((prev) => (prev > 0 ? prev - 1 : filteredRegions.length - 1))
} else if (key.downArrow && filteredRegions.length > 0) {
} else if (key.downArrow) {
setRegionIndex((prev) => (prev < filteredRegions.length - 1 ? prev + 1 : 0))
} else if (key.return && (filteredRegions.length > 0 || regionSearch.trim())) {
} else if (key.return && filteredRegions.length > 0) {
setStep("options")
} else if (key.backspace || key.delete) {
setRegionSearch((prev) => prev.slice(0, -1))
@@ -341,7 +330,7 @@ export const BedrockSetup: React.FC<BedrockSetupProps> = ({ isActive, onComplete
<Text color="white">AWS Region</Text>
<Text> </Text>
<Box>
<Text color="gray">Search or enter custom region: </Text>
<Text color="gray">Search: </Text>
<Text color="white">{regionSearch}</Text>
<Text inverse> </Text>
</Box>
@@ -361,6 +350,7 @@ export const BedrockSetup: React.FC<BedrockSetupProps> = ({ isActive, onComplete
{showRegionBottom && (
<Text color="gray">... {filteredRegions.length - regionVisibleStart - regionVisibleCount} more below</Text>
)}
{filteredRegions.length === 0 && <Text color="gray">No regions match "{regionSearch}"</Text>}
<Text> </Text>
<Text color="gray">Type to search, arrows to navigate, Enter to select, Esc to go back</Text>
</Box>
-106
View File
@@ -1,106 +0,0 @@
import type { ClineMessage } from "@shared/ExtensionMessage"
import { render } from "ink-testing-library"
import React from "react"
import { describe, expect, it, vi } from "vitest"
import { ChatMessage } from "./ChatMessage"
vi.mock("../hooks/useTerminalSize", () => ({
useTerminalSize: () => ({
columns: 120,
rows: 40,
resizeKey: 0,
}),
}))
describe("ChatMessage subagent rendering", () => {
it("renders subagent approval prompts as a tree", () => {
const message: ClineMessage = {
ts: Date.now(),
type: "ask",
ask: "use_subagents",
text: JSON.stringify({
prompts: [
"Find codebase stats and size",
"Find funny comments and easter eggs",
"Find unusual patterns and history",
],
}),
}
const { lastFrame } = render(React.createElement(ChatMessage, { message, mode: "act" }))
const frame = lastFrame() || ""
expect(frame).toContain("Cline wants to run subagents")
expect(frame).toContain("├─ Find codebase stats and size")
expect(frame).toContain("├─ Find funny comments and easter eggs")
expect(frame).toContain("└─ Find unusual patterns and history")
})
it("renders subagent progress rows with compact token stats and completion checks", () => {
const message: ClineMessage = {
ts: Date.now(),
type: "say",
say: "subagent",
text: JSON.stringify({
status: "running",
total: 3,
completed: 1,
successes: 1,
failures: 0,
toolCalls: 21,
inputTokens: 0,
outputTokens: 0,
contextWindow: 0,
maxContextTokens: 0,
maxContextUsagePercentage: 0,
items: [
{
index: 1,
prompt: "Find codebase stats and size",
status: "completed",
toolCalls: 5,
inputTokens: 0,
outputTokens: 0,
totalCost: 0.034,
contextTokens: 24400,
contextWindow: 200000,
contextUsagePercentage: 12.2,
},
{
index: 2,
prompt: "Find funny comments and easter eggs",
status: "running",
toolCalls: 11,
inputTokens: 0,
outputTokens: 0,
totalCost: 0.056,
contextTokens: 31600,
contextWindow: 200000,
contextUsagePercentage: 15.8,
},
{
index: 3,
prompt: "Find unusual patterns and history",
status: "pending",
toolCalls: 5,
inputTokens: 0,
outputTokens: 0,
totalCost: 0,
contextTokens: 28900,
contextWindow: 200000,
contextUsagePercentage: 14.4,
},
],
}),
}
const { lastFrame } = render(React.createElement(ChatMessage, { isStreaming: true, message, mode: "act" }))
const frame = lastFrame() || ""
expect(frame).toContain("Cline is running subagents")
expect(frame).toContain("✓ Find codebase stats and size")
expect(frame).toContain("5 tool uses · 24.4k tokens · $0.03")
expect(frame).toContain("11 tool uses · 31.6k tokens · $0.06")
expect(frame).toContain("5 tool uses · 28.9k tokens · $0.00")
})
})
+7 -16
View File
@@ -17,7 +17,6 @@ import { useTerminalSize } from "../hooks/useTerminalSize"
import { jsonParseSafe } from "../utils/parser"
import { getToolDescription, isFileEditTool, parseToolFromMessage } from "../utils/tools"
import { DiffView } from "./DiffView"
import { SubagentMessage } from "./SubagentMessage"
/**
* Add "(Tab)" hint after "Act mode" mentions.
@@ -25,7 +24,7 @@ import { SubagentMessage } from "./SubagentMessage"
* Matches just "Act mode" without requiring "to " prefix because markdown
* processing may split "toggle to **Act mode**" into separate text chunks.
*/
function addActModeHint(text: string, keyPrefix: string): React.ReactNode[] {
function addActModeHint(text: string): React.ReactNode[] {
// Match "Act mode" in various capitalizations, but not if already followed by (Tab)
const actModeRegex = /\bact\s+mode\b(?!\s*\(tab\))/gi
const parts = text.split(actModeRegex)
@@ -42,7 +41,7 @@ function addActModeHint(text: string, keyPrefix: string): React.ReactNode[] {
}
if (matches[i]) {
nodes.push(
<React.Fragment key={`${keyPrefix}-act-mode-${i}`}>
<React.Fragment key={`act-mode-${i}`}>
{matches[i]}
<Text color="gray"> (Tab)</Text>
</React.Fragment>,
@@ -60,8 +59,6 @@ function addActModeHint(text: string, keyPrefix: string): React.ReactNode[] {
*/
function renderInlineMarkdown(text: string): React.ReactNode[] {
const nodes: React.ReactNode[] = []
let hintCallIndex = 0
const addHintedText = (value: string) => addActModeHint(value, `hint-${hintCallIndex++}`)
// Match **bold**, *italic*, or `code` - order matters (** before *)
const regex = /(\*\*[^*]+\*\*|\*[^*]+\*|`[^`]+`)/g
let lastIndex = 0
@@ -71,7 +68,7 @@ function renderInlineMarkdown(text: string): React.ReactNode[] {
// Add text before match (with Act Mode hint processing)
if (match.index > lastIndex) {
const beforeText = text.slice(lastIndex, match.index)
nodes.push(...addHintedText(beforeText))
nodes.push(...addActModeHint(beforeText))
}
const fullMatch = match[0]
@@ -80,7 +77,7 @@ function renderInlineMarkdown(text: string): React.ReactNode[] {
if (fullMatch.startsWith("**") && fullMatch.endsWith("**")) {
// Bold - also process for Act Mode hints inside bold text
const boldContent = fullMatch.slice(2, -2)
const hintedContent = addHintedText(boldContent)
const hintedContent = addActModeHint(boldContent)
nodes.push(
<Text bold key={key}>
{hintedContent}
@@ -103,10 +100,10 @@ function renderInlineMarkdown(text: string): React.ReactNode[] {
// Add remaining text (with Act Mode hint processing)
if (lastIndex < text.length) {
nodes.push(...addHintedText(text.slice(lastIndex)))
nodes.push(...addActModeHint(text.slice(lastIndex)))
}
return nodes.length > 0 ? nodes : addHintedText(text)
return nodes.length > 0 ? nodes : addActModeHint(text)
}
/**
@@ -227,7 +224,7 @@ function truncate(text: string, maxLength: number): string {
/**
* Format tool result for display
*/
function formatToolResult(result: string, maxLines = 5): string[] {
function formatToolResult(result: string, maxLines: number = 5): string[] {
const lines = result.split("\n")
if (lines.length <= maxLines) {
return lines
@@ -449,10 +446,6 @@ export const ChatMessage: React.FC<ChatMessageProps> = ({ message, mode, isStrea
)
}
if ((type === "ask" && ask === "use_subagents") || say === "use_subagents" || say === "subagent") {
return <SubagentMessage isStreaming={isStreaming} message={message} mode={mode} />
}
// MCP response
if (say === "mcp_server_response" && text) {
const lines = formatToolResult(text, 8)
@@ -807,8 +800,6 @@ export const ChatMessageList: React.FC<ChatMessageListProps> = ({ messages, maxM
const displayMessages = messages.filter((m) => {
// Skip api_req_finished, they're just markers
if (m.say === "api_req_finished") return false
// Skip hidden aggregated usage messages
if (m.say === "subagent_usage") return false
// Skip empty text messages
if (m.say === "text" && !m.text?.trim()) return false
// Skip checkpoint messages
+31 -119
View File
@@ -109,11 +109,10 @@ import { getApiMetrics, getLastApiReqTotalTokens } from "@shared/getApiMetrics"
import { EmptyRequest, StringRequest } from "@shared/proto/cline/common"
import type { SlashCommandInfo } from "@shared/proto/cline/slash"
import { CLI_ONLY_COMMANDS } from "@shared/slashCommands"
import { getProviderDefaultModelId, getProviderModelIdKey } from "@shared/storage"
import { getProviderModelIdKey } from "@shared/storage"
import type { Mode } from "@shared/storage/types"
import { execSync } from "child_process"
import { Box, Static, Text, useApp, useInput } from "ink"
// biome-ignore lint/style/useImportType: JSX requires React as a value (jsx: "react" in tsconfig)
import React, { useCallback, useEffect, useMemo, useRef, useState } from "react"
import { getAvailableSlashCommands } from "@/core/controller/slash/getAvailableSlashCommands"
import { showTaskWithId } from "@/core/controller/task/showTaskWithId"
@@ -138,7 +137,6 @@ import {
import { isMouseEscapeSequence } from "../utils/input"
import { jsonParseSafe, parseImagesFromInput } from "../utils/parser"
import { extractSlashQuery, filterCommands, insertSlashCommand, sortCommandsWorkflowsFirst } from "../utils/slash-commands"
import { waitFor } from "../utils/timeout"
import { isFileEditTool, parseToolFromMessage } from "../utils/tools"
import { shutdownEvent } from "../vscode-shim"
import { ActionButtons, type ButtonActionType, getButtonConfig, getVisibleButtons } from "./ActionButtons"
@@ -153,24 +151,6 @@ import { SettingsPanelContent } from "./SettingsPanelContent"
import { SlashCommandMenu } from "./SlashCommandMenu"
import { ThinkingIndicator } from "./ThinkingIndicator"
/**
* Persistent input storage that survives React remounts (e.g., during terminal resize).
* Keyed by a stable identifier so each task/session maintains its own input state.
*/
interface PersistedInputState {
text: string
cursorPos: number
pastedTexts: Map<number, string>
pasteCounter: number
}
const inputStateStorage = new Map<string, PersistedInputState>()
function getInputStorageKey(controller: any, taskId?: string): string {
// Use taskId if available, otherwise fall back to controller instance
return taskId || (controller?.task?.taskId ?? "default")
}
interface ChatViewProps {
controller?: any
onExit?: () => void
@@ -229,9 +209,9 @@ function getGitDiffStats(cwd?: string): GitDiffStats | null {
const delMatch = output.match(/(\d+) deletion/)
return {
files: filesMatch ? Number.parseInt(filesMatch[1], 10) : 0,
additions: addMatch ? Number.parseInt(addMatch[1], 10) : 0,
deletions: delMatch ? Number.parseInt(delMatch[1], 10) : 0,
files: filesMatch ? parseInt(filesMatch[1], 10) : 0,
additions: addMatch ? parseInt(addMatch[1], 10) : 0,
deletions: delMatch ? parseInt(delMatch[1], 10) : 0,
}
} catch {
return null
@@ -242,7 +222,7 @@ function getGitDiffStats(cwd?: string): GitDiffStats | null {
* Create a progress bar for context window usage
* Returns { filled, empty } strings to allow different coloring
*/
function createContextBar(used: number, total: number, width = 8): { filled: string; empty: string } {
function createContextBar(used: number, total: number, width: number = 8): { filled: string; empty: string } {
const ratio = Math.min(used / total, 1)
// Use ceil so any usage > 0 shows at least one bar
const filledCount = used > 0 ? Math.max(1, Math.ceil(ratio * width)) : 0
@@ -332,7 +312,7 @@ function parseAskOptions(text: string): string[] {
*/
function expandPastedTexts(text: string, pastedTexts: Map<number, string>): string {
return text.replace(/\[Pasted text #(\d+) \+\d+ lines\]/g, (match, num) => {
const content = pastedTexts.get(Number.parseInt(num, 10))
const content = pastedTexts.get(parseInt(num, 10))
return content ?? match
})
}
@@ -369,14 +349,9 @@ export const ChatView: React.FC<ChatViewProps> = ({
insertText: insertTextAtCursor,
} = useTextInput()
// Get storage key for persisting input across remounts
const storageKey = useMemo(() => getInputStorageKey(ctrl, taskId), [ctrl, taskId])
// Refs for text input and cursor position (used by useHomeEndKeys and to avoid stale closures in useInput)
// Ref for text input (used by useHomeEndKeys)
const textInputRef = useRef(textInput)
textInputRef.current = textInput
const cursorPosRef = useRef(cursorPos)
cursorPosRef.current = cursorPos
const [fileResults, setFileResults] = useState<FileSearchResult[]>([])
const [selectedIndex, setSelectedIndex] = useState(0) // For file menu
@@ -388,10 +363,8 @@ export const ChatView: React.FC<ChatViewProps> = ({
const [userScrolled, setUserScrolled] = useState(false)
// Pasted text storage - maps placeholder number to full pasted content
const [pastedTexts, setPastedTexts] = useState<Map<number, string>>(() => {
return inputStateStorage.get(storageKey)?.pastedTexts ?? new Map()
})
const pasteCounterRef = useRef<number>(inputStateStorage.get(storageKey)?.pasteCounter ?? 0)
const [pastedTexts, setPastedTexts] = useState<Map<number, string>>(new Map())
const pasteCounterRef = useRef(0)
// Track paste timing to combine chunks that arrive in rapid succession
const lastPasteTimeRef = useRef<number>(0)
const activePasteNumRef = useRef<number>(0)
@@ -425,29 +398,6 @@ export const ChatView: React.FC<ChatViewProps> = ({
// Track when we're exiting to hide UI elements before exit
const [isExiting, setIsExiting] = useState(false)
// Restore input state from storage on mount (after resize remount)
useEffect(() => {
const stored = inputStateStorage.get(storageKey)
if (stored) {
setTextInput(stored.text)
setCursorPos(stored.cursorPos)
setPastedTexts(stored.pastedTexts)
pasteCounterRef.current = stored.pasteCounter
}
}, [storageKey, setTextInput, setCursorPos])
// Persist input state to storage whenever it changes (survives remount)
useEffect(() => {
if (textInput || pastedTexts.size > 0) {
inputStateStorage.set(storageKey, {
text: textInput,
cursorPos,
pastedTexts: new Map(pastedTexts),
pasteCounter: pasteCounterRef.current,
})
}
}, [storageKey, textInput, cursorPos, pastedTexts])
// Task switch handling: when switching tasks via /history, we clear the terminal and
// increment a counter used as the root Box's key. This forces React to remount the tree,
// giving us a fresh Static instance. Mirrors how App.tsx handles resize with resizeKey.
@@ -502,12 +452,11 @@ export const ChatView: React.FC<ChatViewProps> = ({
// Get model ID based on current mode and provider
// Different providers use different state keys (e.g., cline uses actModeOpenRouterModelId)
// Re-read when activePanel changes (settings panel closes) to pick up changes
// Falls back to provider's default model if no model has been explicitly set
const modelId = useMemo(() => {
if (!provider) return ""
const stateManager = StateManager.get()
const modelKey = getProviderModelIdKey(provider as ApiProvider, mode)
return (stateManager.getGlobalSettingsKey(modelKey) as string) || getProviderDefaultModelId(provider as ApiProvider) || ""
return (stateManager.getGlobalSettingsKey(modelKey) as string) || ""
}, [mode, provider, activePanel])
const toggleMode = useCallback(async () => {
@@ -540,14 +489,12 @@ export const ChatView: React.FC<ChatViewProps> = ({
clearState() // Force clear React state (bypasses empty messages check)
setTextInput("")
setCursorPos(0)
// Clear persisted state
inputStateStorage.delete(storageKey)
// Post the now-empty state
if (ctrl) {
ctrl.postStateToWebview()
}
}, [ctrl, clearState, storageKey])
}, [ctrl, clearState])
const refs = useRef({
searchTimeout: null as NodeJS.Timeout | null,
@@ -808,8 +755,6 @@ export const ChatView: React.FC<ChatViewProps> = ({
setCursorPos(0)
setPastedTexts(new Map()) // Clear stored pastes
pasteCounterRef.current = 0
// Clear persisted state
inputStateStorage.delete(storageKey)
try {
await ctrl.task.handleWebviewAskResponse(responseType, expandedText)
@@ -817,7 +762,7 @@ export const ChatView: React.FC<ChatViewProps> = ({
// Controller may be disposed
}
},
[ctrl, pendingAsk, pastedTexts, storageKey],
[ctrl, pendingAsk, pastedTexts],
)
// Handle cancel/interrupt
@@ -908,8 +853,6 @@ export const ChatView: React.FC<ChatViewProps> = ({
setCursorPos(0)
setPastedTexts(new Map()) // Clear stored pastes
pasteCounterRef.current = 0
// Clear persisted state
inputStateStorage.delete(storageKey)
try {
// Convert image paths to data URLs if needed
@@ -937,12 +880,10 @@ export const ChatView: React.FC<ChatViewProps> = ({
onError?.()
}
},
[ctrl, onError, pastedTexts, storageKey],
[ctrl, onError, pastedTexts],
)
// Auto-submit initial prompt if provided
// When taskId is also provided, this sends the prompt to resume the existing task
// When no taskId, this creates a new task with the prompt
useEffect(() => {
const autoSubmit = async () => {
if (!initialPrompt && (!initialImages || initialImages.length === 0)) {
@@ -963,32 +904,8 @@ export const ChatView: React.FC<ChatViewProps> = ({
if (initialPrompt) {
setTerminalTitle(initialPrompt)
}
if (taskId) {
// Resuming an existing task with a prompt - wait for task to load first
// The task loading happens in the other useEffect via showTaskWithId
// We need to wait for it to complete before sending the resume message
const task = await waitFor(() => ctrl.task, 5000)
if (task) {
// Send the prompt as a message to resume the task
await task.handleWebviewAskResponse("messageResponse", initialPrompt || "")
} else {
// Task failed to load, fall back to creating new task
Logger.error(`Failed to load task ${taskId} for resume, creating new task instead`)
await ctrl.initTask(
initialPrompt || "",
initialImages && initialImages.length > 0 ? initialImages : undefined,
)
}
} else {
// New task - use initTask
// initialImages are already data URLs from index.ts processing
await ctrl.initTask(
initialPrompt || "",
initialImages && initialImages.length > 0 ? initialImages : undefined,
)
}
// initialImages are already data URLs from index.ts processing
await ctrl.initTask(initialPrompt || "", initialImages && initialImages.length > 0 ? initialImages : undefined)
} catch (_error) {
onError?.()
}
@@ -1084,11 +1001,11 @@ export const ChatView: React.FC<ChatViewProps> = ({
// 3. Handle Option+arrow via key.meta (backup - Ink sometimes parses these instead of passing raw sequence)
if (key.meta) {
if (key.leftArrow) {
setCursorPos(findWordStart(textInputRef.current, cursorPosRef.current))
setCursorPos(findWordStart(textInput, cursorPos))
return
}
if (key.rightArrow) {
setCursorPos(findWordEnd(textInputRef.current, cursorPosRef.current))
setCursorPos(findWordEnd(textInput, cursorPos))
return
}
}
@@ -1274,8 +1191,7 @@ export const ChatView: React.FC<ChatViewProps> = ({
if (hasPrimary && buttonConfig.primaryAction) {
handleButtonAction(buttonConfig.primaryAction, true)
return
}
if (hasSecondary && !hasPrimary && buttonConfig.secondaryAction) {
} else if (hasSecondary && !hasPrimary && buttonConfig.secondaryAction) {
handleButtonAction(buttonConfig.secondaryAction, false)
return
}
@@ -1296,7 +1212,7 @@ export const ChatView: React.FC<ChatViewProps> = ({
}
// Number selection for options (only when no text typed yet)
if (askType === "options") {
const num = Number.parseInt(input, 10)
const num = parseInt(input, 10)
if (textInput === "" && !Number.isNaN(num) && num >= 1 && num <= askOptions.length) {
const selectedOption = askOptions[num - 1]
sendAskResponse("messageResponse", selectedOption)
@@ -1338,10 +1254,10 @@ export const ChatView: React.FC<ChatViewProps> = ({
}
pasteUpdateTimeoutRef.current = setTimeout(() => {
const newPlaceholder = `[Pasted text #${pasteNum} +${activePasteLinesRef.current} lines]`
const pattern = new RegExp(`\\[Pasted text #${pasteNum} \\+\\d+ lines\\]`)
const newText = textInputRef.current.replace(pattern, newPlaceholder)
textInputRef.current = newText // Update ref immediately so setCursorPos bounds check works
setTextInput(newText)
setTextInput((prev) => {
const pattern = new RegExp(`\\[Pasted text #${pasteNum} \\+\\d+ lines\\]`)
return prev.replace(pattern, newPlaceholder)
})
// Update cursor to be right after the placeholder
setCursorPos(activePasteStartPosRef.current + newPlaceholder.length)
Logger.info(`Paste #${pasteNum} complete: ${activePasteLinesRef.current} lines`)
@@ -1354,8 +1270,7 @@ export const ChatView: React.FC<ChatViewProps> = ({
pasteCounterRef.current += 1
const pasteNum = pasteCounterRef.current
activePasteNumRef.current = pasteNum
const currentCursorPos = cursorPosRef.current // Use ref to avoid stale closure
activePasteStartPosRef.current = currentCursorPos // Track where placeholder starts
activePasteStartPosRef.current = cursorPos // Track where placeholder starts
// Count line breaks in the pasted content (handle both \n and \r)
const extraLines = input.match(/[\r\n]/g)?.length || 0
activePasteLinesRef.current = extraLines // Track total lines
@@ -1367,11 +1282,8 @@ export const ChatView: React.FC<ChatViewProps> = ({
return next
})
const newText =
textInputRef.current.slice(0, currentCursorPos) + placeholder + textInputRef.current.slice(currentCursorPos)
textInputRef.current = newText // Update ref immediately so setCursorPos bounds check works
setTextInput(newText)
setCursorPos(currentCursorPos + placeholder.length)
setTextInput((prev) => prev.slice(0, cursorPos) + placeholder + prev.slice(cursorPos))
setCursorPos(cursorPos + placeholder.length)
return // Exit early - don't also add the raw input via normal handling below
}
@@ -1400,15 +1312,15 @@ export const ChatView: React.FC<ChatViewProps> = ({
return
}
if (key.rightArrow && !inSlashMenu && !inFileMenu) {
setCursorPos((pos) => Math.min(textInputRef.current.length, pos + 1))
setCursorPos((pos) => Math.min(textInput.length, pos + 1))
return
}
if (key.upArrow && !inSlashMenu && !inFileMenu) {
setCursorPos(moveCursorUp(textInputRef.current, cursorPosRef.current))
setCursorPos(moveCursorUp(textInput, cursorPos))
return
}
if (key.downArrow && !inSlashMenu && !inFileMenu) {
setCursorPos(moveCursorDown(textInputRef.current, cursorPosRef.current))
setCursorPos(moveCursorDown(textInput, cursorPos))
return
}
// Normal input (single char or short paste)
@@ -1474,10 +1386,10 @@ export const ChatView: React.FC<ChatViewProps> = ({
{/* Dynamic region - only current streaming message + input */}
<Box flexDirection="column" width="100%">
{/* Animated robot and welcome text - only shown before messages start and user hasn't interacted */}
{/* Animated robot and welcome text - only shown before messages start and user hasn't scrolled */}
{isWelcomeState && (
<Box flexDirection="column" marginBottom={1}>
<AsciiMotionCli onInteraction={() => setUserScrolled(true)} />
<AsciiMotionCli onScroll={() => setUserScrolled(true)} />
<Text> </Text>
<Text bold color="white">
{centerText("What can I do for you?")}
+18 -124
View File
@@ -13,7 +13,6 @@ import {
import { Box, Text, useApp, useInput } from "ink"
import React, { useMemo, useState } from "react"
import { useStdinContext } from "../context/StdinContext"
import { fuzzyFilter } from "../utils/fuzzy-search"
import {
BooleanSelect,
buildConfigEntries,
@@ -22,8 +21,6 @@ import {
HookInfo,
HookRow,
MAX_VISIBLE,
ObjectEditorPanel,
ObjectEditorState,
parseValue,
SEPARATOR,
SectionHeader,
@@ -108,8 +105,6 @@ export const ConfigView: React.FC<ConfigViewProps> = ({
const [isEditing, setIsEditing] = useState(false)
const [selectedIndex, setSelectedIndex] = useState(0)
const [editValue, setEditValue] = useState("")
const [searchQuery, setSearchQuery] = useState("")
const [objectEditor, setObjectEditor] = useState<ObjectEditorState | null>(null)
// Build entries for settings tab
const configEntries = useMemo(
@@ -117,13 +112,6 @@ export const ConfigView: React.FC<ConfigViewProps> = ({
[globalState, workspaceState],
)
const filteredConfigEntries = useMemo(() => {
if (!searchQuery.trim()) {
return configEntries
}
return fuzzyFilter(configEntries, searchQuery, (entry) => `${entry.key} ${String(entry.value ?? "")}`)
}, [configEntries, searchQuery])
// Build entries for rules tab
const ruleEntries = useMemo(() => {
const entries: ToggleEntry[] = []
@@ -171,7 +159,7 @@ export const ConfigView: React.FC<ConfigViewProps> = ({
const currentListLength = useMemo(() => {
switch (currentTab) {
case "settings":
return filteredConfigEntries.length
return configEntries.length
case "rules":
return ruleEntries.length
case "workflows":
@@ -183,14 +171,7 @@ export const ConfigView: React.FC<ConfigViewProps> = ({
default:
return 0
}
}, [
currentTab,
filteredConfigEntries.length,
ruleEntries.length,
workflowEntries.length,
hookEntries.length,
skillEntries.length,
])
}, [currentTab, configEntries.length, ruleEntries.length, workflowEntries.length, hookEntries.length, skillEntries.length])
// Get available tabs
const availableTabs = useMemo(() => {
@@ -210,11 +191,10 @@ export const ConfigView: React.FC<ConfigViewProps> = ({
setCurrentTab(newTab)
setSelectedIndex(0)
setIsEditing(false)
setObjectEditor(null)
}
// Settings tab handlers
const selectedConfigEntry = filteredConfigEntries[selectedIndex]
const selectedConfigEntry = configEntries[selectedIndex]
const handleSettingsSave = (value: string | boolean) => {
if (!selectedConfigEntry) {
@@ -230,43 +210,6 @@ export const ConfigView: React.FC<ConfigViewProps> = ({
setIsEditing(false)
}
const getObjectAtPath = (root: Record<string, unknown>, path: string[]): Record<string, unknown> => {
let current: unknown = root
for (const segment of path) {
if (!current || typeof current !== "object") {
return {}
}
current = (current as Record<string, unknown>)[segment]
}
return current && typeof current === "object" ? (current as Record<string, unknown>) : {}
}
const setObjectValueAtPath = (
root: Record<string, unknown>,
path: string[],
key: string,
value: unknown,
): Record<string, unknown> => {
if (path.length === 0) {
return { ...root, [key]: value }
}
const [head, ...rest] = path
const child = root[head]
const childObj = child && typeof child === "object" ? (child as Record<string, unknown>) : {}
return {
...root,
[head]: setObjectValueAtPath(childObj, rest, key, value),
}
}
const persistObjectEditor = (nextObject: Record<string, unknown>, source: "global" | "workspace", key: string) => {
if (source === "global" && onUpdateGlobal) {
onUpdateGlobal(key as GlobalStateAndSettingsKey, nextObject as never)
} else if (source === "workspace" && onUpdateWorkspace) {
onUpdateWorkspace(key as LocalStateKey, nextObject as never)
}
}
const handleSettingsReset = () => {
if (!selectedConfigEntry?.isEditable || selectedConfigEntry.source !== "global") {
return
@@ -297,22 +240,15 @@ export const ConfigView: React.FC<ConfigViewProps> = ({
// Input handling
useInput(
(input, key) => {
if (objectEditor) {
return
}
if (key.escape) {
if (input.toLowerCase() === "q" || key.escape) {
exit()
}
if (key.leftArrow || key.rightArrow || (input >= "1" && input <= "5")) {
const currentTabIndex = availableTabs.findIndex((t) => t.key === currentTab)
const targetIdx =
input >= "1" && input <= "5"
? Number.parseInt(input) - 1
: key.leftArrow
? (currentTabIndex - 1 + availableTabs.length) % availableTabs.length
: (currentTabIndex + 1) % availableTabs.length
// Tab navigation with Tab key or number keys
if (key.tab || (input >= "1" && input <= "5")) {
const targetIdx = key.tab
? (availableTabs.findIndex((t) => t.key === currentTab) + 1) % availableTabs.length
: parseInt(input) - 1
if (targetIdx >= 0 && targetIdx < availableTabs.length) {
handleTabChange(availableTabs[targetIdx].key)
}
@@ -320,45 +256,21 @@ export const ConfigView: React.FC<ConfigViewProps> = ({
}
// List navigation (arrow keys and vim-style j/k)
if (key.upArrow) {
if (key.upArrow || input === "k") {
setSelectedIndex((i) => (i > 0 ? i - 1 : currentListLength - 1))
} else if (key.downArrow) {
} else if (key.downArrow || input === "j") {
setSelectedIndex((i) => (i < currentListLength - 1 ? i + 1 : 0))
}
// Tab-specific actions
if (currentTab === "settings") {
if ((key.return || key.tab) && selectedConfigEntry?.isEditable) {
if (selectedConfigEntry.type === "boolean") {
handleSettingsSave(!selectedConfigEntry.value)
return
}
if (selectedConfigEntry.type === "object") {
const value =
selectedConfigEntry.value && typeof selectedConfigEntry.value === "object"
? (selectedConfigEntry.value as Record<string, unknown>)
: {}
setObjectEditor({
source: selectedConfigEntry.source,
key: selectedConfigEntry.key,
path: [],
value,
selectedIndex: 0,
isEditingValue: false,
editValue: "",
})
return
}
if ((key.return || input === "e") && selectedConfigEntry?.isEditable) {
setEditValue(selectedConfigEntry.value !== undefined ? String(selectedConfigEntry.value) : "")
setIsEditing(true)
} else if (key.ctrl && input.toLowerCase() === "r") {
} else if (input === "r") {
handleSettingsReset()
} else if (key.backspace || key.delete) {
setSearchQuery((prev) => prev.slice(0, -1))
} else if (input && !key.ctrl && !key.meta && !key.escape && !key.upArrow && !key.downArrow) {
setSearchQuery((prev) => prev + input)
}
} else if (key.return || key.tab || input === " ") {
} else if (key.return || input === " ") {
// Toggle for rules/workflows/hooks/skills
handleToggle()
}
@@ -426,31 +338,13 @@ export const ConfigView: React.FC<ConfigViewProps> = ({
)
}
if (objectEditor && currentTab === "settings") {
return (
<ObjectEditorPanel
getObjectAtPath={getObjectAtPath}
onClose={() => setObjectEditor(null)}
onPersist={(nextObject) => persistObjectEditor(nextObject, objectEditor.source, objectEditor.key)}
setObjectValueAtPath={setObjectValueAtPath}
setState={setObjectEditor}
state={objectEditor}
/>
)
}
// Render tab content
const renderTabContent = () => {
switch (currentTab) {
case "settings": {
const visibleEntries = filteredConfigEntries.slice(startIndex, startIndex + MAX_VISIBLE)
const visibleEntries = configEntries.slice(startIndex, startIndex + MAX_VISIBLE)
return (
<React.Fragment>
<Box>
<Text>Search: </Text>
<Text color="white">{searchQuery}</Text>
<Text inverse> </Text>
</Box>
<Box>
<Text>Data directory: </Text>
<Text color="blue" underline>
@@ -613,12 +507,12 @@ export const ConfigView: React.FC<ConfigViewProps> = ({
// Help text based on current tab
const getHelpText = () => {
const base = "↑/↓ Navigate • ←/→ tabs • 1-5 tabs • Esc Exit"
const base = "↑/↓/j/k Navigate • Tab/1-5 Switch tabs • q/Esc Exit"
if (currentTab === "settings") {
return `${base} Type to search • Enter/Tab Edit (booleans toggle) • Backspace clear search • Ctrl+R Reset`
return `${base} • Enter/e Edit • r Reset`
}
const openFolder = onOpenFolder ? " • o Open folder" : ""
return `${base} • Enter/Tab/Space Toggle${openFolder}`
return `${base} • Enter/Space Toggle${openFolder}`
}
return (
+11 -180
View File
@@ -46,19 +46,16 @@ export interface SkillInfo {
enabled: boolean
}
export interface ObjectEditorState {
source: "global" | "workspace"
key: string
path: string[]
value: Record<string, unknown>
selectedIndex: number
isEditingValue: boolean
editValue: string
}
export const EXCLUDED_KEYS = new Set([
"taskHistory",
"primaryRootIndex",
"subagentsEnabled",
"subagentTerminalOutputLineLimit",
"welcomeViewCompleted",
"isNewUser",
])
export const EXCLUDED_KEYS = new Set(["taskHistory", "primaryRootIndex", "welcomeViewCompleted", "isNewUser"])
export const EDITABLE_TYPES: Set<ValueType> = new Set(["string", "number", "boolean", "object"])
export const EDITABLE_TYPES: Set<ValueType> = new Set(["string", "number", "boolean"])
export const MAX_VISIBLE = 12
export const SEPARATOR = "─".repeat(80)
@@ -138,7 +135,7 @@ export function parseValue(input: string, type: ValueType): unknown {
return input.toLowerCase() === "true" || input === "1"
}
if (type === "number") {
const num = Number.parseFloat(input)
const num = parseFloat(input)
return Number.isNaN(num) ? 0 : num
}
if (type === "object") {
@@ -220,7 +217,7 @@ export const TextInput: React.FC<TextInputProps> = ({ label, onChange, onCancel,
</Text>
<Box>
<Text color="white">{value}</Text>
<Text color="cyan">|</Text>
<Text inverse> </Text>
</Box>
<Text color="gray">Type: {type} Enter to save Esc to cancel</Text>
</Box>
@@ -382,169 +379,3 @@ export const SectionHeader: React.FC<{ title: string }> = ({ title }) => (
</Text>
</Box>
)
interface ObjectEditorPanelProps {
state: ObjectEditorState
setState: React.Dispatch<React.SetStateAction<ObjectEditorState | null>>
onClose: () => void
onPersist: (nextObject: Record<string, unknown>) => void
getObjectAtPath: (root: Record<string, unknown>, path: string[]) => Record<string, unknown>
setObjectValueAtPath: (root: Record<string, unknown>, path: string[], key: string, value: unknown) => Record<string, unknown>
}
export const ObjectEditorPanel: React.FC<ObjectEditorPanelProps> = ({
state,
setState,
onClose,
onPersist,
getObjectAtPath,
setObjectValueAtPath,
}) => {
const { isRawModeSupported } = useStdinContext()
const currentNode = getObjectAtPath(state.value, state.path)
const objectEntries = Object.entries(currentNode).sort(([a], [b]) => a.localeCompare(b))
const selectedEntry = objectEntries[state.selectedIndex]
const breadcrumb = [state.key, ...state.path].join(" ")
useInput(
(input, key) => {
if (state.isEditingValue) {
if (key.escape) {
setState((prev) => (prev ? { ...prev, isEditingValue: false, editValue: "" } : prev))
return
}
if (key.return) {
if (!selectedEntry) {
setState((prev) => (prev ? { ...prev, isEditingValue: false, editValue: "" } : prev))
return
}
const [entryKey, entryValue] = selectedEntry
let parsed: unknown = state.editValue
if (typeof entryValue === "boolean") {
parsed = state.editValue.toLowerCase() === "true" || state.editValue === "1"
} else if (typeof entryValue === "number") {
const maybeNum = Number(state.editValue)
parsed = Number.isNaN(maybeNum) ? 0 : maybeNum
}
const nextObject = setObjectValueAtPath(state.value, state.path, entryKey, parsed)
onPersist(nextObject)
setState((prev) => (prev ? { ...prev, value: nextObject, isEditingValue: false, editValue: "" } : prev))
return
}
if (key.backspace || key.delete) {
setState((prev) => (prev ? { ...prev, editValue: prev.editValue.slice(0, -1) } : prev))
return
}
if (input && !key.ctrl && !key.meta) {
setState((prev) => (prev ? { ...prev, editValue: prev.editValue + input } : prev))
}
return
}
if (key.escape) {
if (state.path.length > 0) {
setState((prev) => (prev ? { ...prev, path: prev.path.slice(0, -1), selectedIndex: 0 } : prev))
} else {
onClose()
}
return
}
if (key.upArrow || input === "k") {
setState((prev) =>
prev
? {
...prev,
selectedIndex:
objectEntries.length > 0
? prev.selectedIndex > 0
? prev.selectedIndex - 1
: objectEntries.length - 1
: 0,
}
: prev,
)
return
}
if (key.downArrow || input === "j") {
setState((prev) =>
prev
? {
...prev,
selectedIndex:
objectEntries.length > 0
? prev.selectedIndex < objectEntries.length - 1
? prev.selectedIndex + 1
: 0
: 0,
}
: prev,
)
return
}
if (key.return || key.tab) {
if (!selectedEntry) {
return
}
const [entryKey, entryValue] = selectedEntry
if (typeof entryValue === "boolean") {
const nextObject = setObjectValueAtPath(state.value, state.path, entryKey, !entryValue)
onPersist(nextObject)
setState((prev) => (prev ? { ...prev, value: nextObject } : prev))
return
}
if (entryValue && typeof entryValue === "object" && !Array.isArray(entryValue)) {
setState((prev) => (prev ? { ...prev, path: [...prev.path, entryKey], selectedIndex: 0 } : prev))
return
}
setState((prev) =>
prev
? { ...prev, isEditingValue: true, editValue: entryValue !== undefined ? String(entryValue) : "" }
: prev,
)
}
},
{ isActive: isRawModeSupported },
)
return (
<Box flexDirection="column">
<Text bold color="white">
Edit Nested Object
</Text>
<Text color="gray">{SEPARATOR}</Text>
<Text color="cyan">{breadcrumb}</Text>
{state.isEditingValue ? (
<Box flexDirection="column" marginTop={1}>
<Box>
<Text color="white">{state.editValue}</Text>
<Text color="cyan">|</Text>
</Box>
<Text color="gray">Enter to save Esc to cancel</Text>
</Box>
) : (
<Box flexDirection="column" marginTop={1}>
{objectEntries.length === 0 ? (
<Text color="gray">No nested keys at this level.</Text>
) : (
objectEntries.map(([key, value], idx) => {
const isSelected = idx === state.selectedIndex
const valueText =
value && typeof value === "object" && !Array.isArray(value) ? "{...}" : String(value)
return (
<Text color={isSelected ? "cyan" : undefined} key={key}>
{isSelected ? " " : " "}
<Text color="cyan">{key}</Text>
<Text color="gray">: </Text>
<Text color="white">{valueText}</Text>
</Text>
)
})
)}
<Text color="gray">/ Navigate Enter/Tab Edit or drill in Esc Back/Close</Text>
</Box>
)}
</Box>
)
}
+5 -5
View File
@@ -45,15 +45,15 @@ export const FeaturedModelPicker: React.FC<FeaturedModelPickerProps> = ({
<Text bold color={isSelected ? COLORS.primaryBlue : "white"}>
{model.name}
</Text>
{model.labels.map((label) => (
<Text key={label}>
{model.label && (
<Text>
<Text> </Text>
<Text backgroundColor={label === "FREE" ? "gray" : COLORS.primaryBlue} color="black">
<Text backgroundColor={model.label === "FREE" ? "gray" : COLORS.primaryBlue} color="black">
{" "}
{label}{" "}
{model.label}{" "}
</Text>
</Text>
))}
)}
</Box>
<Box paddingLeft={2}>
<Text color="gray">{model.description}</Text>
-24
View File
@@ -43,30 +43,6 @@ export const HelpPanelContent: React.FC<HelpPanelContentProps> = ({ onClose }) =
</Text>
</Box>
<Box flexDirection="column">
<Text bold>Keyboard Shortcuts</Text>
<Text>
{" "}
<Text color="white">Ctrl+U</Text> - Clear entire input (delete to start)
</Text>
<Text>
{" "}
<Text color="white">Ctrl+K</Text> - Delete from cursor to end
</Text>
<Text>
{" "}
<Text color="white">Ctrl+W</Text> - Delete word backwards
</Text>
<Text>
{" "}
<Text color="white">Ctrl+A / Ctrl+E</Text> - Jump to start / end of input
</Text>
<Text>
{" "}
<Text color="white">Alt/Option+/</Text> - Move by word
</Text>
</Box>
<Box flexDirection="column">
<Text bold>Slash Commands</Text>
<Text>
+4 -18
View File
@@ -6,7 +6,6 @@
import { Box, Text } from "ink"
import Spinner from "ink-spinner"
import React, { useEffect, useMemo, useState } from "react"
import { refreshOcaModels } from "@/core/controller/models/refreshOcaModels"
import { refreshOpenRouterModels } from "@/core/controller/models/refreshOpenRouterModels"
import {
type ApiProvider,
@@ -65,7 +64,6 @@ import {
xaiDefaultModelId,
xaiModels,
} from "@/shared/api"
import { StringRequest } from "@/shared/proto/cline/common"
import { filterOpenRouterModelIds } from "@/shared/utils/model-filters"
import { COLORS } from "../constants/colors"
import { getOpenRouterDefaultModelId, usesOpenRouterModels } from "../utils/openrouter-models"
@@ -107,7 +105,7 @@ export function hasStaticModels(provider: string): boolean {
}
export function hasModelPicker(provider: string): boolean {
return hasStaticModels(provider) || usesOpenRouterModels(provider) || provider === "oca"
return hasStaticModels(provider) || usesOpenRouterModels(provider)
}
export function getDefaultModelId(provider: string): string {
@@ -134,7 +132,7 @@ export const ModelPicker: React.FC<ModelPickerProps> = ({ provider, controller,
const [isLoading, setIsLoading] = useState(false)
const [asyncModels, setAsyncModels] = useState<string[]>([])
// Fetch async models (OpenRouter or OCA) when needed
// Fetch OpenRouter models when needed using shared core function
useEffect(() => {
if (usesOpenRouterModels(provider)) {
setIsLoading(true)
@@ -147,23 +145,11 @@ export const ModelPicker: React.FC<ModelPickerProps> = ({ provider, controller,
.finally(() => {
setIsLoading(false)
})
} else if (provider === "oca") {
setIsLoading(true)
refreshOcaModels(controller, StringRequest.create({ value: "" }))
.then((result) => {
if (result.models) {
const modelIds = Object.keys(result.models).sort((a, b) => a.localeCompare(b))
setAsyncModels(modelIds)
}
})
.finally(() => {
setIsLoading(false)
})
}
}, [provider, controller])
const modelList = useMemo(() => {
if (usesOpenRouterModels(provider) || provider === "oca") {
if (usesOpenRouterModels(provider)) {
return asyncModels
}
return getModelList(provider)
@@ -194,7 +180,7 @@ export const ModelPicker: React.FC<ModelPickerProps> = ({ provider, controller,
}
// If async fetch returned no models, render nothing
if ((usesOpenRouterModels(provider) || provider === "oca") && modelList.length === 0) {
if (usesOpenRouterModels(provider) && modelList.length === 0) {
return null
}
-88
View File
@@ -1,88 +0,0 @@
/**
* OCA (Oracle Cloud Assist) employee check component.
* Shows a checkbox for "I'm an Oracle Employee" and a sign-in button.
* Sets ocaMode in state before triggering the OAuth flow.
*/
import { Box, Text, useInput } from "ink"
// biome-ignore lint/style/useImportType: React is used as a value by JSX (jsx: "react" in tsconfig)
import React, { useCallback, useState } from "react"
import { StateManager } from "@/core/storage/StateManager"
import { COLORS } from "../constants/colors"
import { useStdinContext } from "../context/StdinContext"
interface OcaEmployeeCheckProps {
/** Whether this component is active and should handle input */
isActive: boolean
/** Called when user confirms and wants to proceed with sign-in */
onSignIn: () => void
/** Called when user presses Escape to go back */
onCancel: () => void
}
export const OcaEmployeeCheck: React.FC<OcaEmployeeCheckProps> = ({ isActive, onSignIn, onCancel }) => {
const { isRawModeSupported } = useStdinContext()
const [isEmployee, setIsEmployee] = useState(true) // Default to checked (internal), matching extension behavior
const [selectedIndex, setSelectedIndex] = useState(0) // 0 = checkbox, 1 = sign in button
const ITEM_COUNT = 2
const handleSignIn = useCallback(async () => {
// Persist ocaMode to state before starting auth
const stateManager = StateManager.get()
stateManager.setGlobalState("ocaMode", isEmployee ? "internal" : "external")
await stateManager.flushPendingState()
onSignIn()
}, [isEmployee, onSignIn])
useInput(
(_input, key) => {
if (key.escape) {
onCancel()
return
}
if (key.upArrow) {
setSelectedIndex((prev) => (prev > 0 ? prev - 1 : ITEM_COUNT - 1))
} else if (key.downArrow) {
setSelectedIndex((prev) => (prev < ITEM_COUNT - 1 ? prev + 1 : 0))
} else if (key.tab || (key.return && selectedIndex === 0)) {
// Toggle checkbox when Tab is pressed or Enter on checkbox item
if (selectedIndex === 0) {
setIsEmployee((prev) => !prev)
}
} else if (key.return && selectedIndex === 1) {
// Sign in button
handleSignIn()
}
},
{ isActive: isRawModeSupported && isActive },
)
return (
<Box flexDirection="column">
<Text color="white">Oracle Code Assist</Text>
<Text> </Text>
{/* Checkbox: I'm an Oracle Employee */}
<Text>
<Text bold color={selectedIndex === 0 ? COLORS.primaryBlue : undefined}>
{selectedIndex === 0 ? "" : " "}{" "}
</Text>
<Text color={selectedIndex === 0 || isEmployee ? COLORS.primaryBlue : "gray"}>{isEmployee ? "[✓]" : "[ ]"}</Text>
<Text color={selectedIndex === 0 ? COLORS.primaryBlue : "white"}> I'm an Oracle Employee</Text>
{selectedIndex === 0 && <Text color="gray"> (Tab to toggle)</Text>}
</Text>
{/* Sign in button */}
<Text>
<Text bold color={selectedIndex === 1 ? COLORS.primaryBlue : undefined}>
{selectedIndex === 1 ? "" : " "}{" "}
</Text>
<Text color={selectedIndex === 1 ? COLORS.primaryBlue : "white"}>Sign in with Oracle Code Assist</Text>
{selectedIndex === 1 && <Text color="gray"> (Enter)</Text>}
</Text>
<Text> </Text>
<Text color="gray">Please ask your IT administrator to set up Oracle Code Assist as a model provider.</Text>
<Text> </Text>
<Text color="gray">Arrows to navigate, Tab to toggle, Enter to continue, Esc to go back</Text>
</Box>
)
}
+6 -5
View File
@@ -5,11 +5,11 @@
import React, { useMemo } from "react"
import { StateManager } from "@/core/storage/StateManager"
import type { ApiConfiguration } from "@/shared/api"
import { getProviderLabel, useValidProviders } from "../utils/providers"
import { SearchableList, type SearchableListItem } from "./SearchableList"
import { CLI_EXCLUDED_PROVIDERS, getProviderLabel, getProviderOrder } from "../utils/providers"
import { SearchableList, SearchableListItem } from "./SearchableList"
// Re-export for backwards compatibility
export { getProviderLabel }
export { CLI_EXCLUDED_PROVIDERS, getProviderLabel, getProviderOrder }
/**
* Check if a provider is configured (has required credentials/settings)
@@ -125,16 +125,17 @@ interface ProviderPickerProps {
export const ProviderPicker: React.FC<ProviderPickerProps> = ({ onSelect, isActive = true }) => {
// Get API configuration to check which providers are configured
const apiConfig = StateManager.get().getApiConfiguration()
const sorted = useValidProviders()
// Use providers.json order, filtered to exclude CLI-incompatible providers
const items: SearchableListItem[] = useMemo(() => {
const sorted = getProviderOrder().filter((p: string) => !CLI_EXCLUDED_PROVIDERS.has(p))
return sorted.map((providerId: string) => ({
id: providerId,
label: getProviderLabel(providerId),
suffix: isProviderConfigured(providerId, apiConfig) ? "(Configured)" : undefined,
}))
}, [apiConfig, sorted])
}, [apiConfig])
return <SearchableList isActive={isActive} items={items} onSelect={(item) => onSelect(item.id)} />
}
+20 -168
View File
@@ -7,21 +7,17 @@ import type { AutoApprovalSettings } from "@shared/AutoApprovalSettings"
import { DEFAULT_AUTO_APPROVAL_SETTINGS } from "@shared/AutoApprovalSettings"
import type { ApiProvider, ModelInfo } from "@shared/api"
import { getProviderModelIdKey, isSettingsKey, ProviderToApiKeyMap } from "@shared/storage"
import { isOpenaiReasoningEffort, OPENAI_REASONING_EFFORT_OPTIONS, type OpenaiReasoningEffort } from "@shared/storage/types"
import type { TelemetrySetting } from "@shared/TelemetrySetting"
import { Box, Text, useInput } from "ink"
import Spinner from "ink-spinner"
import React, { useCallback, useEffect, useMemo, useState } from "react"
import { buildApiHandler } from "@/core/api"
import type { Controller } from "@/core/controller"
import { refreshOcaModels } from "@/core/controller/models/refreshOcaModels"
import { StateManager } from "@/core/storage/StateManager"
import { openAiCodexOAuthManager } from "@/integrations/openai-codex/oauth"
import { ClineAccountService } from "@/services/account/ClineAccountService"
import { AuthService, ClineAccountOrganization } from "@/services/auth/AuthService"
import { StringRequest } from "@/shared/proto/cline/common"
import { openExternal } from "@/utils/env"
import { supportsReasoningEffortForModel } from "@/utils/model-utils"
import { version as CLI_VERSION } from "../../package.json"
import { COLORS } from "../constants/colors"
import { useStdinContext } from "../context/StdinContext"
@@ -39,7 +35,6 @@ import {
} from "./FeaturedModelPicker"
import { LanguagePicker } from "./LanguagePicker"
import { hasModelPicker, ModelPicker } from "./ModelPicker"
import { OcaEmployeeCheck } from "./OcaEmployeeCheck"
import { OrganizationPicker } from "./OrganizationPicker"
import { Panel, PanelTab } from "./Panel"
import { getProviderLabel, ProviderPicker } from "./ProviderPicker"
@@ -56,25 +51,13 @@ type SettingsTab = "api" | "auto-approve" | "features" | "other" | "account"
interface ListItem {
key: string
label: string
type: "checkbox" | "readonly" | "editable" | "separator" | "header" | "spacer" | "action" | "cycle"
type: "checkbox" | "readonly" | "editable" | "separator" | "header" | "spacer" | "action"
value: string | boolean
description?: string
isSubItem?: boolean
parentKey?: string
}
function normalizeReasoningEffort(value: unknown): OpenaiReasoningEffort {
if (isOpenaiReasoningEffort(value)) {
return value
}
return "low"
}
function nextReasoningEffort(current: OpenaiReasoningEffort): OpenaiReasoningEffort {
const idx = OPENAI_REASONING_EFFORT_OPTIONS.indexOf(current)
return OPENAI_REASONING_EFFORT_OPTIONS[(idx + 1) % OPENAI_REASONING_EFFORT_OPTIONS.length]
}
const TABS: PanelTab[] = [
{ key: "api", label: "API" },
{ key: "auto-approve", label: "Auto-approve" },
@@ -85,12 +68,6 @@ const TABS: PanelTab[] = [
// Settings configuration for simple boolean toggles
const FEATURE_SETTINGS = {
subagents: {
stateKey: "subagentsEnabled",
default: false,
label: "Subagents",
description: "Let Cline run focused subagents in parallel to explore the codebase for you",
},
autoCondense: {
stateKey: "useAutoCondense",
default: false,
@@ -121,12 +98,6 @@ const FEATURE_SETTINGS = {
label: "Parallel tool calling",
description: "Allow multiple tools in a single response",
},
doubleCheckCompletion: {
stateKey: "doubleCheckCompletionEnabled",
default: false,
label: "Double-check completion",
description: "Reject first completion attempt and require re-verification",
},
} as const
type FeatureKey = keyof typeof FEATURE_SETTINGS
@@ -165,7 +136,6 @@ export const SettingsPanelContent: React.FC<SettingsPanelContentProps> = ({
const [isEnteringApiKey, setIsEnteringApiKey] = useState(false)
const [isConfiguringBedrock, setIsConfiguringBedrock] = useState(false)
const [isWaitingForCodexAuth, setIsWaitingForCodexAuth] = useState(false)
const [isShowingOcaEmployeeCheck, setIsShowingOcaEmployeeCheck] = useState(false)
const [codexAuthError, setCodexAuthError] = useState<string | null>(null)
const [pendingProvider, setPendingProvider] = useState<string | null>(null)
const [apiKeyValue, setApiKeyValue] = useState("")
@@ -195,12 +165,6 @@ export const SettingsPanelContent: React.FC<SettingsPanelContentProps> = ({
const [planThinkingEnabled, setPlanThinkingEnabled] = useState<boolean>(
() => (stateManager.getGlobalSettingsKey("planModeThinkingBudgetTokens") ?? 0) > 0,
)
const [actReasoningEffort, setActReasoningEffort] = useState<OpenaiReasoningEffort>(() =>
normalizeReasoningEffort(stateManager.getGlobalSettingsKey("actModeReasoningEffort")),
)
const [planReasoningEffort, setPlanReasoningEffort] = useState<OpenaiReasoningEffort>(() =>
normalizeReasoningEffort(stateManager.getGlobalSettingsKey("planModeReasoningEffort")),
)
// Auto-approve settings (complex nested object)
const [autoApproveSettings, setAutoApproveSettings] = useState<AutoApprovalSettings>(() => {
@@ -239,8 +203,6 @@ export const SettingsPanelContent: React.FC<SettingsPanelContentProps> = ({
// OCA auth hook
const handleOcaAuthSuccess = useCallback(async () => {
await applyProviderConfig({ providerId: "oca", controller })
// Fetch OCA models from the API - this sets actModeOcaModelId/planModeOcaModelId in state
await refreshOcaModels(controller!, StringRequest.create({ value: "" }))
setProvider("oca")
refreshModelIds()
}, [controller, refreshModelIds])
@@ -433,12 +395,9 @@ export const SettingsPanelContent: React.FC<SettingsPanelContentProps> = ({
// Build items list based on current tab
const items: ListItem[] = useMemo(() => {
// Some providers/models expose reasoning effort instead of thinking budget controls.
const providerUsesReasoningEffort = provider === "openai-native" || provider === "openai-codex"
const showActReasoningEffort = supportsReasoningEffortForModel(actModelId || "")
const showPlanReasoningEffort = supportsReasoningEffortForModel(planModelId || "")
const showActThinkingOption = !providerUsesReasoningEffort && !showActReasoningEffort
const showPlanThinkingOption = !providerUsesReasoningEffort && !showPlanReasoningEffort
// OpenAI Native, Codex, and GPT models don't support thinking budget (they use reasoning effort)
const isGptModel = actModelId?.toLowerCase().includes("gpt") || planModelId?.toLowerCase().includes("gpt")
const showThinkingOption = provider !== "openai-native" && provider !== "openai-codex" && !isGptModel
switch (currentTab) {
case "api":
@@ -462,7 +421,7 @@ export const SettingsPanelContent: React.FC<SettingsPanelContentProps> = ({
type: "editable" as const,
value: actModelId || "not set",
},
...(showActThinkingOption
...(showThinkingOption
? [
{
key: "actThinkingEnabled",
@@ -472,16 +431,6 @@ export const SettingsPanelContent: React.FC<SettingsPanelContentProps> = ({
},
]
: []),
...(showActReasoningEffort
? [
{
key: "actReasoningEffort",
label: "Reasoning effort",
type: "cycle" as const,
value: actReasoningEffort,
},
]
: []),
{ key: "planHeader", label: "Plan Mode", type: "header" as const, value: "" },
{
key: "planModelId",
@@ -489,7 +438,7 @@ export const SettingsPanelContent: React.FC<SettingsPanelContentProps> = ({
type: "editable" as const,
value: planModelId || "not set",
},
...(showPlanThinkingOption
...(showThinkingOption
? [
{
key: "planThinkingEnabled",
@@ -499,16 +448,6 @@ export const SettingsPanelContent: React.FC<SettingsPanelContentProps> = ({
},
]
: []),
...(showPlanReasoningEffort
? [
{
key: "planReasoningEffort",
label: "Reasoning effort",
type: "cycle" as const,
value: planReasoningEffort,
},
]
: []),
{ key: "spacer1", label: "", type: "spacer" as const, value: "" },
]
: [
@@ -518,7 +457,7 @@ export const SettingsPanelContent: React.FC<SettingsPanelContentProps> = ({
type: "editable" as const,
value: actModelId || "not set",
},
...(showActThinkingOption
...(showThinkingOption
? [
{
key: "actThinkingEnabled",
@@ -528,16 +467,6 @@ export const SettingsPanelContent: React.FC<SettingsPanelContentProps> = ({
},
]
: []),
...(showActReasoningEffort
? [
{
key: "actReasoningEffort",
label: "Reasoning effort",
type: "cycle" as const,
value: actReasoningEffort,
},
]
: []),
]),
{
key: "separateModels",
@@ -700,8 +629,6 @@ export const SettingsPanelContent: React.FC<SettingsPanelContentProps> = ({
separateModels,
actThinkingEnabled,
planThinkingEnabled,
actReasoningEffort,
planReasoningEffort,
autoApproveSettings,
features,
preferredLanguage,
@@ -735,33 +662,6 @@ export const SettingsPanelContent: React.FC<SettingsPanelContentProps> = ({
}
}, [items.length, selectedIndex])
const rebuildTaskApi = useCallback(() => {
if (!controller?.task) {
return
}
const currentMode = stateManager.getGlobalSettingsKey("mode")
const apiConfig = stateManager.getApiConfiguration()
controller.task.api = buildApiHandler({ ...apiConfig, ulid: controller.task.ulid }, currentMode)
}, [controller, stateManager])
const setReasoningEffortForMode = useCallback(
(mode: "act" | "plan", effort: OpenaiReasoningEffort) => {
if (mode === "act") {
setActReasoningEffort(effort)
stateManager.setGlobalState("actModeReasoningEffort", effort)
if (!separateModels) {
setPlanReasoningEffort(effort)
stateManager.setGlobalState("planModeReasoningEffort", effort)
}
} else {
setPlanReasoningEffort(effort)
stateManager.setGlobalState("planModeReasoningEffort", effort)
}
rebuildTaskApi()
},
[separateModels, rebuildTaskApi, stateManager],
)
// Handle toggle/edit for selected item
const handleAction = useCallback(() => {
const item = items[selectedIndex]
@@ -785,15 +685,6 @@ export const SettingsPanelContent: React.FC<SettingsPanelContentProps> = ({
return
}
if (item.type === "cycle") {
const targetMode = item.key === "actReasoningEffort" ? "act" : item.key === "planReasoningEffort" ? "plan" : undefined
if (targetMode) {
const currentEffort = targetMode === "act" ? actReasoningEffort : planReasoningEffort
setReasoningEffortForMode(targetMode, nextReasoningEffort(currentEffort))
}
return
}
if (item.type === "editable") {
// For provider field, use the provider picker
if (item.key === "provider") {
@@ -851,16 +742,7 @@ export const SettingsPanelContent: React.FC<SettingsPanelContentProps> = ({
const actModel = stateManager.getGlobalSettingsKey(actKey)
if (planKey) stateManager.setGlobalState(planKey, actModel)
}
const actThinkingBudget = stateManager.getGlobalSettingsKey("actModeThinkingBudgetTokens") ?? 0
stateManager.setGlobalState("planModeThinkingBudgetTokens", actThinkingBudget)
setPlanThinkingEnabled(actThinkingBudget > 0)
const actEffort = normalizeReasoningEffort(stateManager.getGlobalSettingsKey("actModeReasoningEffort"))
stateManager.setGlobalState("planModeReasoningEffort", actEffort)
setPlanReasoningEffort(actEffort)
}
rebuildTaskApi()
return
}
@@ -868,19 +750,23 @@ export const SettingsPanelContent: React.FC<SettingsPanelContentProps> = ({
if (item.key === "actThinkingEnabled") {
setActThinkingEnabled(newValue)
stateManager.setGlobalState("actModeThinkingBudgetTokens", newValue ? 1024 : 0)
if (!separateModels) {
setPlanThinkingEnabled(newValue)
stateManager.setGlobalState("planModeThinkingBudgetTokens", newValue ? 1024 : 0)
}
// Rebuild API handler to apply thinking budget change
rebuildTaskApi()
if (controller?.task) {
const currentMode = stateManager.getGlobalSettingsKey("mode")
const apiConfig = stateManager.getApiConfiguration()
controller.task.api = buildApiHandler({ ...apiConfig, ulid: controller.task.ulid }, currentMode)
}
return
}
if (item.key === "planThinkingEnabled") {
setPlanThinkingEnabled(newValue)
stateManager.setGlobalState("planModeThinkingBudgetTokens", newValue ? 1024 : 0)
// Rebuild API handler to apply thinking budget change
rebuildTaskApi()
if (controller?.task) {
const currentMode = stateManager.getGlobalSettingsKey("mode")
const apiConfig = stateManager.getApiConfiguration()
controller.task.api = buildApiHandler({ ...apiConfig, ulid: controller.task.ulid }, currentMode)
}
return
}
@@ -937,11 +823,6 @@ export const SettingsPanelContent: React.FC<SettingsPanelContentProps> = ({
handleClineLogin,
handleClineLogout,
accountOrganizations,
separateModels,
actReasoningEffort,
planReasoningEffort,
rebuildTaskApi,
setReasoningEffortForMode,
])
// Handle model selection from picker
@@ -1084,8 +965,8 @@ export const SettingsPanelContent: React.FC<SettingsPanelContentProps> = ({
setProvider("oca")
refreshModelIds()
} else {
// Not logged in - show employee check before auth
setIsShowingOcaEmployeeCheck(true)
// Not logged in - trigger OAuth
startOcaAuth()
}
return
}
@@ -1376,7 +1257,7 @@ export const SettingsPanelContent: React.FC<SettingsPanelContentProps> = ({
return
}
},
{ isActive: isRawModeSupported && !isEnteringApiKey && !isConfiguringBedrock && !isShowingOcaEmployeeCheck },
{ isActive: isRawModeSupported && !isEnteringApiKey && !isConfiguringBedrock },
)
// Render content
@@ -1552,19 +1433,6 @@ export const SettingsPanelContent: React.FC<SettingsPanelContentProps> = ({
)
}
if (isShowingOcaEmployeeCheck) {
return (
<OcaEmployeeCheck
isActive={isShowingOcaEmployeeCheck}
onCancel={() => setIsShowingOcaEmployeeCheck(false)}
onSignIn={() => {
setIsShowingOcaEmployeeCheck(false)
startOcaAuth()
}}
/>
)
}
if (isWaitingForOcaAuth) {
return (
<Box flexDirection="column">
@@ -1701,21 +1569,6 @@ export const SettingsPanelContent: React.FC<SettingsPanelContentProps> = ({
)
}
if (item.type === "cycle") {
return (
<Text key={item.key}>
<Text bold color={isSelected ? COLORS.primaryBlue : undefined}>
{isSelected ? "" : " "}{" "}
</Text>
<Text color={isSelected ? COLORS.primaryBlue : "white"}>{item.label}: </Text>
<Text color={COLORS.primaryBlue}>
{typeof item.value === "string" ? item.value : String(item.value)}
</Text>
{isSelected && <Text color="gray"> (Tab to cycle)</Text>}
</Text>
)
}
// Readonly or editable field
return (
<Text key={item.key}>
@@ -1746,7 +1599,6 @@ export const SettingsPanelContent: React.FC<SettingsPanelContentProps> = ({
!!codexAuthError ||
isPickingOrganization ||
isWaitingForClineAuth ||
isShowingOcaEmployeeCheck ||
isWaitingForOcaAuth ||
isEditing
-361
View File
@@ -1,361 +0,0 @@
import type { ClineAskUseSubagents, ClineMessage, ClineSaySubagentStatus } from "@shared/ExtensionMessage"
import { Box, Text } from "ink"
import Spinner from "ink-spinner"
import React from "react"
import { COLORS } from "../constants/colors"
import { useTerminalSize } from "../hooks/useTerminalSize"
import { jsonParseSafe } from "../utils/parser"
interface SubagentMessageProps {
message: ClineMessage
isStreaming?: boolean
mode?: "act" | "plan"
}
const TREE_PREFIX_WIDTH = 5
const MIN_PROMPT_WIDTH = 20
const DotRow: React.FC<{ children: React.ReactNode; color?: string; flashing?: boolean }> = ({
children,
color,
flashing = false,
}) => (
<Box flexDirection="row">
<Box width={2}>
{flashing ? (
<Text color={color}>
<Spinner type="toggle8" />
</Text>
) : (
<Text color={color}></Text>
)}
</Box>
<Box flexGrow={1}>{children}</Box>
</Box>
)
function formatCompactTokens(tokens: number | undefined): string {
const value = Number.isFinite(tokens) ? Math.max(0, tokens || 0) : 0
return new Intl.NumberFormat("en-US", {
notation: "compact",
maximumFractionDigits: 1,
})
.format(value)
.toLowerCase()
}
function formatCompactCost(cost: number | undefined): string {
const value = Number.isFinite(cost) ? Math.max(0, cost || 0) : 0
const maximumFractionDigits = value >= 0.01 ? 2 : 4
return new Intl.NumberFormat("en-US", {
style: "currency",
currency: "USD",
minimumFractionDigits: 2,
maximumFractionDigits,
}).format(value)
}
function formatSubagentStatsValues(
toolCalls: number | undefined,
contextTokens: number | undefined,
totalCost: number | undefined,
latestToolCall?: string,
) {
const safeToolCalls = Number.isFinite(toolCalls) ? Math.max(0, toolCalls || 0) : 0
const toolUses = safeToolCalls === 1 ? "tool use" : "tool uses"
const tokensUsed = formatCompactTokens(contextTokens || 0)
const formattedCost = formatCompactCost(totalCost || 0)
const stats = `${safeToolCalls} ${toolUses} · ${tokensUsed} tokens · ${formattedCost}`
const latestTool = latestToolCall?.trim()
return latestTool ? `${latestTool} · ${stats}` : stats
}
function wrapPrompt(text: string, width: number): string[] {
if (!text) {
return [""]
}
const normalizedWidth = Math.max(1, width)
const wrappedLines: string[] = []
const paragraphs = text.split("\n")
for (const paragraph of paragraphs) {
const words = paragraph.trim().split(/\s+/).filter(Boolean)
if (words.length === 0) {
wrappedLines.push("")
continue
}
let line = ""
for (const word of words) {
if (!line) {
if (word.length <= normalizedWidth) {
line = word
continue
}
let remaining = word
while (remaining.length > normalizedWidth) {
wrappedLines.push(remaining.slice(0, normalizedWidth))
remaining = remaining.slice(normalizedWidth)
}
line = remaining
continue
}
if (line.length + 1 + word.length <= normalizedWidth) {
line = `${line} ${word}`
continue
}
wrappedLines.push(line)
if (word.length <= normalizedWidth) {
line = word
continue
}
let remaining = word
while (remaining.length > normalizedWidth) {
wrappedLines.push(remaining.slice(0, normalizedWidth))
remaining = remaining.slice(normalizedWidth)
}
line = remaining
}
if (line) {
wrappedLines.push(line)
}
}
return wrappedLines.length > 0 ? wrappedLines : [text]
}
const TreePromptRow: React.FC<{
prefix: React.ReactNode
continuationPrefix: string
prompt: string
promptWidth: number
color?: string
}> = ({ prefix, continuationPrefix, prompt, promptWidth, color }) => {
const lines = wrapPrompt(prompt, promptWidth)
return (
<Box flexDirection="column" width="100%">
{lines.map((line, index) => (
<Box flexDirection="row" key={`${line}-${index}`} width="100%">
<Box flexShrink={0} width={TREE_PREFIX_WIDTH}>
{index === 0 ? prefix : <Text color="gray">{continuationPrefix}</Text>}
</Box>
<Box flexGrow={1}>
<Text color={color}>{line}</Text>
</Box>
</Box>
))}
</Box>
)
}
const TreeStatsRow: React.FC<{ prefix: string; stats: string }> = ({ prefix, stats }) => (
<Box flexDirection="row" width="100%">
<Box flexShrink={0} width={TREE_PREFIX_WIDTH}>
<Text color="gray">{prefix}</Text>
</Box>
<Box flexGrow={1}>
<Text color="gray"> {stats}</Text>
</Box>
</Box>
)
export const SubagentMessage: React.FC<SubagentMessageProps> = ({ message, mode, isStreaming }) => {
const { type, ask, say, text, partial } = message
const toolColor = mode === "plan" ? "yellow" : COLORS.primaryBlue
const { columns } = useTerminalSize()
const promptWidth = Math.max(MIN_PROMPT_WIDTH, columns - 2 - TREE_PREFIX_WIDTH)
if ((type === "ask" && ask === "use_subagents") || say === "use_subagents") {
const parsed = text
? jsonParseSafe<ClineAskUseSubagents>(text, {
prompts: [],
})
: { prompts: [] }
const prompts = (parsed.prompts || []).map((prompt) => prompt?.trim()).filter(Boolean)
if (prompts.length === 0) {
return (
<Box flexDirection="column" marginBottom={1} width="100%">
<DotRow color={toolColor}>
<Text color={toolColor}>Cline wants to run subagents:</Text>
</DotRow>
</Box>
)
}
const singular = prompts.length === 1
return (
<Box flexDirection="column" marginBottom={1} width="100%">
<DotRow color={toolColor} flashing={partial === true && isStreaming}>
<Text color={toolColor}>{singular ? "Cline wants to run a subagent:" : "Cline wants to run subagents:"}</Text>
</DotRow>
<Box flexDirection="column" marginLeft={2} width="100%">
{prompts.map((prompt, index) => {
const isLastPrompt = index === prompts.length - 1
const branch = isLastPrompt ? "└─" : "├─"
const continuationPrefix = isLastPrompt ? " " : "│ "
const shouldShowPromptStats = partial !== true || !isLastPrompt
return (
<Box flexDirection="column" key={`${prompt}-${index}`}>
<TreePromptRow
color={toolColor}
continuationPrefix={continuationPrefix}
prefix={<Text color={toolColor}>{`${branch} `}</Text>}
prompt={prompt}
promptWidth={promptWidth}
/>
{shouldShowPromptStats && (
<TreeStatsRow
prefix={continuationPrefix}
stats={formatSubagentStatsValues(undefined, undefined, undefined)}
/>
)}
</Box>
)
})}
</Box>
</Box>
)
}
if (say === "subagent" && text) {
const parsed = jsonParseSafe<ClineSaySubagentStatus>(text, {
status: "running",
total: 0,
completed: 0,
successes: 0,
failures: 0,
toolCalls: 0,
inputTokens: 0,
outputTokens: 0,
contextWindow: 0,
maxContextTokens: 0,
maxContextUsagePercentage: 0,
items: [],
})
const items = parsed.items || []
if (items.length === 0) {
return null
}
return (
<Box flexDirection="column" marginBottom={1} width="100%">
<DotRow color={toolColor} flashing={partial === true && isStreaming}>
<Text color={toolColor}>
{items.length === 1 ? "Cline is running a subagent:" : "Cline is running subagents:"}
</Text>
</DotRow>
<Box flexDirection="column" marginLeft={2} width="100%">
{items.map((entry, index) => {
const isLastEntry = index === items.length - 1
const branch = isLastEntry ? "└─" : "├─"
const continuationPrefix = isLastEntry ? " " : "│ "
const key = `${entry.index}-${index}`
const shouldShowStats = true
if (entry.status === "completed") {
return (
<Box flexDirection="column" key={key}>
<TreePromptRow
color="green"
continuationPrefix={continuationPrefix}
prefix={
<Box flexDirection="row">
<Text color="gray">{`${branch} `}</Text>
<Text color="green"></Text>
</Box>
}
prompt={entry.prompt}
promptWidth={promptWidth}
/>
<TreeStatsRow
prefix={continuationPrefix}
stats={formatSubagentStatsValues(
entry.toolCalls,
entry.contextTokens,
entry.totalCost,
entry.latestToolCall,
)}
/>
</Box>
)
}
if (entry.status === "failed") {
return (
<Box flexDirection="column" key={key}>
<TreePromptRow
color="red"
continuationPrefix={continuationPrefix}
prefix={
<Box flexDirection="row">
<Text color="gray">{`${branch} `}</Text>
<Text color="red"></Text>
</Box>
}
prompt={entry.prompt}
promptWidth={promptWidth}
/>
<TreeStatsRow
prefix={continuationPrefix}
stats={formatSubagentStatsValues(
entry.toolCalls,
entry.contextTokens,
entry.totalCost,
entry.latestToolCall,
)}
/>
</Box>
)
}
return (
<Box flexDirection="column" key={key}>
<TreePromptRow
color={toolColor}
continuationPrefix={continuationPrefix}
prefix={
<Box flexDirection="row">
<Text color="gray">{branch} </Text>
{entry.status === "running" ? (
<Text color={toolColor}>
<Spinner type="dots" />
</Text>
) : (
<Text color={toolColor}></Text>
)}
</Box>
}
prompt={entry.prompt}
promptWidth={promptWidth}
/>
{shouldShowStats && (
<TreeStatsRow
prefix={continuationPrefix}
stats={formatSubagentStatsValues(
entry.toolCalls,
entry.contextTokens,
entry.totalCost,
entry.latestToolCall,
)}
/>
)}
</Box>
)
})}
</Box>
</Box>
)
}
return null
}
-12
View File
@@ -1,12 +0,0 @@
import { describe, expect, it } from "vitest"
import { getAllFeaturedModels } from "./featured-models"
describe("featured models", () => {
it("includes display names for all featured models", () => {
const models = getAllFeaturedModels()
for (const model of models) {
expect(model.name).toBeTruthy()
}
})
})
+22 -28
View File
@@ -7,56 +7,50 @@ export interface FeaturedModel {
id: string
name: string
description: string
labels: string[]
label: string
}
export const FEATURED_MODELS: { recommended: FeaturedModel[]; free: FeaturedModel[] } = {
export const FEATURED_MODELS = {
recommended: [
{
id: "anthropic/claude-opus-4.6",
name: "Claude Opus 4.6",
id: "anthropic/claude-opus-4.5",
name: "Claude Opus 4.5",
description: "State-of-the-art for complex coding",
labels: ["BEST"],
},
{
id: "anthropic/claude-sonnet-4.6",
name: "Claude Sonnet 4.6",
description: "Best balance of speed, cost, and quality",
labels: ["NEW"],
label: "Best",
},
{
id: "openai/gpt-5.2-codex",
name: "GPT 5.2 Codex",
description: "OpenAI's latest with strong coding abilities",
labels: ["HOT"],
label: "New",
},
],
{
id: "google/gemini-3-pro-preview",
name: "Gemini 3 Pro",
description: "1M context window for large codebases",
label: "Trending",
},
] as FeaturedModel[],
free: [
{
id: "anthropic/claude-sonnet-4.6",
name: "Claude Sonnet 4.6",
description: "Best balance of speed, cost, and quality, now free in Cline",
labels: ["FREE"],
},
{
id: "z-ai/glm-5",
name: "Z-AI GLM5",
description: "Z.AI's latest GLM 5 model with strong coding and agent performance",
labels: ["FREE"],
id: "moonshotai/kimi-k2.5",
name: "Kimi K2.5",
description: "State-of-the-art model topping benchmarks",
label: "FREE",
},
{
id: "kwaipilot/kat-coder-pro",
name: "KAT Coder Pro",
description: "KwaiKAT's most advanced agentic coding model in the KAT-Coder series",
labels: ["FREE"],
description: "Advanced agentic coding model",
label: "FREE",
},
{
id: "arcee-ai/trinity-large-preview:free",
name: "Trinity Large Preview",
description: "Arcee AI's advanced large preview model in the Trinity series",
labels: ["FREE"],
description: "US built open source coding model",
label: "FREE",
},
],
] as FeaturedModel[],
}
export function getAllFeaturedModels(): FeaturedModel[] {
-11
View File
@@ -142,17 +142,6 @@ export class CliEnvServiceClient implements EnvServiceClientInterface {
printInfo("Shutting down...")
return proto.cline.Empty.create()
}
async openExternal(request: proto.cline.StringRequest): Promise<proto.cline.Empty> {
const url = request.value || ""
if (url) {
printInfo(`🌐 Opening: ${url}`)
// Dynamically import 'open' to open URL in default browser
const { default: open } = await import("open")
await open(url)
}
return proto.cline.Empty.create()
}
}
/**
+1 -13
View File
@@ -6,7 +6,6 @@
* - Ctrl+A/E: start/end of line
* - Ctrl+W: delete word backwards
* - Ctrl+U: delete to start of line
* - Ctrl+K: delete to end of line
*
* Note: Home/End keys are handled by useHomeEndKeys hook because Ink doesn't
* expose them in useInput (it sets input='' for these keys).
@@ -153,14 +152,6 @@ export function useTextInput(): UseTextInputReturn {
}
}, [])
const deleteToEnd = useCallback(() => {
const pos = cursorRef.current
if (pos < textRef.current.length) {
setTextState((prev) => prev.slice(0, pos))
// Cursor stays at same position (now at end of text)
}
}, [])
// Cursor movement (internal, used by handlers)
const moveToStart = useCallback(() => setCursorPosState(0), [])
const moveToEnd = useCallback(() => setCursorPosState(textRef.current.length), [])
@@ -199,9 +190,6 @@ export function useTextInput(): UseTextInputReturn {
case "u": // Ctrl+U - delete to start
deleteToStart()
return true
case "k": // Ctrl+K - delete to end
deleteToEnd()
return true
case "w": // Ctrl+W - delete word backwards
deleteWordBefore()
return true
@@ -209,7 +197,7 @@ export function useTextInput(): UseTextInputReturn {
return false
}
},
[moveToStart, moveToEnd, deleteToStart, deleteToEnd, deleteWordBefore],
[moveToStart, moveToEnd, deleteToStart, deleteWordBefore],
)
return {
+2 -42
View File
@@ -30,9 +30,7 @@ describe("CLI Commands", () => {
.option("-v, --verbose", "Show verbose output")
.option("-c, --cwd <path>", "Working directory")
.option("--config <path>", "Configuration directory")
.option("--thinking [tokens]", "Enable extended thinking")
.option("--reasoning-effort <effort>", "Reasoning effort")
.option("--max-consecutive-mistakes <count>", "Maximum consecutive mistakes")
.option("--thinking", "Enable extended thinking")
.action(() => {})
program
@@ -69,9 +67,7 @@ describe("CLI Commands", () => {
.option("-v, --verbose", "Verbose output")
.option("-c, --cwd <path>", "Working directory")
.option("--config <path>", "Configuration directory")
.option("--thinking [tokens]", "Enable extended thinking")
.option("--reasoning-effort <effort>", "Reasoning effort")
.option("--max-consecutive-mistakes <count>", "Maximum consecutive mistakes")
.option("--thinking", "Enable extended thinking")
.action(() => {})
})
@@ -150,27 +146,6 @@ describe("CLI Commands", () => {
expect(taskCmd.opts().thinking).toBe(true)
})
it("should parse --thinking with token budget", () => {
const taskCmd = program.commands.find((c) => c.name() === "task")!
const args = ["test prompt", "--thinking", "8000"]
taskCmd.parse(args, { from: "user" })
expect(taskCmd.opts().thinking).toBe("8000")
})
it("should parse --reasoning-effort option", () => {
const taskCmd = program.commands.find((c) => c.name() === "task")!
const args = ["test prompt", "--reasoning-effort", "high"]
taskCmd.parse(args, { from: "user" })
expect(taskCmd.opts().reasoningEffort).toBe("high")
})
it("should parse --max-consecutive-mistakes option", () => {
const taskCmd = program.commands.find((c) => c.name() === "task")!
const args = ["test prompt", "--max-consecutive-mistakes", "999"]
taskCmd.parse(args, { from: "user" })
expect(taskCmd.opts().maxConsecutiveMistakes).toBe("999")
})
it("should parse short flags", () => {
const taskCmd = program.commands.find((c) => c.name() === "task")!
const args = ["test prompt", "-a", "-v", "-m", "gpt-4"]
@@ -306,21 +281,6 @@ describe("CLI Commands", () => {
program.parse(["node", "cli", "--thinking"])
expect(program.opts().thinking).toBe(true)
})
it("should parse --thinking with token budget", () => {
program.parse(["node", "cli", "--thinking", "4096"])
expect(program.opts().thinking).toBe("4096")
})
it("should parse --reasoning-effort option", () => {
program.parse(["node", "cli", "--reasoning-effort", "medium"])
expect(program.opts().reasoningEffort).toBe("medium")
})
it("should parse --max-consecutive-mistakes option", () => {
program.parse(["node", "cli", "--max-consecutive-mistakes", "7"])
expect(program.opts().maxConsecutiveMistakes).toBe("7")
})
})
describe("command structure", () => {
+179 -433
View File
@@ -15,14 +15,13 @@ import { HostProvider } from "@/hosts/host-provider"
import { FileEditProvider } from "@/integrations/editor/FileEditProvider"
import { openAiCodexOAuthManager } from "@/integrations/openai-codex/oauth"
import { StandaloneTerminalManager } from "@/integrations/terminal/standalone/StandaloneTerminalManager"
import { BannerService } from "@/services/banner/BannerService"
import { ErrorService } from "@/services/error/ErrorService"
import { initializeDistinctId } from "@/services/logging/distinctId"
import { telemetryService } from "@/services/telemetry"
import { PostHogClientProvider } from "@/services/telemetry/providers/posthog/PostHogClientProvider"
import { HistoryItem } from "@/shared/HistoryItem"
import { Logger } from "@/shared/services/Logger"
import { Session } from "@/shared/services/Session"
import { getProviderModelIdKey, ProviderToApiKeyMap } from "@/shared/storage"
import { isOpenaiReasoningEffort, OPENAI_REASONING_EFFORT_OPTIONS, type OpenaiReasoningEffort } from "@/shared/storage/types"
import { version as CLI_VERSION } from "../package.json"
import { runAcpMode } from "./acp/index.js"
import { App } from "./components/App"
@@ -32,7 +31,6 @@ import { CliCommentReviewController } from "./controllers/CliCommentReviewContro
import { CliWebviewProvider } from "./controllers/CliWebviewProvider"
import { restoreConsole } from "./utils/console"
import { printInfo, printWarning } from "./utils/display"
import { selectOutputMode } from "./utils/mode-selection"
import { parseImagesFromInput, processImagePaths } from "./utils/parser"
import { CLINE_CLI_DIR, getCliBinaryPath } from "./utils/path"
import { readStdinIfPiped } from "./utils/piped"
@@ -43,280 +41,11 @@ import { autoUpdateOnStartup, checkForUpdates } from "./utils/update"
import { initializeCliContext } from "./vscode-context"
import { CLI_LOG_FILE, shutdownEvent, window } from "./vscode-shim"
/**
* Common options shared between runTask and resumeTask
*/
interface TaskOptions {
act?: boolean
plan?: boolean
model?: string
verbose?: boolean
cwd?: string
config?: string
thinking?: boolean | string
reasoningEffort?: string
maxConsecutiveMistakes?: string
yolo?: boolean
doubleCheckCompletion?: boolean
timeout?: string
json?: boolean
stdinWasPiped?: boolean
}
let telemetryDisposed = false
async function disposeTelemetryServices(): Promise<void> {
if (telemetryDisposed) {
return
}
telemetryDisposed = true
await Promise.allSettled([telemetryService.dispose(), PostHogClientProvider.getInstance().dispose()])
}
/**
* Restore yoloModeToggled to its original value from before this CLI session.
* This ensures the --yolo flag is session-only and doesn't leak into future runs.
* Must be called before flushPendingState so the restored value gets persisted.
*/
function restoreYoloState(): void {
if (savedYoloModeToggled !== null) {
try {
StateManager.get().setGlobalState("yoloModeToggled", savedYoloModeToggled)
savedYoloModeToggled = null
} catch {
// StateManager may not be initialized (e.g., early exit before init)
}
}
}
async function disposeCliContext(ctx: CliContext): Promise<void> {
restoreYoloState()
await ctx.controller.stateManager.flushPendingState()
await ctx.controller.dispose()
await ErrorService.get().dispose()
await disposeTelemetryServices()
}
function setModeScopedState(currentMode: "act" | "plan", setter: (mode: "act" | "plan") => void): void {
const stateManager = StateManager.get()
setter(currentMode)
const separateModels = stateManager.getGlobalSettingsKey("planActSeparateModelsSetting") ?? false
if (!separateModels) {
const otherMode: "act" | "plan" = currentMode === "act" ? "plan" : "act"
setter(otherMode)
}
}
function normalizeReasoningEffort(value?: string): OpenaiReasoningEffort | undefined {
if (value === undefined) {
return undefined
}
const normalized = value.toLowerCase()
if (isOpenaiReasoningEffort(normalized)) {
return normalized
}
printWarning(
`Invalid --reasoning-effort '${value}'. Using 'medium'. Valid values: ${OPENAI_REASONING_EFFORT_OPTIONS.join(", ")}.`,
)
return "medium"
}
function normalizeMaxConsecutiveMistakes(value?: string): number | undefined {
if (value === undefined) {
return undefined
}
const parsed = Number.parseInt(value, 10)
if (Number.isNaN(parsed) || parsed < 1) {
printWarning(`Invalid --max-consecutive-mistakes value '${value}'. Expected integer >= 1.`)
return undefined
}
return parsed
}
/**
* Apply task-related options (mode, model, thinking, yolo) to StateManager.
* Shared between runTask and resumeTask to avoid duplication.
*/
function applyTaskOptions(options: TaskOptions): void {
// Apply mode flag
if (options.plan) {
StateManager.get().setGlobalState("mode", "plan")
telemetryService.captureHostEvent("mode_flag", "plan")
} else if (options.act) {
StateManager.get().setGlobalState("mode", "act")
telemetryService.captureHostEvent("mode_flag", "act")
}
// Apply model override if specified
if (options.model) {
const selectedMode = (StateManager.get().getGlobalSettingsKey("mode") || "act") as "act" | "plan"
const providerKey = selectedMode === "act" ? "actModeApiProvider" : "planModeApiProvider"
const currentProvider = StateManager.get().getGlobalSettingsKey(providerKey) as ApiProvider
const modelKey = getProviderModelIdKey(currentProvider, selectedMode)
if (modelKey) {
StateManager.get().setGlobalState(modelKey, options.model)
}
telemetryService.captureHostEvent("model_flag", options.model)
}
// Set thinking budget based on --thinking flag (boolean or number)
let thinkingBudget = 0
if (options.thinking) {
if (typeof options.thinking === "string") {
const parsed = Number.parseInt(options.thinking, 10)
if (Number.isNaN(parsed) || parsed < 0) {
printWarning(`Invalid --thinking value '${options.thinking}'. Using default 1024.`)
thinkingBudget = 1024
} else {
thinkingBudget = parsed
}
} else {
thinkingBudget = 1024
}
}
const currentMode = (StateManager.get().getGlobalSettingsKey("mode") || "act") as "act" | "plan"
setModeScopedState(currentMode, (mode) => {
const thinkingKey = mode === "act" ? "actModeThinkingBudgetTokens" : "planModeThinkingBudgetTokens"
StateManager.get().setGlobalState(thinkingKey, thinkingBudget)
})
if (options.thinking) {
telemetryService.captureHostEvent("thinking_flag", "true")
}
const reasoningEffort = normalizeReasoningEffort(options.reasoningEffort)
if (reasoningEffort !== undefined) {
setModeScopedState(currentMode, (mode) => {
const reasoningKey = mode === "act" ? "actModeReasoningEffort" : "planModeReasoningEffort"
StateManager.get().setGlobalState(reasoningKey, reasoningEffort)
})
telemetryService.captureHostEvent("reasoning_effort_flag", reasoningEffort)
}
const maxConsecutiveMistakes = normalizeMaxConsecutiveMistakes(options.maxConsecutiveMistakes)
if (maxConsecutiveMistakes !== undefined) {
StateManager.get().setGlobalState("maxConsecutiveMistakes", maxConsecutiveMistakes)
telemetryService.captureHostEvent("max_consecutive_mistakes_flag", String(maxConsecutiveMistakes))
}
// Override yolo mode only if --yolo flag is explicitly passed.
// The original value is saved in initializeCli and restored on exit.
if (options.yolo) {
const state = StateManager.get()
savedYoloModeToggled = state.getGlobalSettingsKey("yoloModeToggled") ?? false
state.setGlobalState("yoloModeToggled", true)
telemetryService.captureHostEvent("yolo_flag", "true")
}
// Set double-check completion based on flag
if (options.doubleCheckCompletion) {
StateManager.get().setGlobalState("doubleCheckCompletionEnabled", true)
telemetryService.captureHostEvent("double_check_completion_flag", "true")
}
}
/**
* Get mode selection result using the extracted, testable selectOutputMode function.
* This wrapper provides the current process TTY state.
*/
function getModeSelection(options: TaskOptions) {
return selectOutputMode({
stdoutIsTTY: process.stdout.isTTY === true,
stdinIsTTY: process.stdin.isTTY === true,
stdinWasPiped: options.stdinWasPiped ?? false,
json: options.json,
yolo: options.yolo,
})
}
/**
* Determine if plain text mode should be used based on options and environment.
*/
function shouldUsePlainTextMode(options: TaskOptions): boolean {
return getModeSelection(options).usePlainTextMode
}
/**
* Get the reason for using plain text mode (for telemetry).
*/
function getPlainTextModeReason(options: TaskOptions): string {
return getModeSelection(options).reason
}
/**
* Run a task in plain text mode (no Ink UI).
* Handles auth check, task execution, cleanup, and exit.
*/
async function runTaskInPlainTextMode(
ctx: CliContext,
options: TaskOptions,
taskConfig: {
prompt?: string
taskId?: string
imageDataUrls?: string[]
},
): Promise<never> {
// Set flag so shutdown handler knows not to clear Ink UI lines
isPlainTextMode = true
// Check if auth is configured before attempting to run the task
// In plain text mode we can't show the interactive auth flow
const hasAuth = await isAuthConfigured()
if (!hasAuth) {
printWarning("Not authenticated. Please run 'cline auth' first to configure your API credentials.")
await disposeCliContext(ctx)
exit(1)
}
const reason = getPlainTextModeReason(options)
telemetryService.captureHostEvent("plain_text_mode", reason)
// Plain text mode: no Ink rendering, just clean text output
const success = await runPlainTextTask({
controller: ctx.controller,
prompt: taskConfig.prompt,
taskId: taskConfig.taskId,
imageDataUrls: taskConfig.imageDataUrls,
verbose: options.verbose,
jsonOutput: options.json,
timeoutSeconds: options.timeout ? Number.parseInt(options.timeout, 10) : undefined,
})
// Cleanup
await disposeCliContext(ctx)
// Ensure stdout is fully drained before exiting - critical for piping
await drainStdout()
exit(success ? 0 : 1)
}
/**
* Create the standard cleanup function for Ink apps.
*/
function createInkCleanup(ctx: CliContext, onTaskError?: () => boolean): () => Promise<void> {
return async () => {
await disposeCliContext(ctx)
if (onTaskError?.()) {
printWarning("Task ended with errors.")
exit(1)
}
exit(0)
}
}
// Track active context for graceful shutdown
let activeContext: CliContext | null = null
let isShuttingDown = false
// Track if we're in plain text mode (no Ink UI) - set by runTask when piped stdin detected
let isPlainTextMode = false
// Track the original yoloModeToggled value from before this CLI session so we can restore it on exit.
// The --yolo flag should only affect the current invocation, not persist across runs.
let savedYoloModeToggled: boolean | null = null
/**
* Wait for stdout to fully drain before exiting.
@@ -358,26 +87,15 @@ function setupSignalHandlers() {
printWarning(`${signal} received, shutting down...`)
try {
// Restore yolo state before any cleanup - this is idempotent and safe
// even if disposeCliContext also calls it (restoreYoloState checks savedYoloModeToggled !== null)
restoreYoloState()
if (activeContext) {
const task = activeContext.controller.task
if (task) {
task.abortTask()
}
await disposeCliContext(activeContext)
} else {
// Best-effort flush of restored yolo state when no active context
try {
await StateManager.get().flushPendingState()
} catch {
// StateManager may not be initialized yet
}
await ErrorService.get().dispose()
await disposeTelemetryServices()
await activeContext.controller.stateManager.flushPendingState()
await activeContext.controller.dispose()
}
await ErrorService.get().dispose()
} catch {
// Best effort cleanup
}
@@ -430,17 +148,8 @@ async function initializeCli(options: InitOptions): Promise<CliContext> {
workspaceDir: workspacePath,
})
// Set up output channel and Logger early so ClineEndpoint.initialize logs are captured
const outputChannel = window.createOutputChannel("Cline CLI")
const logToChannel = (message: string) => outputChannel.appendLine(message)
// Configure the shared Logging class early to capture all initialization logs
Logger.subscribe(logToChannel)
await ClineEndpoint.initialize(EXTENSION_DIR)
// Auto-update check (after endpoints initialized, so we can detect bundled configs)
autoUpdateOnStartup(CLI_VERSION)
await ClineEndpoint.initialize()
await initializeDistinctId(extensionContext)
// Initialize/reset session tracking for this CLI run
Session.reset()
@@ -449,9 +158,11 @@ async function initializeCli(options: InitOptions): Promise<CliContext> {
AuthHandler.getInstance().setEnabled(true)
}
const outputChannel = window.createOutputChannel("Cline CLI")
outputChannel.appendLine(
`Cline CLI initialized. Data dir: ${DATA_DIR}, Extension dir: ${EXTENSION_DIR}, Log dir: ${CLINE_CLI_DIR.log}`,
)
const logToChannel = (message: string) => outputChannel.appendLine(message)
HostProvider.initialize(
() => new CliWebviewProvider(extensionContext as any),
@@ -460,24 +171,28 @@ async function initializeCli(options: InitOptions): Promise<CliContext> {
() => new StandaloneTerminalManager(),
createCliHostBridgeProvider(workspacePath),
logToChannel,
async (path: string) => (options.enableAuth ? AuthHandler.getInstance().getCallbackUrl(path) : ""),
async () => (options.enableAuth ? AuthHandler.getInstance().getCallbackUrl() : ""),
getCliBinaryPath,
EXTENSION_DIR,
DATA_DIR,
)
await StateManager.initialize(extensionContext as any)
await ErrorService.initialize()
// Initialize OpenAI Codex OAuth manager with extension context for secrets storage
openAiCodexOAuthManager.initialize(extensionContext)
// Configure the shared Logging class to use HostProvider's output channel
Logger.subscribe((msg: string) => HostProvider.get().logToChannel(msg))
const webview = HostProvider.get().createWebviewProvider() as CliWebviewProvider
const controller = webview.controller
await telemetryService.captureExtensionActivated()
await telemetryService.captureHostEvent("cline_cli", "initialized")
BannerService.initialize(webview.controller)
telemetryService.captureExtensionActivated()
telemetryService.captureHostEvent("cline_cli", "initialized")
const ctx = { extensionContext, dataDir: DATA_DIR, extensionDir: EXTENSION_DIR, workspacePath, controller }
activeContext = ctx
@@ -513,7 +228,24 @@ async function runInkApp(element: React.ReactElement, cleanup: () => Promise<voi
/**
* Run a task with the given prompt - uses welcome view for consistent behavior
*/
async function runTask(prompt: string, options: TaskOptions & { images?: string[] }, existingContext?: CliContext) {
async function runTask(
prompt: string,
options: {
act?: boolean
plan?: boolean
model?: string
verbose?: boolean
cwd?: string
config?: string
thinking?: boolean
yolo?: boolean
timeout?: string
images?: string[]
json?: boolean
stdinWasPiped?: boolean
},
existingContext?: CliContext,
) {
const ctx = existingContext || (await initializeCli({ ...options, enableAuth: true }))
// Parse images from the prompt text (e.g., @/path/to/image.png)
@@ -530,23 +262,101 @@ async function runTask(prompt: string, options: TaskOptions & { images?: string[
// Task without prompt starts in interactive mode
telemetryService.captureHostEvent("task_command", prompt ? "task" : "interactive")
// Apply shared task options (mode, model, thinking, yolo)
applyTaskOptions(options)
await StateManager.get().flushPendingState()
// Use plain text mode when output is redirected, stdin was piped, JSON mode is enabled, or --yolo flag is used
if (shouldUsePlainTextMode(options)) {
return runTaskInPlainTextMode(ctx, options, {
prompt: taskPrompt,
imageDataUrls: imageDataUrls.length > 0 ? imageDataUrls : undefined,
})
if (options.plan) {
StateManager.get().setGlobalState("mode", "plan")
telemetryService.captureHostEvent("mode_flag", "plan")
} else if (options.act) {
StateManager.get().setGlobalState("mode", "act")
telemetryService.captureHostEvent("mode_flag", "act")
}
if (options.model) {
const selectedMode = (StateManager.get().getGlobalSettingsKey("mode") || "act") as "act" | "plan"
// Get the current provider for the selected mode
const providerKey = selectedMode === "act" ? "actModeApiProvider" : "planModeApiProvider"
const currentProvider = StateManager.get().getGlobalSettingsKey(providerKey) as ApiProvider
// Update model ID using provider-specific key (e.g., cline uses actModeOpenRouterModelId)
const modelKey = getProviderModelIdKey(currentProvider, selectedMode)
if (modelKey) {
StateManager.get().setGlobalState(modelKey, options.model)
}
telemetryService.captureHostEvent("model_flag", options.model)
}
// Set thinking budget based on --thinking flag
const thinkingBudget = options.thinking ? 1024 : 0
const currentMode = StateManager.get().getGlobalSettingsKey("mode") || "act"
const thinkingKey = currentMode === "act" ? "actModeThinkingBudgetTokens" : "planModeThinkingBudgetTokens"
StateManager.get().setGlobalState(thinkingKey, thinkingBudget)
if (options.thinking) {
telemetryService.captureHostEvent("thinking_flag", "true")
}
// Set yolo mode based on --yolo flag
if (options.yolo) {
StateManager.get().setGlobalState("yoloModeToggled", true)
telemetryService.captureHostEvent("yolo_flag", "true")
}
await StateManager.get().flushPendingState()
// Detect if output is a TTY (interactive terminal) or redirected to a file/pipe
const isTTY = process.stdout.isTTY === true
// Use plain text mode when output is redirected, stdin was piped, JSON mode is enabled, or --yolo flag is used
// Ink requires raw mode on stdin which isn't available when stdin is piped
// Note: we use the stdinWasPiped flag passed from the caller because process.stdin.isTTY
// may not be reliable after stdin has been consumed by readStdinIfPiped()
if (!isTTY || options.stdinWasPiped || options.json || options.yolo) {
// Set flag so shutdown handler knows not to clear Ink UI lines
isPlainTextMode = true
// Check if auth is configured before attempting to run the task
// In plain text mode we can't show the interactive auth flow
const hasAuth = await isAuthConfigured()
if (!hasAuth) {
printWarning("Not authenticated. Please run 'cline auth' first to configure your API credentials.")
await ctx.controller.stateManager.flushPendingState()
await ctx.controller.dispose()
await ErrorService.get().dispose()
exit(1)
}
const reason = options.yolo
? "yolo_flag"
: options.json
? "json"
: options.stdinWasPiped
? "piped_stdin"
: "redirected_output"
telemetryService.captureHostEvent("plain_text_mode", reason)
// Plain text mode: no Ink rendering, just clean text output
const success = await runPlainTextTask({
controller: ctx.controller,
prompt: taskPrompt,
imageDataUrls: imageDataUrls.length > 0 ? imageDataUrls : undefined,
verbose: options.verbose,
jsonOutput: options.json,
timeoutSeconds: options.timeout ? parseInt(options.timeout, 10) : undefined,
})
// Cleanup
await ctx.controller.stateManager.flushPendingState()
await ctx.controller.dispose()
await ErrorService.get().dispose()
// Ensure stdout is fully drained before exiting - critical for piping
await drainStdout()
exit(success ? 0 : 1)
}
// Interactive mode: Render the welcome view with optional initial prompt/images
// If prompt provided (cline task "prompt"), ChatView will auto-submit
// If no prompt (cline interactive), user will type it in
let taskError = false
// Render the welcome view with optional initial prompt/images
// If prompt provided (cline task "prompt"), ChatView will auto-submit
// If no prompt (cline interactive), user will type it in
await runInkApp(
React.createElement(App, {
view: "welcome",
@@ -559,10 +369,20 @@ async function runTask(prompt: string, options: TaskOptions & { images?: string[
taskError = true
},
onWelcomeExit: () => {
// User pressed Esc; Ink exits and cleanup handles process exit.
// User pressed Esc
exit(0)
},
}),
createInkCleanup(ctx, () => taskError),
async () => {
await ctx.controller.stateManager.flushPendingState()
await ctx.controller.dispose()
await ErrorService.get().dispose()
if (taskError) {
printWarning("Task ended with errors.")
exit(1)
}
exit(0)
},
)
}
@@ -575,8 +395,8 @@ async function listHistory(options: { config?: string; limit?: number; page?: nu
const taskHistory = StateManager.get().getGlobalStateKey("taskHistory") || []
// Sort by timestamp (newest first) before pagination
const sortedHistory = [...taskHistory].sort((a: any, b: any) => (b.ts || 0) - (a.ts || 0))
const limit = typeof options.limit === "string" ? Number.parseInt(options.limit, 10) : options.limit || 10
const initialPage = typeof options.page === "string" ? Number.parseInt(options.page, 10) : options.page || 1
const limit = typeof options.limit === "string" ? parseInt(options.limit, 10) : options.limit || 10
const initialPage = typeof options.page === "string" ? parseInt(options.page, 10) : options.page || 1
const totalCount = sortedHistory.length
const totalPages = Math.ceil(totalCount / limit)
@@ -584,7 +404,9 @@ async function listHistory(options: { config?: string; limit?: number; page?: nu
if (sortedHistory.length === 0) {
printInfo("No task history found.")
await disposeCliContext(ctx)
await ctx.controller.stateManager.flushPendingState()
await ctx.controller.dispose()
await ErrorService.get().dispose()
exit(0)
}
@@ -598,7 +420,9 @@ async function listHistory(options: { config?: string; limit?: number; page?: nu
isRawModeSupported: checkRawModeSupport(),
}),
async () => {
await disposeCliContext(ctx)
await ctx.controller.stateManager.flushPendingState()
await ctx.controller.dispose()
await ErrorService.get().dispose()
exit(0)
},
)
@@ -627,7 +451,9 @@ async function showConfig(options: { config?: string }) {
isRawModeSupported: checkRawModeSupport(),
}),
async () => {
await disposeCliContext(ctx)
await ctx.controller.stateManager.flushPendingState()
await ctx.controller.dispose()
await ErrorService.get().dispose()
exit(0)
},
)
@@ -703,15 +529,17 @@ async function runAuth(options: {
baseurl: options.baseurl,
})
await ctx.controller.stateManager.flushPendingState()
await ctx.controller.dispose()
await ErrorService.get().dispose()
if (!result.success) {
printWarning(result.error || "Quick setup failed")
await telemetryService.captureHostEvent("auth", "error")
await disposeCliContext(ctx)
telemetryService.captureHostEvent("auth", "error")
exit(1)
}
await telemetryService.captureHostEvent("auth", "completed")
await disposeCliContext(ctx)
telemetryService.captureHostEvent("auth", "completed")
exit(0)
}
@@ -725,6 +553,7 @@ async function runAuth(options: {
isRawModeSupported: checkRawModeSupport(),
onComplete: () => {
telemetryService.captureHostEvent("auth", "completed")
exit(0)
},
onError: () => {
telemetryService.captureHostEvent("auth", "error")
@@ -732,10 +561,16 @@ async function runAuth(options: {
},
}),
async () => {
await disposeCliContext(ctx)
exit(authError ? 1 : 0)
await ctx.controller.stateManager.flushPendingState()
await ctx.controller.dispose()
await ErrorService.get().dispose()
exit(0)
},
)
if (authError) {
process.exit(1)
}
}
// Setup CLI commands
@@ -759,18 +594,9 @@ program
.option("-v, --verbose", "Show verbose output")
.option("-c, --cwd <path>", "Working directory for the task")
.option("--config <path>", "Path to Cline configuration directory")
.option("--thinking [tokens]", "Enable extended thinking (default: 1024 tokens)")
.option("--reasoning-effort <effort>", "Reasoning effort: none|low|medium|high|xhigh")
.option("--max-consecutive-mistakes <count>", "Maximum consecutive mistakes before halting in yolo mode")
.option("--thinking", "Enable extended thinking (1024 token budget)")
.option("--json", "Output messages as JSON instead of styled text")
.option("--double-check-completion", "Reject first completion attempt to force re-verification")
.option("-T, --taskId <id>", "Resume an existing task by ID")
.action((prompt, options) => {
if (options.taskId) {
return resumeTask(options.taskId, { ...options, initialPrompt: prompt })
}
return runTask(prompt, options)
})
.action((prompt, options) => runTask(prompt, options))
program
.command("history")
@@ -790,9 +616,9 @@ program
program
.command("auth")
.description("Authenticate a provider and configure what model is used")
.option("-p, --provider <id>", "Provider ID for quick setup (e.g., openai-native, anthropic, moonshot)")
.option("-p, --provider <id>", "Provider ID for quick setup (e.g., openai-native, anthropic)")
.option("-k, --apikey <key>", "API key for the provider")
.option("-m, --modelid <id>", "Model ID to configure (e.g., gpt-4o, claude-sonnet-4-6, kimi-k2.5)")
.option("-m, --modelid <id>", "Model ID to configure (e.g., gpt-4o, claude-sonnet-4-5-20250929)")
.option("-b, --baseurl <url>", "Base URL (optional, only for openai provider)")
.option("-v, --verbose", "Show verbose output")
.option("-c, --cwd <path>", "Working directory for the task")
@@ -883,67 +709,6 @@ async function checkAnyProviderConfigured(): Promise<boolean> {
return false
}
/**
* Validate that a task exists in history
* @returns The task history item if found, null otherwise
*/
function findTaskInHistory(taskId: string): HistoryItem | null {
const taskHistory = StateManager.get().getGlobalStateKey("taskHistory") || []
return taskHistory.find((item) => item.id === taskId) || null
}
/**
* Resume an existing task by ID
* Loads the task and optionally prefills the input with a prompt
*/
async function resumeTask(taskId: string, options: TaskOptions & { initialPrompt?: string }) {
const ctx = await initializeCli({ ...options, enableAuth: true })
// Validate task exists
const historyItem = findTaskInHistory(taskId)
if (!historyItem) {
printWarning(`Task not found: ${taskId}`)
printInfo("Use 'cline history' to see available tasks.")
await disposeCliContext(ctx)
exit(1)
}
telemetryService.captureHostEvent("resume_task_command", options.initialPrompt ? "with_prompt" : "interactive")
// Apply shared task options (mode, model, thinking, yolo)
applyTaskOptions(options)
await StateManager.get().flushPendingState()
// Use plain text mode for non-interactive scenarios
if (shouldUsePlainTextMode(options)) {
return runTaskInPlainTextMode(ctx, options, {
prompt: options.initialPrompt,
taskId: taskId,
})
}
// Interactive mode: render the task view with the existing task
let taskError = false
await runInkApp(
React.createElement(App, {
view: "task",
taskId: taskId,
verbose: options.verbose,
controller: ctx.controller,
isRawModeSupported: checkRawModeSupport(),
initialPrompt: options.initialPrompt || undefined,
onError: () => {
taskError = true
},
onWelcomeExit: () => {
// User pressed Esc; Ink exits and cleanup handles process exit.
},
}),
createInkCleanup(ctx, () => taskError),
)
}
/**
* Show welcome prompt and wait for user input
* If auth is not configured, show auth flow first
@@ -964,14 +729,16 @@ async function showWelcome(options: { verbose?: boolean; cwd?: string; config?:
controller: ctx.controller,
isRawModeSupported: checkRawModeSupport(),
onWelcomeExit: () => {
// User pressed Esc; Ink exits and cleanup handles process exit.
exit(0)
},
onError: () => {
hadError = true
},
}),
async () => {
await disposeCliContext(ctx)
await ctx.controller.stateManager.flushPendingState()
await ctx.controller.dispose()
await ErrorService.get().dispose()
exit(hadError ? 1 : 0)
},
)
@@ -988,13 +755,9 @@ program
.option("-v, --verbose", "Show verbose output")
.option("-c, --cwd <path>", "Working directory")
.option("--config <path>", "Configuration directory")
.option("--thinking [tokens]", "Enable extended thinking (default: 1024 tokens)")
.option("--reasoning-effort <effort>", "Reasoning effort: none|low|medium|high|xhigh")
.option("--max-consecutive-mistakes <count>", "Maximum consecutive mistakes before halting in yolo mode")
.option("--thinking", "Enable extended thinking (1024 token budget)")
.option("--json", "Output messages as JSON instead of styled text")
.option("--double-check-completion", "Reject first completion attempt to force re-verification")
.option("--acp", "Run in ACP (Agent Client Protocol) mode for editor integration")
.option("-T, --taskId <id>", "Resume an existing task by ID")
.action(async (prompt, options) => {
// Check for ACP mode first - this takes precedence over everything else
if (options.acp) {
@@ -1009,18 +772,8 @@ program
// Always check for piped stdin content
const stdinInput = await readStdinIfPiped()
// Track whether stdin was actually piped (even if empty) vs not piped (null)
// stdinInput === null means stdin wasn't piped (TTY or not FIFO/file)
// stdinInput === "" means stdin was piped but empty
// stdinInput has content means stdin was piped with data
const stdinWasPiped = stdinInput !== null
// Error if stdin was piped but empty AND no prompt was provided
// This handles:
// - `echo "" | cline` -> error (empty stdin, no prompt)
// - `cline "prompt"` in GitHub Actions -> OK (empty stdin ignored, has prompt)
// - `cat file | cline "explain"` -> OK (has stdin AND prompt)
if (stdinInput === "" && !prompt) {
// Error if stdin was piped but empty (e.g., `echo "" | cline`)
if (stdinInput === "") {
printWarning("Empty input received from stdin. Please provide content to process.")
exit(1)
}
@@ -1043,24 +796,17 @@ program
}
}
// Handle --taskId flag to resume an existing task
if (options.taskId) {
await resumeTask(options.taskId, {
...options,
initialPrompt: effectivePrompt,
stdinWasPiped,
})
return
}
if (effectivePrompt) {
// Pass stdinWasPiped flag so runTask knows to use plain text mode
await runTask(effectivePrompt, { ...options, stdinWasPiped })
await runTask(effectivePrompt, { ...options, stdinWasPiped: !!stdinInput })
} else {
// Show welcome prompt if no prompt given
await showWelcome(options)
}
})
// Background auto-update check (non-blocking)
autoUpdateOnStartup(CLI_VERSION)
// Parse and run
program.parse()
-10
View File
@@ -1,10 +0,0 @@
/**
* Opens a URL in the user's default browser.
* Uses dynamic import of the 'open' package to open URLs.
*
* @param url - The URL to open in the browser
*/
export async function openUrlInBrowser(url: string): Promise<void> {
const { default: open } = await import("open")
await open(url)
}
-194
View File
@@ -1,194 +0,0 @@
import { describe, expect, it } from "vitest"
import { selectOutputMode } from "./mode-selection"
describe("selectOutputMode", () => {
describe("interactive mode (Ink)", () => {
it("should use interactive mode when both stdin and stdout are TTY", () => {
const result = selectOutputMode({
stdoutIsTTY: true,
stdinIsTTY: true,
stdinWasPiped: false,
})
expect(result.usePlainTextMode).toBe(false)
expect(result.reason).toBe("interactive")
})
})
describe("yolo flag", () => {
it("should use plain text mode when --yolo flag is set", () => {
const result = selectOutputMode({
stdoutIsTTY: true,
stdinIsTTY: true,
stdinWasPiped: false,
yolo: true,
})
expect(result.usePlainTextMode).toBe(true)
expect(result.reason).toBe("yolo_flag")
})
it("should prioritize yolo over other flags", () => {
const result = selectOutputMode({
stdoutIsTTY: false,
stdinIsTTY: false,
stdinWasPiped: true,
json: true,
yolo: true,
})
expect(result.reason).toBe("yolo_flag")
})
})
describe("json flag", () => {
it("should use plain text mode when --json flag is set", () => {
const result = selectOutputMode({
stdoutIsTTY: true,
stdinIsTTY: true,
stdinWasPiped: false,
json: true,
})
expect(result.usePlainTextMode).toBe(true)
expect(result.reason).toBe("json")
})
})
describe("piped stdin", () => {
it("should use plain text mode when stdin was piped (echo x | cline)", () => {
const result = selectOutputMode({
stdoutIsTTY: true,
stdinIsTTY: false, // piped stdin is not a TTY
stdinWasPiped: true,
})
expect(result.usePlainTextMode).toBe(true)
expect(result.reason).toBe("piped_stdin")
})
it("should use plain text mode when stdin was piped but empty (echo '' | cline 'prompt')", () => {
const result = selectOutputMode({
stdoutIsTTY: true,
stdinIsTTY: false,
stdinWasPiped: true, // empty pipe still counts as piped
})
expect(result.usePlainTextMode).toBe(true)
expect(result.reason).toBe("piped_stdin")
})
})
describe("stdin redirected (< /dev/null)", () => {
it("should use plain text mode when stdin is redirected from /dev/null", () => {
// cline "prompt" < /dev/null
// stdin is not a TTY, but also not a FIFO/file, so stdinWasPiped=false
const result = selectOutputMode({
stdoutIsTTY: true,
stdinIsTTY: false, // redirected, not a TTY
stdinWasPiped: false, // /dev/null is a character device, not FIFO
})
expect(result.usePlainTextMode).toBe(true)
expect(result.reason).toBe("stdin_redirected")
})
})
describe("stdout redirected", () => {
it("should use plain text mode when stdout is redirected to file", () => {
// cline "prompt" > output.txt
const result = selectOutputMode({
stdoutIsTTY: false,
stdinIsTTY: true,
stdinWasPiped: false,
})
expect(result.usePlainTextMode).toBe(true)
expect(result.reason).toBe("stdout_redirected")
})
it("should use plain text mode when stdout is piped", () => {
// cline "prompt" | grep something
const result = selectOutputMode({
stdoutIsTTY: false,
stdinIsTTY: true,
stdinWasPiped: false,
})
expect(result.usePlainTextMode).toBe(true)
expect(result.reason).toBe("stdout_redirected")
})
})
describe("GitHub Actions scenarios", () => {
it("should use plain text mode in GitHub Actions (stdin is empty FIFO)", () => {
// In GitHub Actions: stdin is an empty FIFO pipe
// stdinIsTTY=false, stdinWasPiped=true (FIFO detected)
const result = selectOutputMode({
stdoutIsTTY: true, // GitHub Actions stdout is TTY-like
stdinIsTTY: false,
stdinWasPiped: true, // empty FIFO still counts as piped
})
expect(result.usePlainTextMode).toBe(true)
})
it("should use plain text mode with --yolo in CI", () => {
const result = selectOutputMode({
stdoutIsTTY: false,
stdinIsTTY: false,
stdinWasPiped: false,
yolo: true,
})
expect(result.usePlainTextMode).toBe(true)
expect(result.reason).toBe("yolo_flag")
})
})
describe("real-world scenarios", () => {
it("cline (no args, interactive terminal)", () => {
const result = selectOutputMode({
stdoutIsTTY: true,
stdinIsTTY: true,
stdinWasPiped: false,
})
expect(result.usePlainTextMode).toBe(false)
})
it('cline "prompt" (prompt arg, interactive terminal)', () => {
const result = selectOutputMode({
stdoutIsTTY: true,
stdinIsTTY: true,
stdinWasPiped: false,
})
expect(result.usePlainTextMode).toBe(false)
})
it('cat file | cline "explain"', () => {
const result = selectOutputMode({
stdoutIsTTY: true,
stdinIsTTY: false,
stdinWasPiped: true,
})
expect(result.usePlainTextMode).toBe(true)
})
it('cline --yolo "prompt"', () => {
const result = selectOutputMode({
stdoutIsTTY: true,
stdinIsTTY: true,
stdinWasPiped: false,
yolo: true,
})
expect(result.usePlainTextMode).toBe(true)
})
it('cline "prompt" < /dev/null', () => {
const result = selectOutputMode({
stdoutIsTTY: true,
stdinIsTTY: false,
stdinWasPiped: false,
})
expect(result.usePlainTextMode).toBe(true)
})
it('cline "prompt" > output.log', () => {
const result = selectOutputMode({
stdoutIsTTY: false,
stdinIsTTY: true,
stdinWasPiped: false,
})
expect(result.usePlainTextMode).toBe(true)
})
})
})
-63
View File
@@ -1,63 +0,0 @@
/**
* Mode selection logic for CLI - determines whether to use Ink (interactive) or plain text mode
*
* This is extracted as a pure function for testability. The decision tree:
* - Plain text mode when output is redirected (stdout not TTY)
* - Plain text mode when input is redirected (stdin not TTY) - Ink requires raw mode
* - Plain text mode when stdin was piped (e.g., echo "x" | cline)
* - Plain text mode when --json flag is used
* - Plain text mode when --yolo flag is used
* - Otherwise: Interactive Ink mode
*/
export interface ModeSelectionInput {
/** Is stdout connected to a TTY (interactive terminal)? */
stdoutIsTTY: boolean
/** Is stdin connected to a TTY (interactive terminal)? */
stdinIsTTY: boolean
/** Was stdin piped (FIFO or file), even if empty? */
stdinWasPiped: boolean
/** --json flag for machine-readable output */
json?: boolean
/** --yolo flag for auto-approve mode */
yolo?: boolean
}
export interface ModeSelectionResult {
/** Use plain text mode instead of Ink */
usePlainTextMode: boolean
/** Reason for the mode selection (for telemetry/debugging) */
reason: "interactive" | "yolo_flag" | "json" | "piped_stdin" | "stdin_redirected" | "stdout_redirected"
}
/**
* Determine whether to use plain text mode or interactive Ink mode
*
* @param input - Environment and option flags
* @returns Mode selection result with reason
*/
export function selectOutputMode(input: ModeSelectionInput): ModeSelectionResult {
// Priority order matters - check most specific flags first
if (input.yolo) {
return { usePlainTextMode: true, reason: "yolo_flag" }
}
if (input.json) {
return { usePlainTextMode: true, reason: "json" }
}
if (input.stdinWasPiped) {
return { usePlainTextMode: true, reason: "piped_stdin" }
}
if (!input.stdinIsTTY) {
return { usePlainTextMode: true, reason: "stdin_redirected" }
}
if (!input.stdoutIsTTY) {
return { usePlainTextMode: true, reason: "stdout_redirected" }
}
return { usePlainTextMode: false, reason: "interactive" }
}
-28
View File
@@ -1,28 +0,0 @@
import { afterEach, describe, expect, it, vi } from "vitest"
import { emitTaskStartedMessage } from "./task-start-output"
describe("emitTaskStartedMessage", () => {
afterEach(() => {
vi.restoreAllMocks()
})
it("writes structured task_started JSON to stdout in json mode", () => {
const stdoutWriteSpy = vi.spyOn(process.stdout, "write").mockImplementation(() => true)
const stderrWriteSpy = vi.spyOn(process.stderr, "write").mockImplementation(() => true)
emitTaskStartedMessage("task-123", true)
expect(stdoutWriteSpy).toHaveBeenCalledWith('{"type":"task_started","taskId":"task-123"}\n')
expect(stderrWriteSpy).not.toHaveBeenCalled()
})
it("writes human-readable task started line to stderr in non-json mode", () => {
const stdoutWriteSpy = vi.spyOn(process.stdout, "write").mockImplementation(() => true)
const stderrWriteSpy = vi.spyOn(process.stderr, "write").mockImplementation(() => true)
emitTaskStartedMessage("task-456", false)
expect(stderrWriteSpy).toHaveBeenCalledWith("Task started: task-456\n")
expect(stdoutWriteSpy).not.toHaveBeenCalled()
})
})
+6 -59
View File
@@ -12,24 +12,18 @@
// Console output is intentional here for plain text mode
import type { ClineMessage, ExtensionState } from "@shared/ExtensionMessage"
import { StringRequest } from "@shared/proto/cline/common"
import type { Controller } from "@/core/controller"
import { getRequestRegistry } from "@/core/controller/grpc-handler"
import { subscribeToState } from "@/core/controller/state/subscribeToState"
import { showTaskWithId } from "@/core/controller/task/showTaskWithId"
import { emitTaskStartedMessage } from "./task-start-output"
export interface PlainTextTaskOptions {
controller: Controller
/** Prompt for new task or message to send to resumed task */
prompt?: string
prompt: string
imageDataUrls?: string[]
verbose?: boolean
jsonOutput?: boolean
/** Timeout in seconds (default: 600 = 10 minutes) */
timeoutSeconds?: number
/** Task ID to resume an existing task */
taskId?: string
}
/**
@@ -45,39 +39,17 @@ export interface PlainTextTaskOptions {
export async function runPlainTextTask(options: PlainTextTaskOptions): Promise<boolean> {
const { controller, prompt, imageDataUrls, verbose, jsonOutput } = options
let completionResolve: (reason?: any) => void
let completionResolve: () => void
let completionReject: (reason?: any) => void
const completionPromise = new Promise<string>((res, rej) => {
const completionPromise = new Promise<void>((res, rej) => {
completionResolve = res
completionReject = rej
})
let hasError = false
let hasEmittedTaskStarted = false
// Track which messages have been processed (by timestamp)
const processedMessages = new Map<number, string>()
const isViewTaskOnly = Boolean(options.taskId) && !prompt
// When resuming a task, we need to ignore completion_result messages that existed
// before we sent our new prompt. This timestamp marks the cutoff - only completion
// results AFTER this time should trigger task completion.
const completionCutoffTs = Date.now()
const emitTaskStarted = () => {
if (hasEmittedTaskStarted) {
return
}
const taskId = controller.task?.taskId
if (!taskId) {
return
}
emitTaskStartedMessage(taskId, Boolean(jsonOutput))
hasEmittedTaskStarted = true
}
// Helper to process a message and track completion state
const processMessage = (message: ClineMessage) => {
const ts = message.ts || 0
@@ -95,12 +67,8 @@ export async function runPlainTextTask(options: PlainTextTaskOptions): Promise<b
processedMessages.set(ts, message.text ?? "")
// Check for completion (only on non-partial messages)
// When resuming a task, only consider completion_result messages that appeared
// AFTER we sent our resume message (ts > completionCutoffTs)
if (message.say === "completion_result" || message.ask === "completion_result") {
if (isViewTaskOnly || ts > completionCutoffTs) {
completionResolve()
}
completionResolve()
} else if (message.say === "error" || message.ask === "api_req_failed") {
completionReject(message.text ?? "message.say error || message.ask api_req_failed")
}
@@ -131,29 +99,8 @@ export async function runPlainTextTask(options: PlainTextTaskOptions): Promise<b
)
try {
// Either resume an existing task or start a new one
if (options.taskId) {
// Load the existing task
await showTaskWithId(controller, StringRequest.create({ value: options.taskId }))
emitTaskStarted()
// If a prompt was provided, send it as a message to the resumed task
if (prompt && controller.task) {
// Wait a moment for the task to fully load
await new Promise((resolve) => setTimeout(resolve, 100))
// Send the prompt as a response to any pending ask, or as a new message
await controller.task.handleWebviewAskResponse("messageResponse", prompt)
}
} else if (prompt) {
// Start a new task with the prompt
await controller.initTask(prompt, imageDataUrls)
emitTaskStarted()
} else {
throw new Error("Either taskId or prompt must be provided")
}
// Normal mode: wait for task completion
// Start the task
await controller.initTask(prompt, imageDataUrls)
const timeoutMs = (options.timeoutSeconds ?? 600) * 1000 // default 10 minutes
const timeoutPromise = new Promise((_, reject) => setTimeout(() => reject(new Error("Timeout")), timeoutMs))
await Promise.race([completionPromise, timeoutPromise])
+2 -21
View File
@@ -3,10 +3,7 @@
* Used by both UI components and CLI commands
*/
import { useMemo } from "react"
import { StateManager } from "@/core/storage/StateManager"
import providersData from "@/shared/providers/providers.json"
import type { RemoteConfigFields } from "@/shared/storage/state-keys"
// Create a lookup map from provider value to display label
const providerLabels: Record<string, string> = Object.fromEntries(
@@ -20,7 +17,7 @@ const providerOrder: string[] = providersData.list.map((p: { value: string }) =>
* Providers that are not supported in CLI.
* - vscode-lm: Requires VS Code's Language Model API (see ENG-1490 for OAuth-based support)
*/
const CLI_EXCLUDED_PROVIDERS = new Set<string>(["vscode-lm"])
export const CLI_EXCLUDED_PROVIDERS = new Set<string>(["vscode-lm"])
/**
* Get the display label for a provider ID
@@ -32,7 +29,7 @@ export function getProviderLabel(providerId: string): string {
/**
* Get the ordered list of all provider IDs (from providers.json)
*/
function getProviderOrder(): string[] {
export function getProviderOrder(): string[] {
return providerOrder
}
@@ -49,19 +46,3 @@ export function getValidCliProviders(): string[] {
export function isValidCliProvider(providerId: string): boolean {
return providerOrder.includes(providerId) && !CLI_EXCLUDED_PROVIDERS.has(providerId)
}
const getValidProviders = (remoteConfig: Partial<RemoteConfigFields> | undefined) => {
if (remoteConfig?.remoteConfiguredProviders?.length) {
return remoteConfig.remoteConfiguredProviders
}
return getProviderOrder().filter((p: string) => !CLI_EXCLUDED_PROVIDERS.has(p))
}
export const useValidProviders = () => {
const remoteConfig = StateManager.get().getRemoteConfigSettings()
return useMemo(() => {
return getValidProviders(remoteConfig)
}, [remoteConfig])
}
-8
View File
@@ -1,8 +0,0 @@
export function emitTaskStartedMessage(taskId: string, jsonOutput: boolean): void {
if (jsonOutput) {
process.stdout.write(JSON.stringify({ type: "task_started", taskId }) + "\n")
return
}
process.stderr.write(`Task started: ${taskId}\n`)
}
-36
View File
@@ -1,36 +0,0 @@
/**
* Wait for a condition to become truthy, with a timeout.
* Uses Promise.race for clean timeout handling instead of polling.
*
* @param condition - Function that returns the value to check (truthy = done)
* @param timeoutMs - Maximum time to wait in milliseconds
* @param pollIntervalMs - How often to check the condition (default: 100ms)
* @returns The truthy value if condition is met, or undefined if timeout
*/
export async function waitFor<T>(
condition: () => T | undefined | null,
timeoutMs: number,
pollIntervalMs: number = 100,
): Promise<T | undefined> {
// Check immediately first
const immediate = condition()
if (immediate) {
return immediate
}
return new Promise((resolve) => {
const intervalId = setInterval(() => {
const result = condition()
if (result) {
clearInterval(intervalId)
clearTimeout(timeoutId)
resolve(result)
}
}, pollIntervalMs)
const timeoutId = setTimeout(() => {
clearInterval(intervalId)
resolve(undefined)
}, timeoutMs)
})
}
+1 -7
View File
@@ -1,7 +1,6 @@
import { spawn } from "node:child_process"
import { realpathSync } from "node:fs"
import { exit } from "node:process"
import { ClineEndpoint } from "@/config"
import { fetch } from "@/shared/net"
import { printInfo, printWarning } from "./display"
@@ -108,7 +107,7 @@ async function getLatestVersion(currentVersion: string): Promise<string | null>
* process to install if a newer version is available.
*
* Supports npm, pnpm, yarn, and bun global installs.
* Skipped for npx, local dev, unknown installations, and bundled enterprise packages.
* Skipped for npx, local dev, and unknown installations.
* Can be disabled with CLINE_NO_AUTO_UPDATE=1 environment variable.
*/
export function autoUpdateOnStartup(currentVersion: string): void {
@@ -122,11 +121,6 @@ export function autoUpdateOnStartup(currentVersion: string): void {
return
}
// Skip if using bundled enterprise config (single source of truth)
if (ClineEndpoint.isBundledConfig()) {
return
}
const { updateCommand } = getInstallationInfo(currentVersion)
if (!updateCommand) {
return
+2 -7
View File
@@ -4,7 +4,6 @@
*/
import { mkdirSync } from "node:fs"
import { fileURLToPath } from "node:url"
import os from "os"
import path from "path"
import { ExtensionRegistryInfo } from "@/registry"
@@ -12,10 +11,6 @@ import { ClineExtensionContext } from "@/shared/cline"
import { ClineFileStorage } from "@/shared/storage"
import { EnvironmentVariableCollection, ExtensionKind, ExtensionMode, readJson, URI } from "./vscode-shim"
// ES module equivalent of __dirname
const __filename = fileURLToPath(import.meta.url)
const __dirname = path.dirname(__filename)
const SETTINGS_SUBFOLDER = "data"
/**
@@ -145,8 +140,8 @@ export function initializeCliContext(config: CliContextConfig = {}): CliContextR
mkdirSync(DATA_DIR, { recursive: true })
mkdirSync(WORKSPACE_STORAGE_DIR, { recursive: true })
// For CLI, extension dir is the package root (one level up from dist/)
const EXTENSION_DIR = path.resolve(__dirname, "..")
// For CLI, extension dir is the root of the project (parent of cli)
const EXTENSION_DIR = path.resolve(__dirname, "..", "..")
const EXTENSION_MODE = process.env.IS_DEV === "true" ? ExtensionMode.Development : ExtensionMode.Production
const extension: ClineExtensionContext["extension"] = {
-1
View File
@@ -13,7 +13,6 @@ export default defineConfig({
},
resolve: {
alias: {
vscode: path.resolve(__dirname, "src/vscode-shim.ts"),
// Match tsconfig paths - baseUrl is parent directory
"@": path.resolve(__dirname, "../src"),
"@api": path.resolve(__dirname, "../src/core/api"),
+4 -4
View File
@@ -202,15 +202,15 @@ If Cline can't access files or run commands:
Learn about Cline CLI's core capabilities and use cases.
</Card>
<Card title="Headless Mode" icon="robot" href="/cline-cli/three-core-flows">
Run Cline autonomously in scripts, CI/CD pipelines, and automated workflows.
<Card title="Three Core Flows" icon="route" href="/cline-cli/three-core-flows">
Master interactive mode, headless automation, and multi-instance workflows.
</Card>
<Card title="Skills" icon="graduation-cap" href="/customization/skills">
<Card title="Skills" icon="graduation-cap" href="/features/skills">
Understand how Cline's Skills work across all editors via ACP.
</Card>
<Card title="Hooks" icon="link" href="/customization/hooks">
<Card title="Hooks" icon="link" href="/features/hooks/index">
Learn how to enforce policies with Hooks in any editor.
</Card>
</Columns>
+482
View File
@@ -0,0 +1,482 @@
---
title: "CLI Reference (Deprecated)"
description: "Command reference for Cline CLI versions earlier than 2.0.0 (deprecated). For the latest commands and options, see the current Cline CLI reference."
---
Complete command reference for Cline CLI. Use this for detailed documentation on all commands, options, and configuration.
For quick help in your terminal:
```bash
cline --help # Show all commands
cline task --help # Show task-specific commands
man cline # View the full manual page
```
## Manual Page
The complete manual page for the Cline CLI:
```
CLINE(1) User Commands CLINE(1)
NAME
cline - orchestrate and interact with Cline AI coding agents
SYNOPSIS
cline [prompt] [options]
cline command [subcommand] [options] [arguments]
DESCRIPTION
Try: cat README.md | cline "Summarize this for me:"
cline is a command-line interface for orchestrating multiple Cline AI
coding agents. Cline is an autonomous AI agent who can read, write,
and execute code across your projects. He operates through a
client-server architecture where Cline Core runs as a standalone
service, and the CLI acts as a scriptable interface for managing tasks,
instances, and agent interactions.
The CLI is designed for both interactive use and automation, making it
ideal for CI/CD pipelines, parallel task execution, and terminal-based
workflows. Multiple frontends (CLI, VSCode, JetBrains) can attach to
the same Cline Core instance, enabling seamless task handoff between
environments.
MODES OF OPERATION
Instant Task Mode
The simplest invocation: cline "prompt here" immediately spawns
an instance, creates a task, and enters chat mode. This is
equivalent to running cline instance new && cline task new &&
cline task chat in sequence.
Subcommand Mode
Advanced usage with explicit control: cline <command>
[subcommand] [options] provides fine-grained control over
instances, tasks, authentication, and configuration.
AGENT BEHAVIOR
Cline operates in two primary modes:
ACT MODE
Cline actively uses tools to accomplish tasks. He can read
files, write code, execute commands, use a headless browser, and
more. This is the default mode for task execution.
PLAN MODE
Cline gathers information and creates a detailed plan before
implementation. He explores the codebase, asks clarifying
questions, and presents a strategy for user approval before
switching to ACT MODE.
INSTANT TASK OPTIONS
When using the instant task syntax cline "prompt" the following options
are available:
-o, --oneshot
Full autonomous mode. Cline completes the task and stops
following after completion. Example: cline -o "what's 6 + 8?"
-s, --setting setting value
Override a setting for this task
-y, --no-interactive, --yolo
Enable fully autonomous mode. Disables all interactivity:
• ask_followup_question tool is disabled
• attempt_completion happens automatically
• execute_command runs in non-blocking mode with timeout
• PLAN MODE automatically switches to ACT MODE
-m, --mode mode
Starting mode. Options: act (default), plan
-w, --workspace path
Additional workspace paths. Can be specified multiple times to
include multiple directories. The current working directory is
always included as the first workspace. Example: cline -w
/path/to/other/project "refactor shared code"
GLOBAL OPTIONS
These options apply to all subcommands:
-F, --output-format format
Output format. Options: rich (default), json, plain
-h, --help
Display help information for the command.
-v, --verbose
Enable verbose output for debugging.
COMMANDS
Authentication
cline auth [provider] [key]
cline a [provider] [key]
Configure authentication for AI model providers. Launches an
interactive wizard if no arguments provided. If provider is
specified without a key, prompts for the key or launches the
appropriate OAuth flow.
Instance Management
Cline Core instances are independent agent processes that can run in
the background. Multiple instances can run simultaneously, enabling
parallel task execution.
cline instance
cline i
Display instance management help.
cline instance new [-d|--default]
cline i n [-d|--default]
Spawn a new Cline Core instance. Use --default to set it as
the default instance for subsequent commands.
cline instance list
cline i l
List all running Cline Core instances with their addresses and
status.
cline instance default address
cline i d address
Set the default instance to avoid specifying --address in task
commands.
cline instance kill address [-a|--all]
cline i k address [-a|--all]
Terminate a Cline Core instance. Use --all to kill all running
instances.
Task Management
Tasks represent individual work items that Cline executes. Tasks
maintain conversation history, checkpoints, and settings.
cline task [-a|--address ADDR]
cline t [-a|--address ADDR]
Display task management help. The --address flag specifies
which Cline Core instance to use (e.g., localhost:50052).
cline task new prompt [options]
cline t n prompt [options]
Create a new task in the default or specified instance.
Options:
-s, --setting setting value
Set task-specific settings
-y, --no-interactive, --yolo
Enable autonomous mode
-m, --mode mode
Starting mode (act or plan)
cline task open task-id [options]
cline t o task-id [options]
Resume a previous task from history. Accepts the same options
as task new.
cline task list
cline t l
List all tasks in history with their id and snippet
cline task chat
cline t c
Enter interactive chat mode for the current task. Allows
back-and-forth conversation with Cline.
cline task send [message] [options]
cline t s [message] [options]
Send a message to Cline. If no message is provided, reads from
stdin. Options:
-a, --approve
Approve Cline's proposed action
-d, --deny
Deny Cline's proposed action
-f, --file FILE
Attach a file to the message
-y, --no-interactive, --yolo
Enable autonomous mode
-m, --mode mode
Switch mode (act or plan)
cline task view [-f|--follow] [-c|--follow-complete]
cline t v [-f|--follow] [-c|--follow-complete]
Display the current conversation. Use --follow to stream
updates in real-time, or --follow-complete to follow until task
completion.
cline task restore checkpoint
cline t r checkpoint
Restore the task to a previous checkpoint state.
cline task pause
cline t p
Pause task execution.
Configuration
Configuration can be set globally. Override these global settings for
a task using the --setting flag
cline config
cline c
cline config set key value
cline c s key value
Set a configuration variable.
cline config get key
cline c g key
Read a configuration variable.
cline config list
cline c l
List all configuration variables and their values.
Context Window Configuration
For local model providers, you can configure the context window size:
Ollama
cline config s ollama-api-options-ctx-num=32768
LM Studio
cline config s lm-studio-max-tokens=32768
For other providers (Anthropic, OpenRouter, etc.), the context window
is defined per model in the model metadata and is not user-settable.
Cline uses each model's built-in context limits automatically.
TASK SETTINGS
Task settings are persisted in the ~/.cline/x/tasks directory. When
resuming a task with cline task open, task settings are automatically
restored.
Common settings include:
yolo Enable autonomous mode (true/false)
mode Starting mode (act/plan)
hooks_enabled
Enable or disable hooks for the task (true/false)
HOOKS INTEGRATION
Hooks let you inject custom logic into Cline's workflow at key moments.
They can validate operations before they execute, monitor tool usage,
and shape AI decisions. This allows you to integrate hooks into
automated workflows, CI/CD pipelines, and headless task execution.
Enable hooks for a task:
cline "prompt" -s hooks_enabled=true
Configure hooks globally:
cline config set hooks-enabled=true
cline config get hooks-enabled
Note: Hooks in the CLI are only supported on macOS and Linux.
For complete hooks documentation, see:
<https://docs.cline.bot/features/hooks/index>
NOTES & EXAMPLES
The cline task send and cline task new commands support reading from
stdin, enabling powerful pipeline compositions:
cat requirements.txt | cline task send
echo "Refactor this code" | cline -y
Instance Management
Manage multiple Cline instances:
# Start a new instance and make it default
cline instance new --default
# List all running instances
cline instance list
# Kill a specific instance
cline instance kill localhost:50052
# Kill all CLI instances
cline instance kill --all-cli
Task History
Work with task history:
# List previous tasks
cline task list
# Resume a previous task
cline task open 1760501486669
# View conversation history
cline task view
# Start interactive chat with this task
cline task chat
ARCHITECTURE
Cline operates on a three-layer architecture:
Presentation Layer
User interfaces (CLI, VSCode, JetBrains) that connect to Cline
Core via gRPC
Cline Core
The autonomous agent service handling task management, AI model
integration, state management, tool orchestration, and real-time
streaming updates
Host Provider Layer
Environment-specific integrations (VSCode APIs, JetBrains APIs,
shell APIs) that Cline Core uses to interact with the host
system
BUGS
Report bugs at: <https://github.com/cline/cline/issues>
For real-time help, join the Discord community at:
<https://discord.gg/cline>
SEE ALSO
Full documentation: <https://docs.cline.bot>
AUTHORS
Cline is developed by the Cline Bot Inc. and the open source community.
COPYRIGHT
Copyright © 2025 Cline Bot Inc. Licensed under the Apache License 2.0.
```
## JSON output (-F json)
When you run a command with `-F json` (or `--output-format json`), Cline prints each client message as JSON.
### ClineMessage schema
| Field | Type | Required | Notes |
|-------|------|----------|-------|
| `type` | `"ask" or "say"` | Yes | Top-level message category. |
| `text` | `string` | Yes | Human-readable message content. |
| `ts` | `number` | Yes | Unix epoch timestamp in milliseconds. |
| `reasoning` | `string` | No | Omitted when empty. |
| `say` | `string` | No | Omitted when empty. Present when `type` is `"say"`. |
| `ask` | `string` | No | Omitted when empty. Present when `type` is `"ask"`. |
| `partial` | `boolean` | No | Omitted when false. `true` for streaming updates. |
| `images` | `string[]` | No | Omitted when empty. Image URIs when included with a message. |
| `files` | `string[]` | No | Omitted when empty. File paths when attached to a message. |
| `lastCheckpointHash` | `string` | No | Omitted when empty. Git checkpoint hash when available. |
| `isCheckpointCheckedOut` | `boolean` | No | Omitted when false. `true` if Cline checked out a checkpoint. |
| `isOperationOutsideWorkspace` | `boolean` | No | Omitted when false. `true` if an operation happened outside the workspace. |
<Note>
Most fields are optional and omitted when empty. If you parse this output, treat missing fields as “not present”, not as empty strings.
</Note>
### Example
```json
{
"type": "say",
"text": "Cline is about to run a command.",
"ts": 1760501486669,
"say": "command",
"partial": false
}
```
### Shell Completion
Generate autocompletion scripts for various shells:
#### Bash
```bash
# Generate bash completion
cline completion bash > /etc/bash_completion.d/cline
# Or for user-level installation
cline completion bash > ~/.local/share/bash-completion/completions/cline
```
#### Zsh
```bash
# Generate zsh completion
cline completion zsh > "${fpath[1]}/_cline"
# Or add to your .zshrc
echo 'source <(cline completion zsh)' >> ~/.zshrc
```
#### Fish
```bash
# Generate fish completion
cline completion fish > ~/.config/fish/completions/cline.fish
```
#### PowerShell
```powershell
# Generate PowerShell completion
cline completion powershell > cline.ps1
# Add to your PowerShell profile
Add-Content $PROFILE "cline completion powershell | Out-String | Invoke-Expression"
```
### Version Command
```bash
# Show version information
cline version
```
### Environment Variables
#### CLINE_DIR
Override the default Cline directory location:
```bash
# Override default Cline directory
export CLINE_DIR=/custom/path
# Default: ~/.cline
```
This directory is used for:
- Instance registry database
- Configuration files
- Task history
- Checkpoints
+5 -3
View File
@@ -3,6 +3,8 @@ title: "CLI Reference"
description: "Complete command reference for Cline CLI including all commands, flags, and configuration options"
---
# CLI Reference
This page documents all available commands, flags, and configuration options for Cline CLI. For quick help in your terminal, use:
```bash
@@ -314,7 +316,7 @@ When using `--json`, each message is output as a JSON object (one per line):
Cline stores all data in `~/.cline/` by default:
```text
```
~/.cline/
├── data/ # Configuration directory
│ ├── globalState.json # Global settings
@@ -414,8 +416,8 @@ cline auth -p openai -k your-key -b https://api.example.com/v1
Keyboard shortcuts, slash commands, and file mentions.
</Card>
<Card title="Headless Mode" icon="robot" href="/cline-cli/three-core-flows">
Run Cline autonomously in scripts, CI/CD pipelines, and automated workflows.
<Card title="CLI Workflows" icon="route" href="/cline-cli/three-core-flows">
Interactive mode, direct execution, and automation patterns.
</Card>
<Card title="Configuration" icon="gear" href="/cline-cli/configuration">
+6 -6
View File
@@ -38,7 +38,7 @@ Rules help Cline understand your project's conventions, coding standards, and pr
### Workflows Tab
View and manage [workflows](/customization/workflows):
View and manage [workflows](/features/slash-commands/workflows/index):
- List available workflows
- View workflow definitions
@@ -46,7 +46,7 @@ View and manage [workflows](/customization/workflows):
### Hooks Tab
Configure [hooks](/customization/hooks) for custom logic integration:
Configure [hooks](/features/hooks/index) for custom logic integration:
- Enable/disable hooks globally
- View configured hook scripts
@@ -58,7 +58,7 @@ Hooks must be enabled via settings. Use `cline config` to toggle `hooks-enabled`
### Skills Tab
Manage [skills](/customization/skills) that extend Cline's capabilities:
Manage [skills](/features/skills) that extend Cline's capabilities:
- View available skills
- Enable/disable specific skills
@@ -68,7 +68,7 @@ Manage [skills](/customization/skills) that extend Cline's capabilities:
Cline stores configuration in `~/.cline/data/`:
```text
```
~/.cline/
├── data/ # Configuration directory
│ ├── globalState.json # Global settings
@@ -268,8 +268,8 @@ cline auth # Re-authenticate
## Next Steps
<Columns cols={2}>
<Card title="Headless Mode" icon="robot" href="/cline-cli/three-core-flows">
Run Cline autonomously in scripts, CI/CD pipelines, and automated workflows.
<Card title="CLI Workflows" icon="route" href="/cline-cli/three-core-flows">
Learn about interactive mode, direct execution, and automation patterns.
</Card>
<Card title="CLI Reference" icon="terminal" href="/cline-cli/cli-reference">
-457
View File
@@ -1,457 +0,0 @@
---
title: "Getting Started"
description: "Run Cline AI coding agents directly in your terminal with an interactive CLI or automated workflows"
---
## What is Cline CLI?
Cline CLI brings the full power of Cline to your terminal. Whether you prefer an interactive experience or automated workflows for CI/CD pipelines, the CLI adapts to your needs.
The CLI supports macOS, Linux, and Windows, and works with all the same AI providers as the VS Code extension.
## Two Ways to Use Cline CLI
The CLI operates in two distinct modes, automatically selecting the appropriate one based on how you invoke it:
### Interactive Mode
Interactive mode is designed for **hands-on development sessions** where you want to collaborate with Cline in real-time. It provides a rich terminal interface that feels like chatting with an AI assistant.
**When it activates:** Running `cline` without arguments, or when stdin is a TTY (terminal).
```bash
cline
```
Key features:
- **Real-time conversation** - Type messages, see Cline's responses, and iterate on tasks
- **Visual feedback** - Animated welcome screen, syntax-highlighted code, and progress indicators
- **File mentions** with `@` - Reference workspace files with fuzzy search autocomplete
- **Slash commands** with `/` - Quick access to `/settings`, `/history`, `/models`, and workflows
- **Keyboard shortcuts** - `Tab` to toggle Plan/Act, `Shift+Tab` for auto-approve all
- **Session summaries** - See tasks completed, files modified, and token usage on exit
- **Settings panel** - Configure providers, models, and features without leaving the CLI
Interactive mode keeps you in control. You review Cline's plan, approve or modify actions, and guide the conversation.
[Learn more about interactive mode →](/cline-cli/interactive-mode)
### Headless Mode (Non-Interactive)
Headless mode is designed for **automation, scripting, and CI/CD pipelines** where human interaction isn't possible or desired.
**When it activates:** Using the `-y`/`--yolo` flag, `--json` flag, piping input/output, or when stdin is not a TTY.
```bash
# Headless with auto-approval (YOLO mode)
cline -y "Run tests and fix any failures"
# Headless with JSON output for parsing
cline --json "List all TODO comments" | jq '.text'
# Headless via piped input
cat README.md | cline "Summarize this document"
# Chain multiple headless commands
git diff | cline -y "explain these changes" | cline -y "write a commit message"
```
Key features:
- **No visual interface** - Clean text or JSON output suitable for scripting
- **Automatic execution** - With `-y`, Cline approves all actions and runs autonomously
- **Process control** - Exits automatically when the task completes
- **Piped workflows** - Read from stdin, write to stdout, chain with other commands
- **Machine-readable output** - Use `--json` to get structured output for parsing
<Warning>
Headless mode with `-y` gives Cline full autonomy. Run on a clean git branch so you can easily revert changes if needed.
</Warning>
### Mode Detection Summary
Cline automatically detects which mode to use based on your invocation. This table shows how different command patterns trigger each mode, helping you predict behavior in scripts and interactive sessions.
| Invocation | Mode | Reason |
|------------|------|--------|
| `cline` | Interactive | No arguments, TTY connected |
| `cline "task"` | Interactive | TTY connected |
| `cline -y "task"` | Headless | YOLO flag forces headless |
| `cline --json "task"` | Headless | JSON flag forces headless |
| `cat file \| cline "task"` | Headless | stdin is piped |
| `cline "task" > output.txt` | Headless | stdout is redirected |
[Learn more about headless mode →](/cline-cli/three-core-flows)
## Supported Model Providers
Cline CLI supports all providers available in the VS Code extension:
- **Anthropic** (Claude)
- **OpenAI** (GPT-4o, GPT-4)
- **OpenAI Codex** (ChatGPT subscription)
- **OpenRouter**
- **AWS Bedrock**
- **Google Gemini**
- **X AI (Grok)**
- **Cerebras**
- **DeepSeek**
- **Ollama** (local models)
- **LM Studio** (local models)
- **OpenAI Compatible** (any compatible API)
During setup, authenticate with `cline auth` to configure your preferred provider. [See authentication →](#authenticate)
## What You Can Build
### Automated Code Maintenance
Keep your codebase healthy with automated fixes. Cline scans for issues and applies corrections across multiple files.
```bash
cline -y "Fix all ESLint errors in src/"
```
Finds and fixes linting violations throughout your source directory.
```bash
cline -y "Update all deprecated React lifecycle methods"
```
Migrates legacy code patterns to modern equivalents (e.g., `componentWillMount` → `useEffect`).
```bash
cline -y "Update dependencies with known vulnerabilities"
```
Identifies outdated packages with security issues and updates them to safe versions.
### CI/CD Integration
Integrate Cline into your continuous integration pipelines for automated code review and documentation.
```bash
git diff origin/main | cline -y "Review these changes for issues"
```
Pipes your PR diff to Cline for automated code review, catching bugs and style issues before merge.
```bash
git log --oneline v1.0..v1.1 | cline -y "Write release notes"
```
Generates human-readable release notes from your commit history between two tags.
```bash
cline -y "Run tests and fix failures" --timeout 600
```
Executes your test suite, analyzes failures, and attempts fixes with a 10-minute timeout.
### Development Workflows
From quick edits to complex refactors, Cline adapts to your workflow.
```bash
cline
```
Launches interactive mode for exploratory development and back-and-forth collaboration.
```bash
cline "Refactor this function to use async/await"
```
Executes a focused task directly from the command line with approval prompts at key steps.
```bash
cline "Based on @src/api.ts, add error handling to all endpoints"
```
Uses file mentions (`@`) to give Cline context about specific files in your workspace.
### Custom Shell Pipelines
Chain Cline with other CLI tools to build powerful automation workflows.
```bash
gh pr diff 123 | cline -y "Review this PR"
```
Fetches a GitHub PR diff and pipes it directly to Cline for review.
```bash
cline --json "List all TODO comments" | jq '.text'
```
Outputs structured JSON that you can process with tools like `jq` for scripting.
```bash
git diff | cline -y "explain" | cline -y "write a haiku about these changes"
```
Chains multiple Cline invocations together for creative multi-step workflows.
## Features at a Glance
| Feature | Interactive Mode | Non-Interactive Mode |
|---------|------------------|----------------------|
| Interactive chat | ✓ | - |
| File mentions (@) | ✓ | ✓ (inline) |
| Slash commands (/) | ✓ | - |
| Settings panel | ✓ | `cline config` |
| Plan/Act toggle | ✓ (Tab) | `-p` / `-a` flags |
| Auto-approve | ✓ (Shift+Tab) | `-y` flag |
| Session summary | ✓ | - |
| JSON output | - | `--json` |
| Piped input | - | ✓ |
---
## Installation & Setup
In just a few minutes, you can install the CLI, authenticate with your preferred AI provider, and start running tasks from any directory on your machine.
### Prerequisites
Cline CLI requires **Node.js version 20 or higher**. We recommend Node.js 22 for the best experience.
Check your Node.js version:
```bash
node --version
```
If you need to install or update Node.js, visit [nodejs.org](https://nodejs.org) or use a version manager like [nvm](https://github.com/nvm-sh/nvm).
### Install Cline CLI
Install globally via npm:
```bash
npm install -g cline
```
Verify the installation:
```bash
cline version
```
<Tip>
To install a specific version, use `npm install -g cline@2.0.0`. Check [npm](https://www.npmjs.com/package/cline) for available versions.
</Tip>
### Authenticate
After installation, run the authentication wizard:
```bash
cline auth
```
This launches an interactive wizard with multiple options. Choose the method that works best for your workflow.
#### Option 1: Sign in with Cline (Recommended)
Select **"Sign in with Cline"** to authenticate with your Cline account via OAuth. Your browser opens automatically to complete sign-in.
#### Option 2: Sign in with ChatGPT Subscription
If you have a ChatGPT Plus or Pro subscription, select **"Sign in with ChatGPT Subscription"**. This uses OpenAI's Codex OAuth to authenticate with your existing subscription.
#### Option 3: Import from Existing Tools
Already using another AI coding CLI? Cline can import your existing configuration:
- **Import from Codex CLI** - Imports credentials from `~/.codex/auth.json`
- **Import from OpenCode** - Imports configuration from `~/.local/share/opencode/auth.json`
#### Option 4: Bring Your Own API Key
Select **"Bring your own API key"** to manually configure any supported provider. Or skip the wizard entirely with flags:
```bash
# Anthropic (Claude)
cline auth -p anthropic -k sk-ant-api-xxxxx -m claude-sonnet-4-5-20250929
# OpenAI
cline auth -p openai-native -k sk-xxxxx -m gpt-4o
# OpenRouter
cline auth -p openrouter -k sk-or-xxxxx -m anthropic/claude-sonnet-4-5-20250929
# OpenAI-compatible provider with custom base URL
cline auth -p openai -k your-api-key -b https://api.example.com/v1
```
**Quick Setup Flags:**
| Flag | Description |
|------|-------------|
| `-p, --provider <id>` | Provider ID (e.g., `anthropic`, `openai-native`, `openrouter`) |
| `-k, --apikey <key>` | Your API key |
| `-m, --modelid <id>` | Model ID (e.g., `claude-sonnet-4-5-20250929`, `gpt-4o`) |
| `-b, --baseurl <url>` | Base URL for OpenAI-compatible providers |
<Tip>
Flags are especially useful for scripting, CI/CD environments, or setting up multiple machines.
</Tip>
#### Supported Providers
| Provider | Provider ID | Notes |
|----------|-------------|-------|
| Anthropic | `anthropic` | Direct Claude API access |
| OpenAI | `openai-native` | GPT-4o, GPT-4, etc. |
| OpenAI Codex | `openai-codex` | ChatGPT subscription OAuth |
| OpenRouter | `openrouter` | Access multiple providers |
| AWS Bedrock | `bedrock` | Claude via AWS |
| Google Gemini | `gemini` | Gemini Pro, etc. |
| X AI (Grok) | `xai` | Grok models |
| Cerebras | `cerebras` | Fast inference |
| DeepSeek | `deepseek` | DeepSeek models |
| Ollama | `ollama` | Local models |
| LM Studio | `lmstudio` | Local models |
| OpenAI Compatible | `openai` | Any OpenAI-compatible API |
### Verify Your Setup
Confirm everything is working with a simple test:
```bash
cline "What is 2 + 2?"
```
If Cline responds with an answer, your installation and authentication are complete.
Check your current configuration:
```bash
cline config
```
### Quick Start
Now you're ready to use Cline. Choose how you want to work:
#### Interactive Mode
Launch the interactive CLI for development:
```bash
cline
```
You'll see the Cline welcome screen. Type your task and press Enter. Use:
- `Tab` to toggle between Plan and Act modes
- `Shift+Tab` to enable auto-approve
- `/help` for available commands
[Learn more about interactive mode →](/cline-cli/interactive-mode)
#### Direct Task Execution
Run a task directly from your shell:
```bash
cline "Add error handling to utils.js"
```
For non-interactive execution (perfect for scripts and CI/CD):
```bash
cline -y "Run tests and fix any failures"
```
[Learn more about headless mode →](/cline-cli/three-core-flows)
### Switching Providers
To change your configured provider at any time:
```bash
cline auth
```
You can also use the settings panel in interactive mode:
```bash
cline
# Then type: /settings
# Navigate to the API tab
```
### Updating
Check for updates and install the latest version:
```bash
cline update
```
Or update manually via npm:
```bash
npm update -g cline
```
### Troubleshooting
#### Command Not Found
If `cline` is not found after installation:
1. Ensure npm global bin is in your PATH:
```bash
npm bin -g
```
2. Add the path to your shell configuration (`.bashrc`, `.zshrc`, etc.):
```bash
export PATH="$PATH:$(npm bin -g)"
```
3. Restart your terminal or source your shell config.
#### Permission Errors
If you get permission errors during installation:
```bash
# Option 1: Use a Node version manager (recommended)
# nvm, fnm, or volta handle permissions automatically
# Option 2: Fix npm permissions
# See: https://docs.npmjs.com/resolving-eacces-permissions-errors-when-installing-packages-globally
```
#### OAuth Flow Issues
If the browser doesn't open automatically during OAuth:
1. Copy the URL from the terminal
2. Paste it in your browser manually
3. Complete the sign-in flow
4. Return to the terminal
#### API Key Validation
If your API key is rejected:
1. Verify the key is correct and hasn't expired
2. Check that you've selected the correct provider
3. Ensure your API account has the necessary permissions
**Provider-specific tips:**
- **Anthropic**: Keys start with `sk-ant-`
- **OpenAI**: Keys start with `sk-`
- **AWS Bedrock**: Requires AWS credentials configured separately. See [AWS Bedrock documentation](/provider-config/aws-bedrock/api-key).
### Uninstallation
To remove Cline CLI:
```bash
npm uninstall -g cline
```
To also remove configuration data:
```bash
rm -rf ~/.cline
```
## Next Steps
- **[Interactive Mode](/cline-cli/interactive-mode)** - Master the interactive CLI with shortcuts and slash commands
- **[Headless Mode](/cline-cli/three-core-flows)** - Run Cline autonomously in scripts, CI/CD pipelines, and automated workflows
- **[Configuration](/cline-cli/configuration)** - Configure settings, rules, workflows, and environment variables
- **[CLI Reference](/cline-cli/cli-reference)** - Complete command documentation with all flags and options
+12 -16
View File
@@ -22,7 +22,7 @@ If you need to install or update Node.js, visit [nodejs.org](https://nodejs.org)
Install globally via npm:
```bash
npm install -g cline
bun install -g cline
```
Verify the installation:
@@ -32,7 +32,7 @@ cline version
```
<Tip>
To install a specific version, use `npm install -g cline@2.0.0`. Check [npm](https://www.npmjs.com/package/cline) for available versions.
To install a specific version, use `bun install -g cline@2.0.0`. Check [npm](https://www.npmjs.com/package/cline) for available versions.
</Tip>
## Authenticate
@@ -74,9 +74,6 @@ cline auth -p openai-native -k sk-xxxxx -m gpt-4o
# OpenRouter
cline auth -p openrouter -k sk-or-xxxxx -m anthropic/claude-sonnet-4-5-20250929
# Moonshot
cline auth -p moonshot -k sk-xxxxx -m kimi-k2.5
# OpenAI-compatible provider with custom base URL
cline auth -p openai -k your-api-key -b https://api.example.com/v1
```
@@ -85,7 +82,7 @@ cline auth -p openai -k your-api-key -b https://api.example.com/v1
| Flag | Description |
|------|-------------|
| `-p, --provider <id>` | Provider ID (e.g., `anthropic`, `openai-native`, `openrouter`, `moonshot`) |
| `-p, --provider <id>` | Provider ID (e.g., `anthropic`, `openai-native`, `openrouter`) |
| `-k, --apikey <key>` | Your API key |
| `-m, --modelid <id>` | Model ID (e.g., `claude-sonnet-4-5-20250929`, `gpt-4o`) |
| `-b, --baseurl <url>` | Base URL for OpenAI-compatible providers |
@@ -107,7 +104,6 @@ Flags are especially useful for scripting, CI/CD environments, or setting up mul
| X AI (Grok) | `xai` | Grok models |
| Cerebras | `cerebras` | Fast inference |
| DeepSeek | `deepseek` | DeepSeek models |
| Moonshot | `moonshot` | Kimi models via Moonshot AI |
| Ollama | `ollama` | Local models |
| LM Studio | `lmstudio` | Local models |
| OpenAI Compatible | `openai` | Any OpenAI-compatible API |
@@ -161,7 +157,7 @@ For non-interactive execution (perfect for scripts and CI/CD):
cline -y "Run tests and fix any failures"
```
[Learn more about headless mode →](/cline-cli/three-core-flows)
[Learn more about CLI workflows →](/cline-cli/three-core-flows)
## Switching Providers
@@ -190,7 +186,7 @@ cline update
Or update manually via npm:
```bash
npm update -g cline
bun update -g cline
```
## Troubleshooting
@@ -199,14 +195,14 @@ npm update -g cline
If `cline` is not found after installation:
1. Ensure npm global bin is in your PATH:
1. Ensure bun global bin is in your PATH:
```bash
npm bin -g
bun bin -g
```
2. Add the path to your shell configuration (`.bashrc`, `.zshrc`, etc.):
```bash
export PATH="$PATH:$(npm bin -g)"
export PATH="$PATH:$(bun bin -g)"
```
3. Restart your terminal or source your shell config.
@@ -219,7 +215,7 @@ If you get permission errors during installation:
# Option 1: Use a Node version manager (recommended)
# nvm, fnm, or volta handle permissions automatically
# Option 2: Fix npm permissions
# Option 2: Fix bun permissions
# See: https://docs.npmjs.com/resolving-eacces-permissions-errors-when-installing-packages-globally
```
@@ -248,7 +244,7 @@ If your API key is rejected:
To remove Cline CLI:
```bash
npm uninstall -g cline
bun uninstall -g cline
```
To also remove configuration data:
@@ -264,8 +260,8 @@ rm -rf ~/.cline
Master the interactive CLI with shortcuts and slash commands.
</Card>
<Card title="Headless Mode" icon="robot" href="/cline-cli/three-core-flows">
Run Cline autonomously in scripts, CI/CD pipelines, and automated workflows.
<Card title="CLI Workflows" icon="route" href="/cline-cli/three-core-flows">
Learn interactive mode, direct execution, and automation patterns.
</Card>
<Card title="Configuration" icon="gear" href="/cline-cli/configuration">
+10 -10
View File
@@ -65,7 +65,7 @@ Keyboard shortcuts are the primary way to navigate and control the interactive C
Reference files from your workspace by typing `@` followed by the filename:
```text
```
@src/utils.ts can you add error handling to this file?
```
@@ -79,7 +79,7 @@ File search uses ripgrep for fast, fuzzy matching. You can type partial paths li
Include multiple files in a single message:
```text
```
Compare @src/old-api.ts with @src/new-api.ts and list the breaking changes
```
@@ -100,9 +100,9 @@ Type `/` to see available commands. Slash commands provide quick access to setti
### Workflow Commands
If you have [workflows](/customization/workflows) configured, they appear as additional slash commands. For example, if you have a workflow named `code-review`, you can invoke it with:
If you have [workflows](/features/slash-commands/workflows/index) configured, they appear as additional slash commands. For example, if you have a workflow named `code-review`, you can invoke it with:
```text
```
/code-review
```
@@ -120,7 +120,7 @@ Access the settings panel with `/settings`. Navigate between tabs using arrow ke
## Plan and Act Modes
Cline operates in two modes, toggled with `Tab`. These modes work the same way in the CLI as they do in the VS Code extension. For a deeper explanation of how Plan and Act modes work, see the [Plan and Act documentation](/core-workflows/plan-and-act).
Cline operates in two modes, toggled with `Tab`. These modes work the same way in the CLI as they do in the VS Code extension. For a deeper explanation of how Plan and Act modes work, see the [Plan and Act documentation](/features/plan-and-act).
### Plan Mode
@@ -211,7 +211,7 @@ Use terminal multiplexers like tmux or split terminals to run multiple Cline ins
Give Cline context about what you're working on:
```text
```
I'm building a REST API with Express. The routes are in @src/routes/ and models in @src/models/. Help me add user authentication.
```
@@ -219,7 +219,7 @@ I'm building a REST API with Express. The routes are in @src/routes/ and models
When you're unsure about the best approach:
```text
```
[Tab to Plan mode]
How should I structure the database schema for a multi-tenant SaaS app?
```
@@ -228,7 +228,7 @@ How should I structure the database schema for a multi-tenant SaaS app?
The interactive CLI maintains conversation context. Build on previous messages:
```text
```
> Add a login endpoint
[Cline creates the endpoint]
@@ -242,8 +242,8 @@ The interactive CLI maintains conversation context. Build on previous messages:
## Next Steps
<Columns cols={2}>
<Card title="Headless Mode" icon="robot" href="/cline-cli/three-core-flows">
Run Cline autonomously in scripts, CI/CD pipelines, and automated workflows.
<Card title="CLI Workflows" icon="route" href="/cline-cli/three-core-flows">
Learn about interactive mode, direct execution, and automation patterns.
</Card>
<Card title="Configuration" icon="gear" href="/cline-cli/configuration">
+5 -14
View File
@@ -15,15 +15,6 @@ Ready to get started? Check out the [installation guide](/cline-cli/installation
## Two Ways to Use Cline CLI
<Columns cols={2}>
<Card title="Interactive Mode" icon="terminal" href="/cline-cli/interactive-mode">
**For hands-on development.** Launch `cline` in your terminal and collaborate with Cline in real-time — chat, review plans, approve actions, and iterate on tasks with a rich visual interface.
</Card>
<Card title="Headless Mode" icon="robot" href="/cline-cli/three-core-flows">
**For automation & CI/CD.** Run `cline -y "task"` to let Cline work autonomously — no interaction needed. Pipe input/output, get JSON results, and chain commands in scripts and pipelines.
</Card>
</Columns>
The CLI operates in two distinct modes, automatically selecting the appropriate one based on how you invoke it:
### Interactive Mode
@@ -95,7 +86,7 @@ Cline automatically detects which mode to use based on your invocation. This tab
| `cat file \| cline "task"` | Headless | stdin is piped |
| `cline "task" > output.txt` | Headless | stdout is redirected |
[Learn more about headless mode →](/cline-cli/three-core-flows)
[Learn more about CLI workflows →](/cline-cli/three-core-flows)
## Supported Model Providers
@@ -219,8 +210,8 @@ Chains multiple Cline invocations together for creative multi-step workflows.
Master the interactive CLI with keyboard shortcuts and slash commands.
</Card>
<Card title="Headless Mode" icon="robot" href="/cline-cli/three-core-flows">
Run Cline autonomously in scripts, CI/CD pipelines, and automated workflows.
<Card title="CLI Workflows" icon="route" href="/cline-cli/three-core-flows">
Learn interactive mode, direct execution, and automation patterns.
</Card>
<Card title="Configuration" icon="gear" href="/cline-cli/configuration">
@@ -231,7 +222,7 @@ Chains multiple Cline invocations together for creative multi-step workflows.
Run Cline as an ACP agent in JetBrains, Neovim, Zed, and more.
</Card>
<Card title="CLI Samples" icon="flask" href="/cline-cli/samples/overview">
Real-world examples of headless workflows and automation patterns.
<Card title="Use in Other Editors" icon="code" href="/cline-cli/acp-editor-integrations">
Run Cline as an ACP agent in JetBrains, Neovim, Zed, and more.
</Card>
</Columns>
@@ -3,6 +3,8 @@ title: "GitHub Actions Integration"
description: "Automatically respond to GitHub issues by mentioning @cline in comments using Cline CLI in GitHub Actions."
---
# GitHub Integration Sample
Automate GitHub issue analysis with AI. Mention `@cline` in any issue comment to trigger an autonomous investigation that reads files, analyzes code, and provides actionable insights - all running automatically in GitHub Actions.
@@ -271,7 +273,7 @@ git push
Once set up, simply mention `@cline` in any issue comment:
```text
```
@cline what's causing this error?
@cline analyze the root cause
+4 -2
View File
@@ -3,6 +3,8 @@ title: "GitHub Issue RCA Sample"
description: "Automated GitHub issue analysis using Cline CLI to identify root causes."
---
# GitHub Root Cause Analysis
Automated GitHub issue analysis using Cline CLI. This script uses Cline's autonomous AI capabilities to fetch, analyze, and identify root causes of GitHub issues, outputting clean, parseable results that can be easily integrated into your development workflows.
<Note>
@@ -201,7 +203,7 @@ fi
This is where the magic happens:
```bash
# Ask Cline for its analysis, showing only the summary
# Ask Cline for his analysis, showing only the summary
cline -y "$PROMPT: $ISSUE_URL" --mode act $ADDRESS -F json | \
sed -n '/^{/,$p' | \
jq -r 'select(.say == "completion_result") | .text' | \
@@ -378,4 +380,4 @@ This pattern can be adapted for many other automation scenarios, from pull reque
- [CLI Installation Guide](https://docs.cline.bot/cline-cli/installation)
- [CLI Reference Documentation](https://docs.cline.bot/cline-cli/cli-reference)
- [Headless Mode](https://docs.cline.bot/cline-cli/three-core-flows)
- [Three Core Flows](https://docs.cline.bot/cline-cli/three-core-flows)
@@ -3,8 +3,14 @@ title: "GitHub PR Review"
description: "Automatically review Pull Requests with AI using Cline CLI in GitHub Actions."
---
# GitHub PR Review Sample
Automate code review for every Pull Request. Detailed analysis, security checks, and code suggestions provided by Cline running autonomously in GitHub Actions.
<Frame>
<img src="https://storage.googleapis.com/cline_public_images/cli-pr-review.png" alt="Cline PR Review Comment" width="600" />
</Frame>
## The Workflow
When a PR is opened or marked ready for review, this workflow:
+13 -8
View File
@@ -3,21 +3,26 @@ title: "Model Orchestration"
description: "Use multiple AI models strategically: optimize costs, reduce bias, and leverage model-specific strengths in your workflows"
---
# Model Orchestration
Cline CLI's `--config` and `--thinking` flags enable sophisticated multi-model workflows. Instead of using a single model for all tasks, you can route different work to different models based on cost, capability, and specialization.
## Why Orchestrate Multiple Models?
**Cost Optimization**
By routing work to the right model for the job, you can dramatically reduce API costs. Fast, inexpensive models like Haiku and Gemini Flash handle simple tasks such as summarization, while expensive models like Opus and O1 are reserved for complex reasoning and planning. This approach can reduce costs by 10-100x on routine operations.
- Use fast, cheap models (Haiku, Gemini Flash) for simple tasks like summarization
- Reserve expensive models (Opus, O1) for complex reasoning and planning
- Reduce API costs by 10-100x on routine operations
**Bias Reduction**
Different models catch different issues, so cross-validating solutions with multiple AI perspectives helps reduce blind spots that come from relying on a single model. In code reviews especially, combining viewpoints surfaces problems that any one model might miss.
- Different models catch different issues in code reviews
- Cross-validate solutions with multiple AI perspectives
- Reduce blind spots from single-model thinking
**Specialization**
Certain models excel in specific domains: Codex and DeepSeek are strong at code generation, while GPT-4 and Claude shine at documentation and prose. Security analysis in particular benefits from combining multiple model viewpoints, since each brings different training data and heuristics to the table.
- Some models excel at code (Codex, DeepSeek)
- Others are better at documentation (GPT-4, Claude)
- Security analysis benefits from multiple viewpoints
## Pattern 1: CI/CD Code Review
@@ -208,8 +213,8 @@ cat *-sec.md | cline -y "find security issues all 3 reviews mentioned"
Complete documentation for --config and --thinking flags
</Card>
<Card title="Headless Mode" icon="robot" href="/cline-cli/three-core-flows">
Run Cline autonomously in scripts, CI/CD pipelines, and automated workflows.
<Card title="Three Core Flows" icon="route" href="/cline-cli/three-core-flows">
Learn about interactive mode, headless automation, and multi-instance workflows
</Card>
<Card title="Model Selection Guide" icon="brain" href="/core-features/model-selection-guide">
+1 -1
View File
@@ -53,4 +53,4 @@ This section provides sample implementations that demonstrate various Cline CLI
- [CLI Installation Guide](/cline-cli/installation)
- [CLI Reference Documentation](/cline-cli/cli-reference)
- [Headless Mode](/cline-cli/three-core-flows)
- [Three Core Flows](/cline-cli/three-core-flows)
@@ -3,6 +3,8 @@ title: "Worktree Workflows"
description: "Use Git worktrees with Cline CLI to run parallel tasks, test different approaches, and pipe context between isolated environments"
---
# Worktree Workflows
Git worktrees let you have multiple branches checked out simultaneously in different folders. Combined with Cline CLI's `--cwd` flag, this enables powerful parallel development workflows and isolated experimentation.
<Tip>
@@ -267,7 +269,7 @@ git worktree remove ~/cline-worktrees/feature-auth
Complete documentation for --cwd and all other CLI flags
</Card>
<Card title="Headless Mode" icon="robot" href="/cline-cli/three-core-flows">
Run Cline autonomously in scripts, CI/CD pipelines, and automated workflows.
<Card title="Three Core Flows" icon="route" href="/cline-cli/three-core-flows">
Learn about interactive mode, task mode, and plain text workflows
</Card>
</Columns>
+129 -81
View File
@@ -1,62 +1,72 @@
---
title: "Headless Mode"
description: "Run Cline autonomously in scripts, CI/CD pipelines, and automated workflows"
title: "CLI Workflows"
description: "Learn the three ways to use Cline CLI: interactive mode, direct task execution, and automation"
---
Headless mode runs Cline without an interactive interface — perfect for automation, scripting, and CI/CD pipelines where human interaction isn't possible or desired. Cline executes tasks, produces clean text or JSON output, and exits when complete.
For collaborative, conversational development, see [Interactive Mode](/cline-cli/interactive-mode) instead.
Cline CLI supports three primary workflows, each optimized for different use cases. Choose the approach that best fits your needs.
<Note>
**Migrating from an older CLI version?** Instance commands (`cline instance new/list/kill`) have been removed in Cline CLI 2.0. The new architecture is simpler — just use `cline -y "task"` for headless execution.
**Migrating from an older CLI version?** Instance commands (`cline instance new/list/kill`) have been removed in Cline CLI 2.0. The new architecture is simpler. Just run `cline` for interactive mode or `cline "task"` for direct execution.
</Note>
## When Headless Mode Activates
## 1. Interactive Mode
Cline automatically enters headless mode when any of these conditions are met:
The interactive CLI provides the richest experience for interactive development.
| Invocation | Reason |
|------------|--------|
| `cline -y "task"` | `-y`/`--yolo` flag forces headless |
| `cline --json "task"` | `--json` flag forces headless |
| `cat file \| cline "task"` | stdin is piped |
| `cline "task" > output.txt` | stdout is redirected |
If none of these apply (e.g., running `cline` or `cline "task"` in a terminal), Cline launches in [interactive mode](/cline-cli/interactive-mode).
## YOLO Mode (Fully Autonomous)
The `-y` or `--yolo` flag enables fully autonomous operation — Cline approves all actions and runs without prompts:
### Getting Started
```bash
cline -y "Run the test suite and fix any failures"
cline
```
In YOLO mode:
- All actions are auto-approved
- Output is plain text (non-interactive)
- Process exits automatically when complete
- Perfect for CI/CD and scripts
This launches an interactive session in your current directory. Type your task, and Cline will analyze and execute it.
<Warning>
YOLO mode gives Cline full autonomy. Run on a clean git branch so you can easily revert changes if needed.
</Warning>
### Key Features
### Mode Selection
**Plan/Act Mode Toggle** - Press `Tab` to switch between modes:
- **Plan Mode**: Cline analyzes your request and presents a strategy
- **Act Mode**: Cline executes actions directly
Control whether Cline plans first or acts immediately:
**Auto-approve Toggle** - Press `Shift+Tab` to enable automatic approval for all actions.
**Slash Commands** - Type `/` for quick access to:
- `/settings` - Configure providers, models, and features
- `/models` - Quick model switching
- `/history` - Browse and resume previous tasks
- `/clear` - Start a fresh task
- `/help` - Show available commands
**File Mentions** - Type `@` to reference workspace files:
```
@src/utils.ts add error handling to this file
```
**Session Summary** - When you exit with `Ctrl+C`, Cline displays a summary of your session including tasks completed, files modified, and token usage.
### When to Use Interactive Mode
- Exploring a new codebase
- Complex refactoring that requires back-and-forth
- Learning how Cline approaches problems
- Tasks where you want to review before executing
[Full Interactive Mode Guide →](/cline-cli/interactive-mode)
## 2. Direct Task Execution
Execute tasks directly from the command line without entering interactive mode.
### Basic Usage
```bash
# Start in Plan mode (analyze before acting)
cline -y -p "Design a REST API for user management"
# Start in Act mode (default)
cline -y -a "Fix the typo in README.md"
cline "Add unit tests to utils.js"
```
## Piping Context
Cline analyzes your task, creates a plan, and executes it. You'll be prompted for approval at key decision points.
Pipe file contents or command output into Cline to provide context:
### Piping Context
Pipe file contents or command output into Cline:
```bash
# Explain a file
@@ -67,16 +77,11 @@ git diff | cline "Review these changes and suggest improvements"
# Analyze command output
npm test 2>&1 | cline "Analyze these test failures and fix them"
# Pipe a GitHub PR diff
gh pr diff 123 | cline -y "Review this PR"
```
When stdin is piped, Cline automatically enters headless mode — the piped content becomes part of the task context.
### Chaining Cline Commands
## Chaining Commands
Pipe Cline's output into another Cline instance for multi-step workflows:
Pipe Cline's output into another Cline instance for creative workflows:
```bash
# Explain changes, then write a commit message
@@ -89,9 +94,59 @@ cline -y "create a fibonacci function" | cline -y "write unit tests for this cod
git diff | cline -y "explain" | cline -y "write a haiku about this"
```
## JSON Output
### Including Images
Use `--json` for machine-readable output that's easy to parse in scripts:
Attach images to your task:
```bash
cline task -i screenshot.png "Fix the layout issue shown in this screenshot"
# Or reference inline
cline "Fix the UI shown in @./design-mockup.png"
```
### Mode Selection
```bash
# Start in Plan mode (analyze before acting)
cline -p "Design a REST API for user management"
# Start in Act mode (default)
cline -a "Fix the typo in README.md"
```
### When to Use Direct Execution
- Quick, well-defined tasks
- Tasks with sufficient context in the prompt
- Scripting and shell workflows
- When you don't need interactive conversation
## 3. Automation & CI/CD
For fully autonomous operation in scripts, CI/CD pipelines, and automated workflows.
### YOLO Mode (Yes Mode)
The `-y` or `--yolo` flag enables fully autonomous operation:
```bash
cline -y "Run the test suite and fix any failures"
```
In YOLO mode:
- All actions are auto-approved
- Output is plain text (non-interactive)
- Process exits automatically when complete
- Perfect for CI/CD and scripts
<Warning>
Run YOLO mode on a clean git branch or directory. You get speed in exchange for oversight, so be ready to revert if needed.
</Warning>
### JSON Output
Use `--json` for machine-readable output:
```bash
cline --json "List all TODO comments in the codebase" | jq '.text'
@@ -109,36 +164,25 @@ JSON output follows the same format as task files in `~/.cline/data/tasks/<id>/u
| `reasoning` | `string` | (Optional) Model reasoning |
| `partial` | `boolean` | (Optional) Streaming flag |
## Including Images
### Timeout Control
Attach images to your headless task:
```bash
cline -y -i screenshot.png "Fix the layout issue shown in this screenshot"
# Or reference inline
cline -y "Fix the UI shown in @./design-mockup.png"
```
## Timeout Control
Set a maximum execution time to prevent runaway tasks:
Set a maximum execution time:
```bash
cline -y --timeout 600 "Run full test suite"
```
## Environment Variables
### Environment Variables
Control Cline behavior via environment variables — useful for CI/CD where you can't use interactive configuration.
Control Cline behavior via environment variables:
**CLINE_DIR** Custom configuration directory:
**CLINE_DIR** - Custom configuration directory:
```bash
export CLINE_DIR=/path/to/config
cline -y "your task"
```
**CLINE_COMMAND_PERMISSIONS** Restrict allowed commands:
**CLINE_COMMAND_PERMISSIONS** - Restrict allowed commands:
```bash
export CLINE_COMMAND_PERMISSIONS='{"allow": ["npm *", "git *"], "deny": ["rm -rf *"]}'
cline -y "your task"
@@ -146,8 +190,6 @@ cline -y "your task"
See [Configuration](/cline-cli/configuration#environment-variables) for full documentation.
## CI/CD Integration
### GitHub Actions Example
Automate PR reviews with Cline:
@@ -190,7 +232,7 @@ jobs:
### Shell Script Example
Create a reusable code review script:
Create a code review script:
```bash
#!/bin/bash
@@ -210,24 +252,30 @@ fi
echo "$DIFF" | cline -y --json "Review this code diff for issues" | jq -r '.text'
```
## Common Use Cases
### When to Use Automation Mode
| Use Case | Example |
|----------|---------|
| Code review | `git diff \| cline -y "Review these changes"` |
| Fix test failures | `cline -y "Run tests and fix any failures"` |
| Generate release notes | `git log --oneline v1.0..v1.1 \| cline -y "Write release notes"` |
| Fix lint errors | `cline -y "Fix all ESLint errors in src/"` |
| Update dependencies | `cline -y "Update dependencies with known vulnerabilities"` |
| Migrate code patterns | `cline -y "Update all deprecated React lifecycle methods"` |
| PR automation | `gh pr diff 123 \| cline -y "Review this PR"` |
| Batch processing | `cline -y --json "List all TODO comments" \| jq '.text'` |
- CI/CD pipelines
- Scheduled maintenance tasks
- Batch processing
- Any workflow requiring non-interactive execution
## Choosing the Right Flow
| Use Case | Recommended Flow |
|----------|------------------|
| Exploring a new codebase | Interactive Mode |
| Complex refactoring | Interactive Mode (Plan first) |
| Quick file edits | Direct Execution |
| Code review | Direct Execution with pipe |
| CI/CD integration | Automation (`-y` flag) |
| Scheduled tasks | Automation (`-y` flag) |
| Learning Cline | Interactive Mode |
## Next Steps
<Columns cols={2}>
<Card title="Interactive Mode" icon="terminal" href="/cline-cli/interactive-mode">
For hands-on development with keyboard shortcuts, slash commands, and file mentions.
Master keyboard shortcuts, slash commands, and file mentions.
</Card>
<Card title="CLI Reference" icon="book" href="/cline-cli/cli-reference">
@@ -238,7 +286,7 @@ echo "$DIFF" | cline -y --json "Review this code diff for issues" | jq -r '.text
Environment variables, rules, and advanced settings.
</Card>
<Card title="CLI Samples" icon="flask" href="/cline-cli/samples/overview">
Real-world examples of headless workflows and automation patterns.
<Card title="YOLO Mode" icon="zap" href="/features/yolo-mode">
Deep dive into autonomous execution and safety considerations.
</Card>
</Columns>
-316
View File
@@ -1,316 +0,0 @@
---
title: "Documentation Templates"
sidebarTitle: "Templates"
description: "Templates for different types of Cline documentation"
---
Use these templates as starting points for new documentation. Each template is designed for a specific purpose. Choose the one that best fits what you're documenting.
## Choosing a Template
| If you're documenting... | Use this template |
|--------------------------|-------------------|
| What a feature does and how to use it | Feature Doc |
| How to accomplish a specific task | How-To Guide |
| Technical specifications or API details | Reference Doc |
| A complete project walkthrough | Tutorial |
## Feature Doc
Use this template when explaining a Cline feature. Focus on what it does, how to use it, and real examples.
````text
---
title: "Feature Name"
sidebarTitle: "Feature Name"
---
[One sentence explaining what this feature does.]
<Frame>
<img src="..." alt="Feature in action" />
</Frame>
[1-2 paragraphs explaining the feature in plain terms. What problem does it
solve? Why would someone use it?]
## How It Works
[Explain the mechanics without jargon. What happens when you use this feature?]
## Using [Feature Name]
[Show how to access and use it. Include the exact UI path.]
### [Option or Variation 1]
[Details with examples]
### [Option or Variation 2]
[Details with examples]
## Inspiration
[Share how you personally use this feature. Use "I" voice. Give 2-3 real
examples that spark imagination about what's possible.]
<Note>
[Important caveat, limitation, or requirement]
</Note>
````
### Example: Checkpoints Feature
Here's how the [Checkpoints](/core-workflows/checkpoints) doc follows this pattern:
- Opens with one clear sentence about what checkpoints do
- Shows a screenshot of the feature in action
- Explains how checkpoints work under the hood
- Shows exact steps to create and restore checkpoints
- Includes real examples of when checkpoints save the day
## How-To Guide
Use this template when showing how to accomplish a specific task. Focus on clear steps and troubleshooting.
````text
---
title: "How to [Accomplish Task]"
sidebarTitle: "[Short Title]"
description: "[One sentence describing what the reader will learn]"
---
[Brief intro explaining what problem this guide solves and what you'll end up
with after following it.]
## Prerequisites
[What the reader needs before starting. Keep it short. Link to other docs
rather than explaining setup here.]
- Cline installed and configured
- [Other requirement]
## Steps
<Steps>
<Step title="[First Action]">
[Clear instructions. Show exactly what to click or type.]
```bash
example command if needed
```
</Step>
<Step title="[Second Action]">
[Next step. Include screenshots for complex UI interactions.]
<Frame>
<img src="..." alt="What you should see" />
</Frame>
</Step>
<Step title="[Final Action]">
[Complete the task. Show the expected result.]
</Step>
</Steps>
## Troubleshooting
Common issues and how to fix them:
- **Problem description**: Solution in one or two sentences.
- **Another problem**: Another solution.
## Next Steps
<Card title="Related Feature" icon="arrow-right" href="/path/to/related">
Continue learning with this related guide.
</Card>
````
### Example: Your First Project
The [Your First Project](/getting-started/your-first-project) guide follows this pattern:
- Clear goal stated upfront
- Prerequisites listed briefly
- Step-by-step instructions with the Steps component
- Troubleshooting section for common issues
## Reference Doc
Use this template for technical specifications, API documentation, or detailed configuration options.
````text
---
title: "[Component/API] Reference"
sidebarTitle: "[Short Title]"
description: "[What this reference covers]"
---
[Brief description of what this reference documents and when you'd need it.]
## Overview
[High-level explanation. What is this component? What role does it play?]
## [Category 1]
### [Item Name]
[What it does in one sentence.]
| Property | Type | Default | Description |
|----------|------|---------|-------------|
| `propertyName` | `string` | `"default"` | What this property controls |
| `anotherProp` | `boolean` | `false` | What this does |
**Example:**
```typescript
// Show practical usage
const example = {
propertyName: "custom value",
anotherProp: true
}
```
### [Another Item]
[Continue for each item in this category.]
## [Category 2]
[Continue with other categories as needed.]
## Examples
[Show 2-3 complete, practical examples that combine multiple concepts.]
### [Example 1 Title]
```typescript
// Complete working example
```
### [Example 2 Title]
```typescript
// Another complete example
```
## Related
- [Related Doc 1](/path/to/doc) - Brief description
- [Related Doc 2](/path/to/doc) - Brief description
````
### Example: Cline Tools Guide
The [Cline Tools Guide](/tools-reference/all-cline-tools) follows this pattern:
- Overview of the tool system
- Each tool documented with parameters and examples
- Practical examples showing tools in context
## Tutorial
Use this template for comprehensive project walkthroughs where users build something from start to finish.
````text
---
title: "[Build/Create X] Tutorial"
sidebarTitle: "[Short Title]"
description: "[What the reader will build]"
---
In this tutorial, you'll build [specific outcome]. By the end, you'll have
[tangible result you can see/use].
<Frame>
<img src="..." alt="Preview of what you'll build" />
</Frame>
## What You'll Learn
- [Skill or concept 1]
- [Skill or concept 2]
- [Skill or concept 3]
## Prerequisites
[Required setup. Link to installation guides rather than repeating them.]
- [Prerequisite 1]
- [Prerequisite 2]
## Part 1: [First Major Section]
[Introduction to this section. What are we doing and why?]
### [Subsection]
[Detailed walkthrough with code blocks and explanations.]
```typescript
// Code that the reader should write or understand
```
[Explain what the code does and why.]
## Part 2: [Second Major Section]
[Continue building on Part 1.]
### [Subsection]
[More detailed walkthrough.]
## Part 3: [Final Section]
[Complete the project.]
## Summary
You built [what they built]. Along the way, you learned:
- [Key takeaway 1]
- [Key takeaway 2]
- [Key takeaway 3]
## Next Steps
<CardGroup cols={2}>
<Card title="Go Deeper" icon="book" href="/path/to/advanced">
Learn more advanced techniques.
</Card>
<Card title="Related Tutorial" icon="code" href="/path/to/related">
Build something else with similar concepts.
</Card>
</CardGroup>
````
### Example Structure
A good tutorial:
- Shows the end result upfront so readers know what they're building
- Breaks the work into logical parts
- Explains the "why" alongside the "how"
- Ends with clear next steps
## Quick Tips
When using these templates:
1. **Delete sections you don't need.** Templates are starting points, not rigid structures.
2. **Add sections that make sense.** If your doc needs something not in the template, add it.
3. **Keep the reader moving forward.** Every section should lead naturally to the next.
4. **Test your own instructions.** Follow your guide from scratch to catch missing steps.
<Tip>
Use the `/write-docs` workflow to generate documentation from these templates automatically.
Cline helps you fill in each section based on your project.
</Tip>
-200
View File
@@ -1,200 +0,0 @@
---
title: "Documentation Guide"
sidebarTitle: "Documentation Guide"
description: "How to write and contribute to Cline documentation"
---
Cline's documentation lives in the `docs/` directory and uses [Mintlify](https://mintlify.com) for rendering. This guide covers how to write docs that match Cline's established style.
## Using the Documentation Workflow
The fastest way to create documentation is using the `/write-docs` workflow. Type `/write-docs` in Cline and describe what you want to document. Cline guides you through a 4-step process:
1. **Research**: Examine existing docs structure and patterns
2. **Scope**: Clarify audience, doc type, and key use cases
3. **Outline**: Select a template and create structure
4. **Write**: Generate documentation following style guidelines
The workflow file lives at `.clinerules/workflows/write-docs.md` and contains templates, style rules, and examples.
## Documentation Principles
### Write for Developers
Your audience is developers who value their time. Get to the point. Every sentence should either help them understand something or help them do something.
```markdown
# Good
Switch to bash in Cline Settings → Terminal → Default Terminal Profile.
# Bad
Users who are experiencing issues may find it helpful to navigate to the
Cline settings menu where they can locate the terminal configuration
options and subsequently modify the default terminal profile setting.
```
### Show Real Examples
Abstract descriptions don't help anyone. Show actual code, real file paths, and concrete implementations.
```markdown
# Good
I use `/deep-planning` whenever I'm building features that touch multiple
parts of the codebase. For example, when adding authentication, Cline
mapped every endpoint and created a migration plan that avoided breaking changes.
# Bad
The deep planning feature can be utilized for various complex tasks
that may require careful consideration and planning.
```
### Use Active Voice
Cline does things. Files don't get created by Cline, Cline creates files.
```markdown
# Good
Cline reads your project files and builds context automatically.
# Bad
Project files are read and context is built automatically.
```
### Use Neutral Pronouns for Cline
Refer to Cline as "it" not "he". Cline is software, not a person.
```markdown
# Good
When Cline encounters an error, it suggests fixes.
# Bad
When Cline encounters an error, he suggests fixes.
```
## File Format
All documentation uses MDX format with YAML frontmatter:
```yaml
---
title: "Full Page Title"
sidebarTitle: "Shorter Nav Title" # optional
description: "One sentence for SEO" # optional but recommended
---
```
### Adding New Pages
After creating a new `.mdx` file, add it to `docs/docs.json` in the appropriate navigation group:
```json
{
"group": "Features",
"pages": [
"features/existing-page",
"features/your-new-page"
]
}
```
## Mintlify Components
Use these components appropriately throughout your docs.
### Frame
Wrap all images and videos:
```jsx
<Frame>
<img
src="https://storage.googleapis.com/cline_public_images/docs/assets/filename.png"
alt="Descriptive alt text"
/>
</Frame>
```
### Callouts
Use sparingly and purposefully:
```jsx
<Tip>Helpful suggestions that improve the experience.</Tip>
<Note>Important information the reader needs to know.</Note>
<Warning>Something that could cause problems if ignored.</Warning>
```
### Steps
For sequential procedures:
```jsx
<Steps>
<Step title="Install the Extension">
Search for "Cline" in the VS Code marketplace.
</Step>
<Step title="Configure Your Model">
Open settings and add your API key.
</Step>
</Steps>
```
### Cards
For navigation and feature overviews:
```jsx
<CardGroup cols={2}>
<Card title="Getting Started" icon="rocket" href="/getting-started/installing-cline">
Install Cline and set up your first project.
</Card>
<Card title="Features" icon="wand-magic-sparkles" href="/core-workflows/plan-and-act">
Explore what Cline can do.
</Card>
</CardGroup>
```
## Style Rules
Quick reference for consistent documentation:
| Do | Don't |
|---|---|
| Use "use" | Use "utilize" |
| Keep sentences under 25 words | Write run-on sentences |
| Use bullet points for lists | Write walls of text |
| Show where things are in the UI | Assume users can find features |
| Cross-link related docs | Leave readers stranded |
| Use code blocks with language tags | Use inline code for long snippets |
### Avoid These Patterns
- Em dashes and emojis
- Starting with "This document explains..."
- The **Bold Text**: description pattern
- Explaining obvious things
- Passive voice
## Previewing Changes
Run the docs locally to preview your changes:
```bash
cd docs
npm install # first time only
npm run dev
```
Open `http://localhost:3000` to see your changes in real time.
## Related Resources
<CardGroup cols={2}>
<Card title="Documentation Templates" icon="file-lines" href="/contributing/doc-templates">
Templates for different documentation types.
</Card>
<Card title="Workflows" icon="diagram-project" href="/customization/workflows">
Learn about Cline's workflow system.
</Card>
</CardGroup>
+8 -26
View File
@@ -1,6 +1,6 @@
---
title: "Model Selection Guide"
description: "Choose the right AI model for your workflow based on reliability, speed, cost, and context window size."
description: "Last updated: August 20, 2025."
---
New models drop constantly, so this guide focuses on what's working well with Cline right now. We'll keep it updated as the landscape shifts.
@@ -59,17 +59,12 @@ Choose your preferred AI provider from the dropdown menu.
| **Cline** | Easiest setup | No API keys needed, access to multiple models including stealth models |
| **OpenRouter** | Value seekers | Multiple models, competitive pricing |
| **Anthropic** | Reliability | Claude models, most dependable tool usage |
| **OpenAI** | Latest tech | GPT-5, o3, o4-mini models |
| **OpenAI Codex** | ChatGPT subscribers | Use your ChatGPT subscription — no API key needed |
| **Google Gemini** | Large context | Gemini 3/2.5 with up to 2M context |
| **DeepSeek** | Budget reasoning | V3.2, R1 models at low cost |
| **Alibaba Qwen** | Open source coding | Qwen3 Coder with 1M context |
| **Moonshot** | Agentic coding | Kimi K2.5 with 262K context |
| **Cerebras** | Speed | Up to 2,600 tokens/sec |
| **OpenAI** | Latest tech | GPT models |
| **Google Gemini** | Large context | Google's AI models |
| **AWS Bedrock** | Enterprise | Advanced features |
| **Ollama** | Privacy | Run models locally |
See the [full provider list](/getting-started/authorizing-with-cline) for all 30+ supported providers including xAI Grok, Mistral, Groq, Fireworks, Together, Baseten, SambaNova, Nebius, Hugging Face, and more.
See the [full provider list](/provider-config) for more options including Cerebras, Vertex AI, Azure, and more.
<Info>
**Recommended for beginners:** Start with **Cline** as your provider - no API key management needed, instant access to multiple models, and occasional free inferencing through partner providers.
@@ -86,19 +81,6 @@ The next step depends on which provider you selected.
- You'll be redirected to [app.cline.bot](https://app.cline.bot) to authenticate
- After signing in, return to your IDE
<Note>
For detailed information about the Cline authentication flow, OAuth tokens, and troubleshooting, see [Authorizing with Cline](/getting-started/authorizing-with-cline).
</Note>
#### If you selected **OpenAI Codex** as your provider:
- **No API key needed!** If you have a ChatGPT subscription (Plus, Pro, or Team), you can use it directly in Cline
- Click **"Sign in with OpenAI"** to authenticate via your browser
- Once authorized, all models available on your OpenAI plan will appear automatically
- Usage is governed by your ChatGPT subscription — no separate API billing
See the full [OpenAI Codex setup guide](/provider-config/openai-codex) for details.
#### If you selected any other provider:
You'll need to get an API key from your chosen provider:
@@ -108,7 +90,7 @@ You'll need to get an API key from your chosen provider:
- **OpenRouter**: [openrouter.ai/keys](https://openrouter.ai/keys)
- **OpenAI**: [platform.openai.com/api-keys](https://platform.openai.com/api-keys)
- **Google**: [aistudio.google.com/apikey](https://aistudio.google.com/apikey)
- **Others**: See [Provider Setup Guide](/getting-started/authorizing-with-cline)
- **Others**: See [Provider Setup Guide](/provider-config)
2. **Generate a new API key** on the provider's website
@@ -186,7 +168,8 @@ Selecting the right model involves balancing several factors. Use this framework
## Model Comparison Resources
For detailed model comparisons and performance metrics, see:
For detailed model comparisons, pricing, and performance metrics, see:
- [**Model Comparison & Pricing**](/model-config/model-comparison) - Complete pricing tables and performance benchmarks
- [**Context Window Guide**](/model-config/context-windows) - Understanding and optimizing context usage
## Open Source vs Closed Source
@@ -212,9 +195,8 @@ For detailed model comparisons and performance metrics, see:
| Huge context windows | Gemini 2.5 Pro or Claude Sonnet 4.5 |
| Open source | Qwen3 Coder, Z AI GLM 4.5, or Kimi K2 |
| Latest tech | GPT-5 |
| To use your ChatGPT subscription | [OpenAI Codex](/provider-config/openai-codex) — sign in with your OpenAI account, no API key needed |
| Speed | Qwen3 Coder on Cerebras (fastest available) |
## What Others Are Using
Check [Vercel's leaderboard](https://vercel.com/ai-gateway/leaderboards) to see real usage patterns from the community.
Check [OpenRouter's Cline usage stats](https://openrouter.ai/apps?url=https%3A%2F%2Fcline.bot%2F) to see real usage patterns from the community.
-95
View File
@@ -1,95 +0,0 @@
---
title: "Checkpoints"
sidebarTitle: "Checkpoints"
description: "Roll back code changes while keeping your conversation. Experiment freely."
---
Checkpoints let you undo code changes without losing your conversation. Every time Cline modifies a file or runs a command, it saves a snapshot of your project files. You can restore to any checkpoint, keeping the context you've built while reverting the code.
This changes how you work with Cline. Instead of carefully reviewing every change before approving, you can let Cline move fast and roll back if something goes wrong. The cost of a mistake drops to nearly zero.
<Tip>
Checkpoints are enabled by default. See [Enable or Disable Checkpoints](#enable-or-disable-checkpoints) if you need to turn them off.
</Tip>
## How It Works
Cline maintains a shadow Git repository separate from your project's actual Git history. After each tool use (file edits, commands, etc.), Cline commits the current state of your files to this shadow repo. Your main Git repository stays untouched.
This means:
- Your Git history remains clean and under your control
- Checkpoints capture everything, including files not tracked by Git
- You can restore to any point in a task without affecting commits you've made
- Checkpoints persist across editor sessions
Each checkpoint captures the complete file state at that moment. If Cline edits three files in sequence, you get three checkpoints and can restore to any of them independently.
## Enable or Disable Checkpoints
Checkpoints are enabled by default. To toggle them:
1. Open Cline settings (gear icon in the Cline sidebar)
2. Scroll to the "Feature Settings" section
3. Toggle "Enable Checkpoints"
<Note>
For very large repositories, checkpoints may use significant storage and slow down Cline as it commits file snapshots after each tool use. Consider disabling them if you notice performance issues.
</Note>
## Viewing and Comparing Changes
After each tool use, a checkpoint indicator appears in your conversation. Look for a bookmark icon labeled "Checkpoint" with a dotted line connecting to **Compare** and **Restore** buttons.
Click **Compare** to open a diff view showing exactly what changed at that checkpoint. This opens in your editor's diff viewer, letting you see additions, deletions, and modifications across all affected files.
This is useful when Cline makes changes you want to understand before deciding whether to keep them. You can review the diff, then either continue or restore to undo.
## Restoring Checkpoints
Click **Restore** next to any step to open the restore menu. You have three options:
| Option | What It Does | When to Use It |
|--------|--------------|----------------|
| **Restore Files** | Reverts your project's files to the snapshot at this checkpoint | Undoing code changes while keeping the conversation |
| **Restore Task Only** | Deletes messages after this point, does not affect files | Trying a different prompt while keeping current code |
| **Restore Files & Task** | Reverts files and deletes messages after this point | Starting over completely from a known good state |
The right choice depends on what went wrong:
- If the conversation is productive but the code changes broke something, use **Restore Files**. Cline keeps all the context you've discussed and can try a different implementation.
- If Cline's code changes are good but the conversation went off track, use **Restore Task Only**. You keep the files and can guide the conversation differently.
- If you want to start over from a clean slate, use **Restore Files & Task**. This resets both your files and the conversation to that checkpoint.
## When to Use Checkpoints
| Scenario | Recommended Action |
|----------|-------------------|
| Cline refactored code and broke something | Restore Files, ask for a different approach |
| Experimenting with multiple solutions | Compare each checkpoint, restore to the best one |
| Cline misunderstood your intent | Restore Files & Task, rephrase your request |
| Want to try a different prompt | Restore Task Only, keep the files, resubmit |
| Reviewing changes before committing to Git | Use Compare to inspect, then commit manually |
| Testing risky changes | Let Cline proceed, restore if it fails |
## Working with Auto-Approve
Checkpoints make [auto-approve](/features/auto-approve) practical. Without checkpoints, auto-approve feels risky because Cline can make many changes before you notice a problem. With checkpoints, you can let Cline work autonomously and roll back if needed.
A typical workflow:
1. Enable auto-approve for file edits and commands
2. Let Cline work through your task quickly
3. Review the final result
4. If something is wrong, restore to the last good checkpoint
5. Give Cline more specific guidance
This approach is faster than reviewing every change individually, and checkpoints provide the safety net.
## Checkpoints and Message Editing
The message editing feature integrates with checkpoints. When you edit a previous message and select "Restore All," Cline restores your files to the checkpoint at that point before resubmitting your edited message.
This lets you fix a poorly worded prompt and undo all the changes that resulted from it in one action.

Some files were not shown because too many files have changed in this diff Show More