removing shadow

Testing Platform - Support partial response validation via meta.expected

.changeset/dull-beds-confess.md

@@ -0,0 +1,5 @@
+---
+"claude-dev": patch
+---
+
+Checkpoints multiroot pt.1: Accept array of workspaces when initializting checkpoints

.changeset/fine-hats-try.md

@@ -0,0 +1,5 @@
+---
+"claude-dev": patch
+---
+
+Support partial response validation via meta.expected

.changeset/heavy-llamas-smile.md

@@ -0,0 +1,5 @@
+---
+"claude-dev": patch
+---
+
+Run Testing platform within Test workflow

.changeset/modern-rivers-appear.md

@@ -0,0 +1,5 @@
+---
+"claude-dev": patch
+---
+
+Refactor Task Header UI with interactive context window management

.changeset/proud-cougars-kick.md

@@ -0,0 +1,5 @@
+---
+"claude-dev": patch
+---
+
+remove temperature settings in z.ai models

.changeset/six-drinks-wink.md

@@ -0,0 +1,5 @@
+---
+"claude-dev": patch
+---
+
+Feature flags are now accessible too all users instead of authenticated users only.

.changeset/slimy-cougars-hope.md

@@ -0,0 +1,5 @@
+---
+"claude-dev": patch
+---
+
+Update anthropic input token usage calculation

.changeset/stupid-laws-jam.md

@@ -0,0 +1,5 @@
+---
+"claude-dev": patch
+---
+
+Support new Cline endpoint auth flow

.changeset/tasty-rocks-sit.md

@@ -0,0 +1,5 @@
+---
+"claude-dev": patch
+---
+
+Add speech-to-text dictation feature for Cline account users

.github/CODEOWNERS

@@ -1,3 +1,3 @@
 /docs/
-/.github/   @saoudrizwan @dcbartlett
-/README.md  @saoudrizwan @nickbaumann98
+/.github/   @saoudrizwan
+/README.md  @saoudrizwan @nickbaumann98

.github/ISSUE_TEMPLATE/bug_report.yml

@@ -1,64 +1,69 @@
 name: 🐛 Bug Report
 description: File a bug report
-labels: ["bug"]
+labels: ['bug']
 body:
-    - type: markdown
-      attributes:
-          value: |
-              **Important:** All bug reports must be reproducible using Claude 4 Sonnet. Cline uses complex prompts so less capable models may not work as expected.
-    - type: textarea
-      id: what-happened
-      attributes:
-          label: What happened?
-          description: Also tell us, what did you expect to happen?
-          placeholder: Tell us what you see!
-      validations:
-          required: true
-    - type: textarea
-      id: steps
-      attributes:
-          label: Steps to reproduce
-          description: How do you trigger this bug? Please walk us through it step by step.
-          value: |
-              1.
-              2.
-              3.
-      validations:
-          required: false
-    - type: textarea
-      id: logs
-      attributes:
-          label: Relevant API REQUEST output
-          description: Please copy and paste any relevant output. This will be automatically formatted into code, so no need for backticks.
-          render: shell
-    - type: input
-      id: provider-model
-      attributes:
-          label: Provider/Model
-          description: What provider and model were you using when the issue occurred?
-          placeholder: "e.g., cline:anthropic/claude-3.7-sonnet, gemini:gemini-2.5-pro-exp-03-25"
-      validations:
-          required: true
-    - type: textarea
-      id: system-info
-      attributes:
-          label: System Information
-          description: What operating system and hardware are you using?
-          placeholder: |
-              Operating System: Windows 11, macOS Sonoma, Ubuntu 22.04, etc.
-              Hardware: CPU, GPU, RAM specifications if relevant
-              e.g.,
-              OS: Windows 11
-              CPU: Intel Core i7-11700K
-              GPU: NVIDIA GeForce RTX 3070
-              RAM: 32GB DDR4
-      validations:
-          required: true
-    - type: input
-      id: cline-version
-      attributes:
-          label: Cline Version
-          description: What version of Cline are you using? (You can find this at the bottom of the Settings view)
-          placeholder: "e.g., 1.2.3"
-      validations:
-          required: true
+  - type: markdown
+    attributes:
+      value: |
+        **Important:** All bug reports must be reproducible using Claude Sonnet 4. Cline uses complex prompts so less capable models may not work as expected.
+  - type: dropdown
+    id: plugin-type
+    attributes:
+      label: Plugin Type
+      description: Which plugin are you reporting a bug for?
+      options:
+        - VSCode Extension
+        - JetBrains Plugin
+      default: 0
+    validations:
+      required: true
+  - type: input
+    id: cline-version
+    attributes:
+      label: Cline Version
+      description: What version of Cline are you using? (You can find this at the bottom of the Settings view)
+      placeholder: 'e.g., 1.2.3'
+    validations:
+      required: true
+  - type: textarea
+    id: what-happened
+    attributes:
+      label: What happened?
+      description: Also tell us, what did you expect to happen?
+      placeholder: Tell us what you see!
+    validations:
+      required: true
+  - type: textarea
+    id: steps
+    attributes:
+      label: Steps to reproduce
+      description: How do you trigger this bug? Please walk us through it step by step.
+      value: |
+        1.
+        2.
+        3.
+    validations:
+      required: false
+  - type: input
+    id: provider-model
+    attributes:
+      label: Provider/Model
+      description: What provider and model were you using when the issue occurred?
+      placeholder: 'e.g., cline:anthropic/claude-sonnet-4, gemini:gemini-2.5-pro-exp-03-25'
+    validations:
+      required: false
+  - type: textarea
+    id: system-info
+    attributes:
+      label: System Information
+      description: What operating system and hardware are you using?
+      placeholder: |
+        Operating System: Windows 11, macOS Sonoma, Ubuntu 22.04, etc.
+        Hardware: CPU, GPU, RAM specifications if relevant
+        e.g.,
+        OS: Windows 11
+        CPU: Intel Core i7-11700K
+        GPU: NVIDIA GeForce RTX 3070
+        RAM: 32GB DDR4
+    validations:
+      required: false

.github/workflows/publish-nightly.yml

@@ -0,0 +1,75 @@
+name: "Publish Nightly Release"
+
+on:
+  schedule:
+    - cron: '0 12 * * *'  # 4 AM PST (UTC-8) = 12 UTC
+  workflow_dispatch:
+
+permissions:
+    contents: write
+    packages: write
+    checks: write
+    pull-requests: write
+
+jobs:
+    test:
+        uses: ./.github/workflows/test.yml
+
+    publish:
+        needs: test
+        name: Publish Cline (Nightly) Extension
+        if: github.repository == 'cline/cline'
+        runs-on: ubuntu-latest
+        environment: PublishNightly
+
+        steps:
+            - uses: actions/checkout@v4
+
+            - name: Check for recent commits
+              run: |
+                if [ $(git rev-list --count HEAD --since="24 hours ago") -eq 0 ]; then
+                  echo "No commits in last 24 hours, exiting"
+                  exit 0
+                fi
+                echo "Found recent commits, proceeding with build"
+                
+            - name: Setup Node.js
+              uses: actions/setup-node@v4
+              with:
+                  node-version: "lts/*"
+
+            # 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') }}
+
+            - name: Install root dependencies
+              if: steps.root-cache.outputs.cache-hit != 'true'
+              run: npm ci --include=optional
+
+            - name: Install webview-ui dependencies
+              if: steps.webview-cache.outputs.cache-hit != 'true'
+              run: cd webview-ui && npm ci --include=optional
+
+            - name: Install Publishing Tools
+              run: npm install -g @vscode/vsce ovsx
+
+            - name: Publish Extension as Pre-release
+              env:
+                  VSCE_PAT: ${{ secrets.VSCE_PAT }}
+                  OVSX_PAT: ${{ secrets.OVSX_PAT }}
+                  TELEMETRY_SERVICE_API_KEY: ${{ secrets.TELEMETRY_SERVICE_API_KEY }}
+                  ERROR_SERVICE_API_KEY: ${{ secrets.ERROR_SERVICE_API_KEY }}
+                  CLINE_ENVIRONMENT: production
+              run: npm run publish:marketplace:nightly

.github/workflows/publish.yml

@@ -95,6 +95,8 @@ jobs:
                   VSCE_PAT: ${{ secrets.VSCE_PAT }}
                   OVSX_PAT: ${{ secrets.OVSX_PAT }}
                   CLINE_ENVIRONMENT: production
+                  TELEMETRY_SERVICE_API_KEY: ${{ secrets.TELEMETRY_SERVICE_API_KEY }}
+                  ERROR_SERVICE_API_KEY: ${{ secrets.ERROR_SERVICE_API_KEY }}
               run: |
                   # Required to generate the .vsix
                   vsce package  --allow-package-secrets sendgrid --out "cline-${{ steps.get_version.outputs.version }}.vsix"

.github/workflows/test.yml

@@ -90,8 +90,14 @@ jobs:
             - name: Build Tests and Extension
               run: npm run pretest
 
-            - name: Unit Tests
-              run: npm run test:unit
+            - name: Unit Tests (with coverage on Linux)
+              run: |
+                if [ "${{ runner.os }}" = "Linux" ]; then
+                  npm install --no-save nyc
+                  npx nyc --nycrc-path .nycrc.unit.json --reporter=lcov npm run test:unit
+                else
+                  npm run test:unit
+                fi
 
             # Run extension tests with coverage
             - name: Extension Integration Tests with Coverage
@@ -125,6 +131,8 @@ jobs:
                   path: |
                       extension_coverage.txt
                       webview-ui/webview_coverage.txt
+                      coverage-unit/lcov.info
+                      webview-ui/coverage/lcov.info
 
             # Set the check as failed if any of the tests failed
             - name: Check for test failures
@@ -141,6 +149,66 @@ jobs:
                       exit 1
                   fi
 
+    test-platform-integration:
+        runs-on: ubuntu-latest
+        steps:
+            - name: Checkout code
+              uses: actions/checkout@v4
+
+            - name: Setup Node.js environment
+              uses: actions/setup-node@v4
+              with:
+                  node-version: 22
+
+            - name: Cache root dependencies
+              uses: actions/cache@v4
+              id: root-cache
+              with:
+                  path: node_modules
+                  key: ${{ runner.os }}-npm-${{ hashFiles('package-lock.json') }}
+
+            - 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') }}
+
+            # Cache testing-platform dependencies
+            - name: Cache testing-platform dependencies
+              uses: actions/cache@v4
+              id: testing-platform-cache
+              with:
+                path: testing-platform/node_modules
+                key: ${{ runner.os }}-npm-testing-platform-${{ hashFiles('testing-platform/package-lock.json') }}
+
+            - name: Install root dependencies
+              if: steps.root-cache.outputs.cache-hit != 'true'
+              run: npm ci
+
+            - name: Install webview-ui dependencies
+              if: steps.webview-cache.outputs.cache-hit != 'true'
+              run: cd webview-ui && npm ci
+
+            - name: Compile standalone
+              run: npm run compile-standalone
+
+            - name: Install testing platform dependencies
+              if: steps.testing-platform-cache.outputs.cache-hit != 'true'
+              run: cd testing-platform && npm ci
+
+            - name: Running testing platform integration spec tests
+              continue-on-error: true
+              timeout-minutes: 7
+              run: |
+                npm run test:tp-orchestrator -- tests/specs/ --count=1 --coverage
+
+            - name: Save Coverage Reports
+              uses: actions/upload-artifact@v4
+              with:
+                name: test-platform-integration-core-coverage
+                path: coverage/**/lcov.info
+
     coverage:
         needs: test
         runs-on: ubuntu-latest
@@ -219,3 +287,51 @@ jobs:
                     --verbose
               env:
                   GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+
+    qlty:
+        needs: [test, test-platform-integration]
+        runs-on: ubuntu-latest
+        # Run on PRs to main, pushes to main, and manual dispatches
+        steps:
+            - name: Checkout code
+              uses: actions/checkout@v4
+              with:
+                  fetch-depth: 0 # Fetch all history for accurate comparison
+
+            - name: Download unit tests coverage reports
+              uses: actions/download-artifact@v4
+              with:
+                  name: pr-coverage-reports
+                  path: .
+
+            - name: Upload core unit tests coverage to Qlty
+              uses: qltysh/qlty-action/coverage@v2
+              with:
+                  token: ${{ secrets.QLTY_COVERAGE_TOKEN }}
+                  # we can merge multiple files if necessary
+                  files: |
+                    coverage-unit/lcov.info
+                  tag: unit:core
+
+            - name: Upload webview-ui unit tests coverage to Qlty
+              uses: qltysh/qlty-action/coverage@v2
+              with:
+                  token: ${{ secrets.QLTY_COVERAGE_TOKEN }}
+                  # we can merge multiple files if necessary
+                  files: |
+                    webview-ui/coverage/lcov.info
+                  tag: unit:webview-ui
+                  add-prefix: webview-ui/
+
+            - name: Download test platform integration core coverage artifact
+              uses: actions/download-artifact@v4
+              with:
+                name: test-platform-integration-core-coverage
+                path: integration-core-coverage-reports
+
+            - name: Upload core integration tests coverage to Qlty
+              uses: qltysh/qlty-action/coverage@v2
+              with:
+                token: ${{ secrets.QLTY_COVERAGE_TOKEN }}
+                files: integration-core-coverage-reports/**/lcov.info
+                tag: integration:core

.github/workflows/trigger-jetbrains-tests.yml

@@ -0,0 +1,51 @@
+name: Trigger Jetbrains Plugin <-> Cline Tests
+on:
+  pull_request:
+    types: [opened, synchronize, reopened]
+permissions:
+  contents: read
+concurrency:
+  group: jetbrains-trigger-${{ github.event.number }}
+  cancel-in-progress: true
+
+jobs:
+  trigger-integration-test:
+    name: Run Tests
+    runs-on: ubuntu-latest
+    steps:
+      - name: Generate GitHub App Token
+        id: app-token
+        uses: actions/create-github-app-token@v1
+        with:
+          app-id: 1998650
+          private-key: ${{ secrets.CLINE_JETBRAINS_WORKFLOW_KEY }}
+          owner: cline
+          repositories: intellij-plugin
+
+      - name: Trigger IntelliJ Plugin Integration Test
+        run: |
+          curl -X POST \
+            -H "Authorization: Bearer ${{ steps.app-token.outputs.token }}" \
+            -H "Accept: application/vnd.github.v3+json" \
+            -H "User-Agent: cline-pr-trigger" \
+            -H "Content-Type: application/json" \
+            https://api.github.com/repos/cline/intellij-plugin/dispatches \
+            -d '{
+              "event_type": "cline-pr-check",
+              "client_payload": {
+                "pr_number": "${{ github.event.number }}",
+                "branch_name": "${{ github.head_ref }}",
+                "action": "${{ github.event.action }}",
+                "sha": "${{ github.event.pull_request.head.sha }}",
+                "pr_title": "${{ github.event.pull_request.title }}",
+                "pr_url": "${{ github.event.pull_request.html_url }}"
+              }
+            }'
+
+      - name: Log trigger details
+        run: |
+          echo "Triggered IntelliJ Plugin integration test for:"
+          echo "  PR #${{ github.event.number }}: ${{ github.event.pull_request.title }}"
+          echo "  Branch: ${{ github.head_ref }}"
+          echo "  Action: ${{ github.event.action }}"
+          echo "  SHA: ${{ github.event.pull_request.head.sha }}"

.gitignore

@@ -20,6 +20,8 @@ webview-ui/src/**/*.js.map
 
 # Ignore coverage directories and files
 coverage
+coverage-unit
+.nyc_output
 # But don't ignore the coverage scripts in .github/scripts/
 !.github/scripts/coverage/
 

.husky/pre-commit

@@ -1 +1 @@
-lint-staged --no-stash
+lint-staged

.mocharc.json

@@ -3,13 +3,13 @@
 		"ts"
 	],
 	"spec": [
-		"src/**/__tests__/*.ts",
-		"eslint-rules/__tests__/**/*.test.ts"
+		"src/**/__tests__/*.ts"
 	],
 	"require": [
 		"ts-node/register",
 		"source-map-support/register",
 		"./src/test/requires.ts"
 	],
-	"recursive": true
+	"recursive": true,
+	"exit": true
 }

.nycrc.unit.json

@@ -0,0 +1,48 @@
+{
+	"all": true,
+	"check-coverage": false,
+	"reporter": [
+		"text",
+		"lcov"
+	],
+	"include": [
+		"src/**/*.ts"
+	],
+	"exclude": [
+		"**/*.d.ts",
+
+		"**/*.{test,spec}.{js,jsx,ts,tsx,mjs,cjs}",
+		"**/__tests__/**",
+		"**/test/**",
+		"**/tests/**",
+		"**/.nyc_output/**",
+		"**/.vscode-test/**",
+		"**/tests-results/**",
+		"src/test/**",
+
+		"src/generated/**",
+
+		"**/node_modules/**",
+		"**/dist/**",
+		"**/out/**",
+		"**/build/**",
+		"**/coverage/**",
+		"**/coverage-unit/**",
+		"**/proto/**",
+
+		"**/*.{config,setup}.{js,ts,mjs,cjs}",
+		"**/vite-env.d.ts",
+
+		"**/*.{css,scss,sass,less,styl}",
+		"**/*.{svg,png,jpg,jpeg,gif,ico}",
+		"**/*.{json,yaml,yml}"
+	],
+	"extension": [
+		".ts",
+		".js"
+	],
+	"cache": true,
+	"sourceMap": true,
+	"instrument": true,
+	"report-dir": "./coverage-unit"
+}

.vscode/launch.json

@@ -12,6 +12,7 @@
 			"args": [
 				"--extensionDevelopmentPath=${workspaceFolder}",
 				"--disable-workspace-trust",
+				"--disable-extensions", // Avoid conflicts with installed extensions
 				"${workspaceFolder}"
 			],
 			"outFiles": [
@@ -31,6 +32,7 @@
 			"args": [
 				"--extensionDevelopmentPath=${workspaceFolder}",
 				"--disable-workspace-trust",
+				"--disable-extensions", // Avoid conflicts with installed extensions
 				"${workspaceFolder}"
 			],
 			"outFiles": [
@@ -50,6 +52,7 @@
 			"args": [
 				"--extensionDevelopmentPath=${workspaceFolder}",
 				"--disable-workspace-trust",
+				"--disable-extensions", // Avoid conflicts with installed extensions
 				"${workspaceFolder}"
 			],
 			"outFiles": [
@@ -71,7 +74,7 @@
 				"--user-data-dir=${workspaceFolder}/dist/tmp/user",
 				"--profile-temp",
 				"--sync=off",
-				"--disable-extensions",
+				"--disable-extensions", // Avoid conflicts with installed extensions
 				"--extensionDevelopmentPath=${workspaceFolder}",
 				"${workspaceFolder}"
 			],
@@ -91,7 +94,7 @@
 		{
 			"type": "node",
 			"request": "launch",
-			"name": "Run cline-core service",
+			"name": "Debug Test Standalone Core Api Server (test:sca-server)",
 			"skipFiles": [
 				"<node_internals>/**"
 			],
@@ -100,18 +103,62 @@
 				"${workspaceFolder}/**",
 				"!**/node_modules/**"
 			],
-			"cwd": "${workspaceFolder}/dist-standalone",
+			"cwd": "${workspaceFolder}",
 			"outFiles": [
+				"${workspaceFolder}/dist/**/*.js",
 				"${workspaceFolder}/dist-standalone/**/*.js"
 			],
 			"preLaunchTask": "compile-standalone",
+			"runtimeExecutable": "npx",
+			"runtimeArgs": [
+				"tsx"
+			],
+			"program": "scripts/test-standalone-core-api-server.ts",
 			"env": {
-				// Turns on grpc debug log.
-				//"GRPC_TRACE": "all",
-				//"GRPC_VERBOSITY": "DEBUG",
-				"NODE_PATH": "${workspaceFolder}/dist-standalone/node_modules"
+				"PROTOBUS_PORT": "26040",
+				"HOSTBRIDGE_PORT": "26041",
+				"WORKSPACE_DIR": "${workspaceFolder}",
+				"E2E_TEST": "true",
+				"CLINE_ENVIRONMENT": "local"
 			},
-			"program": "cline-core.js"
+			"console": "integratedTerminal",
+			"internalConsoleOptions": "neverOpen"
+		},
+		{
+			"type": "node",
+			"request": "launch",
+			"name": "Debug Current Test File",
+			"skipFiles": [
+				"<node_internals>/**"
+			],
+			"sourceMaps": true,
+			"resolveSourceMapLocations": [
+				"${workspaceFolder}/**",
+				"!**/node_modules/**"
+			],
+			"cwd": "${workspaceFolder}",
+			"runtimeExecutable": "npx",
+			"runtimeArgs": [
+				"mocha"
+			],
+			"args": [
+				"--require",
+				"ts-node/register",
+				"--require",
+				"source-map-support/register",
+				"--require",
+				"./src/test/requires.ts",
+				"--exit",
+				"${file}"
+			],
+			"env": {
+				"TS_NODE_PROJECT": "./tsconfig.unit-test.json",
+				"NODE_ENV": "test",
+				"IS_DEV": "true",
+				"CLINE_ENVIRONMENT": "local"
+			},
+			"console": "integratedTerminal",
+			"internalConsoleOptions": "openOnSessionStart"
 		}
 	]
 }

.vscode/settings.json

@@ -21,16 +21,11 @@
 	},
 	// Enable Lint and format using Biome
 	"biome.enabled": true,
-	"[typescript]": {
-		"editor.defaultFormatter": "biomejs.biome"
-	},
-	"[typescriptreact]": {
-		"editor.defaultFormatter": "biomejs.biome"
-	},
-	"[jsonc]": {
-		"editor.defaultFormatter": "biomejs.biome"
-	},
-	"[json]": {
-		"editor.defaultFormatter": "biomejs.biome"
+	"editor.defaultFormatter": "biomejs.biome",
+	"editor.codeActionsOnSave": {
+		"source.fixAll.biome": "explicit",
+		"source.removeUnused.biome": "always",
+		"source.removeUnusedImports": "always",
+		"source.organizeImports.biome": "always"
 	}
 }

.vscodeignore

@@ -64,3 +64,9 @@ old_docs/**
 e2e-build.mjs
 e2e.vsix
 test-results/
+
+# Ignore Storybook files
+**/*.stories.tsx
+*storybook.log
+storybook-static
+**/StorybookDecorator.tsx

CHANGELOG.md

@@ -1,5 +1,96 @@
 # Changelog
 
+## [3.30.3]
+
+- Add Oracle Code Assist provider
+
+## [3.30.2]
+
+- Fix UI tests
+
+## [3.30.1]
+
+- Fix model list not being updated in time for user to use shortcut button to update model to stealth model
+- Fix flicker issue when switching modes
+- Fix Sticky header in settings view overlaping with content on scroll
+- Add experimental yolo mode feature that disables all user approvals and automatically executes a task and navigates through plan to act mode until the task is complete
+
+## [3.30.0]
+
+- Add code-supernova stealth model
+
+## [3.29.2]
+
+- Fix: Reverted change that caused formatting issues
+- Fix: Moonshot - Pass max_tokens value to provider
+
+## [3.29.1]
+
+- Changeset bump + Announcement banner update
+
+## [3.29.0]
+
+- Updated Baseten provider to fetch models from server
+- Fix: Updated insufficient balance URL for easy Cline balance top-ups
+- Accessibility: Improvements to screen readers in MCP, Cline Rules, workflows, and history views.
+
+## [3.28.4]
+
+- Fix bug where some Windows machines had API request hanging
+- Fix bug where 'Proceed while running' action button would be disabled after running an interactive command
+- Fix prompt cache info not being displayed in History
+
+## [3.28.3]
+
+- Fixed issue with start new task button
+- Feature to generate commit message for staged changes, with unstaged as fallback
+
+## [3.28.2]
+
+- Fix for focus chain settings
+
+## [3.28.1]
+
+- Requesty: use base URL to get models and API keys
+- Removed focus chain feature flag
+
+## [3.28.0]
+
+- Synchronized Task History: Real-time task history synchronization across all Cline instances
+- Optimized GPT-5 Integration: Fine-tuned system prompts for improved performance with GPT-5 model family
+- Deep Planning Improvements: Optimized prompts for Windows/PowerShell environments and dependency exclusion
+- Streamlined UI Experience: ESC key navigation, cleaner approve/reject buttons, and improved editor panel focus
+- Smart Provider Search: Improved search functionality in API provider dropdown for faster model selection
+- Added per-provider thinking tokens configurability
+- Added Ollama custom prompt options
+- Enhanced SAP AI Core Provider: Orchestration mode support and improved model visibility
+- Added Dify.ai API Integration
+- SambaNova Updates: Added DeepSeek-V3.1 model
+- Better Gemini rate limit handling
+- OpenAI Reasoning Effort: Minimal reasoning effort configuration for OpenAI models
+- Fixed LiteLLM Caching: Anthropic caching compatibility when using LiteLLM
+- Fixed Ollama default endpoint connections
+- Fixed AutoApprove menu overflow
+- Fixed extended thinking token issue with Anthropic models
+- Fixed issue with slash commands removing text from prompt
+
+## [3.27.2]
+
+- Remove `grok-code-fast-1` promotion deadline
+
+## [3.27.1]
+
+- Add new Kimi K2 model to groq and moonshot providers
+
+## [3.27.0]
+
+- Fix `grok-code-fast-1` model information
+- Add call to action for trying free `grok-code-fast-1` in Announcement banner
+
+## [3.26.7]
+
+- Add 200k context window variant for Claude Sonnet 4 to OpenRouter and Cline providers
+
 ## [3.26.6]
 
 - Add free Grok Coder model to Cline provider for users looking for a fast, free coding model option

CONTRIBUTING.md

@@ -157,6 +157,36 @@ Anyone can contribute code to Cline, but we ask that you follow these guidelines
     - Update existing tests if your changes affect them
     - Include both unit tests and integration tests where appropriate
 
+    **End-to-End (E2E) Testing**
+    
+    Cline includes comprehensive E2E tests using Playwright that simulate real user interactions with the extension in VS Code:
+    
+    - **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
+      ```
+    
+    - **Writing E2E tests:**
+      - Tests are located in `src/test/e2e/`
+      - Use the `e2e` fixture for single-root workspace tests
+      - Use `e2eMultiRoot` fixture for multi-root workspace tests
+      - Follow existing patterns in `auth.test.ts`, `chat.test.ts`, `diff.test.ts`, and `editor.test.ts`
+      - See `src/test/e2e/README.md` for detailed documentation
+    
+    - **Debug mode features:**
+      - Interactive Playwright Inspector for step-by-step debugging
+      - Record new interactions and generate test code automatically
+      - Visual VS Code instance for manual testing
+      - Element inspection and selector validation
+    
+    - **Test environment:**
+      - Automated VS Code setup with Cline extension loaded
+      - Mock API server for backend testing
+      - Temporary workspaces with test fixtures
+      - Video recording for failed tests
+
 4. **Version Management with Changesets**
 
     - Create a changeset for any user-facing changes using `npm run changeset`

README.md

@@ -30,9 +30,9 @@ English | <a href="https://github.com/cline/cline/blob/main/locales/es/README.md
 </table>
 </div>
 
-Meet Cline (pronounced /klaɪn/, like "Klein"), an AI assistant that can use your **CLI** a**N**d **E**ditor.
+Meet Cline, an AI assistant that can use your **CLI** a**N**d **E**ditor.
 
-Thanks to [Claude 3.7 Sonnet's agentic coding capabilities](https://www.anthropic.com/claude/sonnet), Cline can handle complex software development tasks step-by-step. With tools that let him create & edit files, explore large projects, use the browser, and execute terminal commands (after you grant permission), he can assist you in ways that go beyond code completion or tech support. Cline can even use the Model Context Protocol (MCP) to create new tools and extend his own capabilities. While autonomous AI scripts traditionally run in sandboxed environments, this extension provides a human-in-the-loop GUI to approve every file change and terminal command, providing a safe and accessible way to explore the potential of agentic AI.
+Thanks to [Claude Sonnet's agentic coding capabilities](https://www.anthropic.com/claude/sonnet), Cline can handle complex software development tasks step-by-step. With tools that let him create & edit files, explore large projects, use the browser, and execute terminal commands (after you grant permission), he can assist you in ways that go beyond code completion or tech support. Cline can even use the Model Context Protocol (MCP) to create new tools and extend his own capabilities. While autonomous AI scripts traditionally run in sandboxed environments, this extension provides a human-in-the-loop GUI to approve every file change and terminal command, providing a safe and accessible way to explore the potential of agentic AI.
 
 1. Enter your task and add images to convert mockups into functional apps or fix bugs with screenshots.
 2. Cline starts by analyzing your file structure & source code ASTs, running regex searches, and reading relevant files to get up to speed in existing projects. By carefully managing what information is added to context, Cline can provide valuable assistance even for large, complex projects without overwhelming the context window.
@@ -87,7 +87,7 @@ All changes made by Cline are recorded in your file's Timeline, providing an eas
 
 ### Use the Browser
 
-With Claude 3.5 Sonnet's new [Computer Use](https://www.anthropic.com/news/3-5-models-and-computer-use) capability, Cline can launch a browser, click elements, type text, and scroll, capturing screenshots and console logs at each step. This allows for interactive debugging, end-to-end testing, and even general web use! This gives him autonomy to fixing visual bugs and runtime issues without you needing to handhold and copy-pasting error logs yourself.
+With Claude Sonnet's new [Computer Use](https://www.anthropic.com/news/3-5-models-and-computer-use) capability, Cline can launch a browser, click elements, type text, and scroll, capturing screenshots and console logs at each step. This allows for interactive debugging, end-to-end testing, and even general web use! This gives him autonomy to fixing visual bugs and runtime issues without you needing to handhold and copy-pasting error logs yourself.
 
 Try asking Cline to "test the app", and watch as he runs a command like `npm run dev`, launches your locally running dev server in a browser, and performs a series of tests to confirm that everything works. [See a demo here.](https://x.com/sdrzn/status/1850880547825823989)
 

biome.jsonc

@@ -37,7 +37,8 @@
 				"useYield": "off",
 				"noConstructorReturn": "off",
 				"noInvalidPositionAtImportRule": "off",
-				"noSwitchDeclarations": "off"
+				"noSwitchDeclarations": "off",
+				"noUnusedImports": "error"
 			},
 			"a11y": "off",
 			"style": {
@@ -122,7 +123,8 @@
 			"!**/node_modules/**",
 			"!**/webview-ui/build/**",
 			"!**/generated/**",
-			"!**/proto/**"
+			"!**/proto/**",
+			"!**/tests/specs/**"
 		]
 	},
 	"plugins": [
@@ -134,7 +136,12 @@
 				"**",
 				"!**/hosts/vscode/**",
 				"!**/test/**",
-				"!src/extension.ts"
+				"!**/*.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"
@@ -147,6 +154,7 @@
 				"!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"
 			],

docs/docs.json

@@ -60,6 +60,7 @@
 					"getting-started/what-is-cline",
 					"getting-started/model-selection-guide",
 					"getting-started/installing-cline",
+					"getting-started/installing-cline-jetbrains",
 					"getting-started/task-management",
 					"getting-started/understanding-context-management",
 					{
@@ -90,6 +91,7 @@
 					"features/focus-chain",
 					"features/auto-compact",
 					"features/editing-messages",
+					"features/dictation",
 					{
 						"group": "@ Mentions",
 						"pages": [
@@ -120,6 +122,12 @@
 							"features/commands-and-shortcuts/git-integration",
 							"features/commands-and-shortcuts/keyboard-shortcuts"
 						]
+					},
+					{
+						"group": "Customization",
+						"pages": [
+							"features/customization/opening-cline-in-sidebar"
+						]
 					}
 				]
 			},
@@ -182,7 +190,8 @@
 					"provider-config/openrouter",
 					"provider-config/sap-aicore",
 					"provider-config/vercel-ai-gateway",
-					"provider-config/requesty"
+					"provider-config/requesty",
+					"provider-config/baseten"
 				]
 			},
 			{

docs/exploring-clines-tools/cline-tools-guide.mdx

@@ -56,7 +56,7 @@ Cline is your AI assistant that can:
 
 ## Available Tools
 
-For the most up-to-date implementation details, you can view the full source code in the [Cline repository](https://github.com/cline/cline/blob/main/src/core/Cline.ts).
+For the most up-to-date implementation details, you can view the full source code in the [Cline repository](https://github.com/cline/cline/blob/main/src/core/prompts/system-prompt/tools).
 
 Cline has access to the following tools for various tasks:
 

docs/features/customization/opening-cline-in-sidebar.mdx

@@ -0,0 +1,52 @@
+---
+title: "Opening Cline in the Sidebar"
+description: "Learn how to open Cline in the right sidebar in VS Code and Cursor"
+---
+
+## VS Code
+
+To open Cline in the right sidebar:
+
+<Steps>
+  <Step title="Align Extension View">
+    Make sure your extension view is aligned vertically to the left
+  </Step>
+  <Step title="Open Right Side View">
+    Click the button that opens the right side view in VS Code (the one VS Code uses for GitHub Copilot)
+  </Step>
+  <Step title="Drag Cline Icon">
+    Drag the Cline icon over to the nav panel at the top of that right view
+  </Step>
+</Steps>
+
+<Frame>
+	<img
+		src="https://storage.googleapis.com/cline_public_images/vscode_right_view.gif"
+		alt="VS Code Right Sidebar Setup"
+	/>
+</Frame>
+
+## Cursor
+
+To open Cline in the right sidebar:
+
+<Steps>
+  <Step title="Align Extensions">
+    Make sure your extensions are [aligned vertically](https://cursor.com/docs/configuration/migrations/vscode#activity-bar-orientation) to the left
+  </Step>
+  <Step title="Open Agent Panel">
+    Click the little cube button that opens Cursor's agent (right side view panel)
+  </Step>
+  <Step title="Drag to Three Dots">
+    Drag the Cline icon specifically to the three dots - it doesn't work if you just drag it to the top, it has to be the three dots
+  </Step>
+</Steps>
+
+<Frame>
+	<img
+		src="https://storage.googleapis.com/cline_public_images/cursor-side-bar.gif"
+		alt="Cursor Right Sidebar Setup"
+	/>
+</Frame>
+
+Once set up, Cline will load on the right side and you can use it as normal.

docs/features/dictation.mdx

@@ -0,0 +1,60 @@
+---
+title: Dictation
+description:
+---
+
+Cline lets you transcribe speech to text in an easy, built-in service
+
+## Get Started
+
+1. **Enable Dictation** in Feature Settings.
+2. **Click the microphone** in the chat input area.
+3. **Speak** - the button turns red while recording.
+4. **Click Stop Recording** when done.
+5. **Wait for transcription** - "[Transcribing...]" will appear and then the finished transcription will appear.
+
+## Settings
+
+Enable or disable dictation in Feature Settings by toggling "Enable Dictation." You can also change the transcription language to one of the supported languages.
+
+## Requirements
+
+Cline uses FFmpeg to capture your voice across all platforms:
+
+-   **macOS**: FFmpeg (via Homebrew: `brew install ffmpeg`)
+-   **Linux**: FFmpeg (via apt: `sudo apt-get install ffmpeg`)
+-   **Windows**: FFmpeg (via winget: `winget install Gyan.FFmpeg`)
+
+If you don't have FFmpeg installed, Cline will automatically detect this and prompt you to install it with a single click.
+
+## Technical Details
+
+### Independent from Chat Provider
+
+The voice transcription feature works completely independently from whatever chat provider you're using. You can use Claude, GPT-4, or any other model for your conversations, and voice transcription will always use Cline's own transcription service. As long as you have a valid Cline account with enough credits, dictation will work regardless of your chat model choice.
+
+### Audio Format
+
+Voice recordings are captured in WebM format using the Opus codec for optimal compression. The system records in mono (single channel) at a 16kHz sample rate, which is specifically optimized for voice recognition. The bitrate is set to 32kbps to keep file sizes efficient while maintaining good audio quality.
+
+### Privacy & Security
+
+Your audio is recorded locally on your machine and only the audio file itself is sent to Cline's transcription service for processing. No audio is stored anywhere after transcription is complete, and all temporary files are automatically cleaned up to protect your privacy.
+
+## Troubleshooting
+
+`Failed to start recording` - Audio recording tools aren't installed. Cline will prompt you to install FFmpeg - just follow the chat instructions.
+
+`Invalid audio format or request data` - Usually an audio recording issue. Make sure FFmpeg is properly installed and working.
+
+`Authentication failed` - You need to reauthenticate your Cline account. Sign out and back in, then check your internet connection.
+
+`Insufficient credits for transcription service` - Your Cline account doesn't have enough credits. Check your balance and purchase more if needed.
+
+`Cannot connect to transcription service` - Connection issue. Check your internet connection and firewall settings aren't blocking Cline's servers.
+
+## API Usage
+
+Voice transcription uses Cline's transcription service, which requires credits from your Cline account. Currently, voice transcription is billed at $0.006 per minute of audio.
+
+**Note:** We are still experimenting with this feature and pricing may change in the future.

docs/getting-started/installing-cline-jetbrains.mdx

@@ -0,0 +1,129 @@
+---
+title: "Installing Cline for JetBrains"
+description: "Install Cline in your favorite JetBrains IDE with the same powerful AI assistance you know from VSCode."
+---
+
+<Frame>
+	<img
+		src="https://storage.googleapis.com/cline_public_images/docs/assets/jetbrains-logo.svg"
+		alt="JetBrains logo"
+		style={{ width: "200px", height: "auto", margin: "0 auto 20px auto", display: "block" }}
+	/>
+</Frame>
+
+Cline for JetBrains works almost identically to Cline in VSCode. All the core features work properly: diff editing, using tools, logging in with different providers, MCP servers, Cline rules and workflows, and more.
+
+<Frame>
+	<img
+		src="https://storage.googleapis.com/cline_public_images/docs/assets/jetbrains-demo-hifi.gif"
+		alt="Cline running in JetBrains IDE showing AI assistance"
+	/>
+</Frame>
+
+Cline is now available on the [JetBrains Marketplace](https://plugins.jetbrains.com/plugin/28247-cline). Works in IntelliJ IDEA, PyCharm, WebStorm, DataSpell, PhpStorm, and all other JetBrains IDEs.
+
+## Installation
+
+**Method 1: Direct from your IDE**
+
+1. Open your JetBrains IDE
+2. Press `Ctrl+Alt+S` (Windows/Linux) or `Cmd+,` (macOS) 
+3. Go to Plugins → Marketplace tab
+4. Search "Cline" and click Install
+5. Restart your IDE
+
+<Frame>
+	<img
+		src="https://storage.googleapis.com/cline_public_images/docs/assets/cline-jetbrains-marketplace-search.png"
+		alt="JetBrains marketplace showing Cline plugin search results"
+	/>
+</Frame>
+
+**Method 2: Browser install**
+
+Visit [plugins.jetbrains.com/plugin/28247-cline](https://plugins.jetbrains.com/plugin/28247-cline) and click the "Install to IDE" button. Your IDE will open and prompt you to install.
+
+<details>
+<summary>Method 3: Manual installation</summary>
+
+Download the plugin from the [marketplace page](https://plugins.jetbrains.com/plugin/28247-cline), then:
+
+1. Go to Settings → Plugins
+2. Click the gear icon → Install Plugin from Disk
+3. Select the downloaded `.zip` file
+4. Restart your IDE
+
+</details>
+
+## Getting Started with Cline
+
+After installation, you'll find Cline in your IDE. Look for the Cline tool window (usually on the right side) or go to **View** → **Tool Windows** → **Cline**.
+
+Sign in is optional - you can also bring your own API key. If you want to sign in, click **Sign In** in the Cline panel. You'll be taken to [app.cline.bot](https://app.cline.bot) to create your account. No credit card needed to get started with free credits.
+
+Try this first prompt: "Hey Cline! Can you help me create a simple Hello World program in this project?"
+
+## Key Differences from VSCode
+
+The terminal integration works differently in JetBrains. Unlike VSCode where terminal output streams directly to the chat, JetBrains shows command output in a collapsible section. Commands still execute successfully - you just need to expand the **Command Output** section to see results.
+
+## What Works
+
+Everything else works exactly like VSCode. Cline can read, write, and edit files with the same precision. All his tools work identically - file operations, web browsing, you name it. 
+
+You get full support for:
+- API providers (Anthropic, OpenAI, local models)
+- MCP servers and custom tools
+- Cline rules and workflows
+- @ mentions for files, folders, and problems
+- Drag & drop for files and images
+
+## Tips for JetBrains Users
+
+Cline automatically understands your project structure, just like in VSCode. He works with any language your JetBrains IDE supports - Java, Python, JavaScript, Go, whatever you're building.
+
+I find it helpful to share error messages and stack traces directly in the chat when debugging. You can also ask him to review your code changes before committing.
+
+## Troubleshooting
+
+### Plugin Installation Issues
+
+If you can't find Cline in the marketplace:
+- Make sure you're searching in the **Marketplace** tab (not Installed)
+- Try searching for "Cline AI" or just "Cline"
+- Check that your IDE version is compatible (2023.1 or later recommended)
+
+If installation fails:
+- Restart your IDE and try again
+- Check your internet connection
+- Try installing from disk as an alternative
+
+### Plugin Not Appearing
+
+If you don't see the Cline tool window after installation:
+- Restart your IDE completely (File → Exit and reopen)
+- Check **View** → **Tool Windows** → **Cline**
+- Verify the plugin is enabled in **Settings** → **Plugins** → **Installed** tab
+- Look for the Cline icon in your IDE's tool window bar (usually on the right side)
+
+### Common Issues
+
+**Plugin appears to be installed but doesn't work:**
+- Ensure you've restarted your IDE after installation
+- Check if there are any error messages in the IDE's event log
+- Try disabling and re-enabling the plugin in Settings
+
+**Performance or compatibility issues:**
+- Make sure you're using a supported JetBrains IDE version
+- Check for IDE updates that might improve compatibility
+- Consider allocating more memory to your IDE if needed
+
+Having other issues? Join our [Discord community](https://discord.gg/cline) for help from the team and other users.
+
+## Next Steps
+
+Now that you have Cline installed, you might want to:
+- Learn about [model selection](/getting-started/model-selection-guide) to choose the best AI provider
+- Explore [@ mentions](/features/at-mentions/overview) to reference files and context efficiently  
+- Set up [Cline rules](/features/cline-rules) for your specific workflow
+- Try [MCP servers](/mcp/mcp-overview) to extend Cline's capabilities

docs/getting-started/installing-cline.mdx

@@ -1,13 +1,17 @@
 ---
 title: "Installing Cline"
-description: "Cline is a VS Code extension that brings AI-powered coding assistance directly
-    to your editor. Install using one of these methods:"
+description: "Cline brings AI-powered coding assistance to your editor. Available for VS Code and JetBrains IDEs."
 ---
 
-### Installation Options
+## Choose Your Editor
 
--   **VS Code Marketplace (Recommended):** Fastest method for standard VS Code and Cursor users.
--   **Open VSX Registry:** For VS Code-compatible editors like VSCodium.
+Cline works across multiple development environments:
+
+-   **VS Code/Cursor:** Install from VS Code Marketplace (most popular)
+-   **JetBrains IDEs:** Install from JetBrains Marketplace - works in IntelliJ IDEA, PyCharm, WebStorm, and more
+-   **VSCodium/Windsurf:** Install from Open VSX Registry
+
+## VS Code Installation
 
 ### VS Code Marketplace: Step-by-Step Setup
 
@@ -46,6 +50,18 @@ For VS Code-compatible editors without Marketplace access (like VSCodium and Win
 4. Select "Cline" by saoudrizwan and click **Install**.
 5. Reload if prompted.
 
+## JetBrains Installation
+
+For IntelliJ IDEA, PyCharm, WebStorm, DataSpell, and other JetBrains IDEs:
+
+1. Open your JetBrains IDE
+2. Press `Ctrl+Alt+S` (Windows/Linux) or `Cmd+,` (macOS) to open Settings
+3. Go to Plugins → Marketplace tab
+4. Search "Cline" and click Install
+5. Restart your IDE
+
+**Need more help?** See our [complete JetBrains installation guide](/getting-started/installing-cline-jetbrains) for screenshots and troubleshooting.
+
 ### Creating Your Cline Account
 
 Now that you have Cline installed, let's get you set up with your account:

docs/hubspot.js

@@ -0,0 +1,14 @@
+// HubSpot Tracking Code for Cline Documentation
+;(() => {
+	// Check if HubSpot script is already loaded to prevent duplicates
+	if (!document.getElementById("hs-script-loader")) {
+		var script = document.createElement("script")
+		script.type = "text/javascript"
+		script.id = "hs-script-loader"
+		script.async = true
+		script.src = "https://js-na2.hs-scripts.com/243656267.js"
+
+		// Append the script to the document head
+		document.head.appendChild(script)
+	}
+})()

docs/package-lock.json

@@ -1386,7 +1386,7 @@
 			"integrity": "sha512-/uR4hAwpcJW9+zbmZL48kKFnWLkOxhIqoGWvZzjg0CniVhR4emtQJAps80WqLAhz0iJgCQxg/axtA7leaznDzQ==",
 			"license": "Elastic-2.0",
 			"dependencies": {
-				"axios": "^1.8.3",
+				"axios": "^1.12.0",
 				"openapi-types": "^12.0.0"
 			},
 			"engines": {
@@ -2688,9 +2688,9 @@
 			}
 		},
 		"node_modules/axios": {
-			"version": "1.11.0",
-			"resolved": "https://registry.npmjs.org/axios/-/axios-1.11.0.tgz",
-			"integrity": "sha512-1Lx3WLFQWm3ooKDYZD1eXmoGO9fxYQjrycfHFC8P0sCfQVXyROp0p9PFWBehewBOdCwHc+f/b8I0fMto5eSfwA==",
+			"version": "1.12.2",
+			"resolved": "https://registry.npmjs.org/axios/-/axios-1.12.2.tgz",
+			"integrity": "sha512-vMJzPewAlRyOgxV2dU0Cuz2O8zzzx9VYtbJOaBgXFeLc4IV/Eg50n4LowmehOOR61S8ZMpc2K5Sa7g6A4jfkUw==",
 			"license": "MIT",
 			"dependencies": {
 				"follow-redirects": "^1.15.6",

docs/provider-config/baseten.mdx

@@ -0,0 +1,115 @@
+---
+title: "Baseten"
+description: "Learn how to configure and use Baseten's Model APIs with Cline. Access frontier open-source models with enterprise-grade performance, reliability, and competitive pricing."
+---
+
+Baseten provides on-demand frontier model APIs designed for production applications, not just experimentation. Built on the Baseten Inference Stack, these APIs deliver enterprise-grade performance and reliability with optimized inference for leading open-source models from OpenAI, DeepSeek, Meta, Moonshot AI, and Alibaba Cloud.
+
+**Website:** [https://www.baseten.co/products/model-apis/](https://www.baseten.co/products/model-apis/)
+
+### Getting an API Key
+
+1.  **Sign Up/Sign In:** Go to [Baseten](https://www.baseten.co/) and create an account or sign in.
+2.  **Navigate to API Keys:** Access your dashboard and go to the API Keys section.
+3.  **Create a Key:** Generate a new API key. Give it a descriptive name (e.g., "Cline").
+4.  **Copy the Key:** Copy the API key immediately and store it securely.
+
+### Supported Models
+
+Cline supports all current models under Baseten Model APIs, including:
+For the most updated pricing, please visit: https://www.baseten.co/products/model-apis/
+
+**Reasoning Models:**
+-   `deepseek-ai/DeepSeek-R1` - DeepSeek's first-generation reasoning model (163K context) - \$2.55/\$5.95 per 1M tokens
+-   `deepseek-ai/DeepSeek-R1-0528` - Latest revision of DeepSeek's reasoning model (163K context) - \$2.55/\$5.95 per 1M tokens
+-   `deepseek-ai/DeepSeek-V3.1` - Hybrid reasoning with advanced tool calling (163K context) - \$0.50/\$1.50 per 1M tokens
+-   `deepseek-ai/DeepSeek-V3-0324` - Fast general-purpose with enhanced reasoning (163K context) - \$0.77/\$0.77 per 1M tokens
+
+**Flagship Models:**
+-   `openai/gpt-oss-120b` (OpenAI) - 120B MoE with strong reasoning capabilities (128K context) - \$0.10/\$0.50 per 1M tokens
+-   `moonshotai/Kimi-K2-Instruct` (Moonshot AI) - 1 trillion parameter model for agentic tasks (131K context) - \$0.60/\$2.50 per 1M tokens
+-   `moonshotai/Kimi-K2-Instruct-0905` (Moonshot AI) - September update with enhanced capabilities (262K context) - \$0.60/\$2.50 per 1M tokens
+
+**Meta Llama 4 Series:**
+-   `meta-llama/Llama-4-Maverick-17B-128E-Instruct` - High-efficiency processing (1M context!) - \$0.19/\$0.72 per 1M tokens
+-   `meta-llama/Llama-4-Scout-17B-16E-Instruct` - Precise context understanding (1M context!) - \$0.13/\$0.50 per 1M tokens
+
+**Coding Specialists:**
+-   `Qwen/Qwen3-Coder-480B-A35B-Instruct`- Advanced coding and reasoning (262K context) - \$0.38/\$1.53 per 1M tokens
+-   `Qwen/Qwen3-235B-A22B-Instruct-2507` - Math and reasoning expert (262K context) - \$0.22/\$0.80 per 1M tokens
+
+### Configuration in Cline
+
+1.  **Open Cline Settings:** Click the settings icon (⚙️) in the Cline panel.
+2.  **Select Provider:** Choose "Baseten" from the "API Provider" dropdown.
+3.  **Enter API Key:** Paste your Baseten API key into the "Baseten API Key" field.
+4.  **Select Model:** Choose your desired model from the "Model" dropdown.
+
+### Production-First Architecture
+
+Baseten's Model APIs are built for production environments with several key advantages:
+
+#### Enterprise-Grade Reliability
+- **Four nines of uptime** (99.99%) through active-active redundancy
+- **Cloud-agnostic, multi-cluster autoscaling** for consistent availability
+- **SOC 2 Type II certified** and **HIPAA compliant** for security requirements
+
+#### Optimized Performance
+- **Pre-optimized models** shipped with the Baseten Inference Stack
+- **Latest-generation GPUs** with multi-cloud infrastructure
+- **Ultra-fast inference** optimized from the bottom up for production workloads
+
+#### Cost Efficiency
+- **5-10x less expensive** than closed alternatives
+- **Optimized multi-cloud infrastructure** for efficient resource utilization
+- **Transparent pricing** with no hidden costs or rate limit surprises
+
+#### Developer Experience
+- **OpenAI compatible API** - migrate by swapping a single URL
+- **Drop-in replacement** for closed models with comprehensive observability
+- **Seamless scaling** from Model APIs to dedicated deployments
+
+### Special Features
+
+#### Function Calling & Tool Use
+All Baseten models support structured outputs, function calling, and tool use as part of the Baseten Inference Stack, making them ideal for agentic applications.
+
+#### Reasoning Capabilities
+DeepSeek models offer enhanced reasoning with step-by-step thought processes, while maintaining production-ready performance.
+
+#### Long Context Support
+- **Up to 1 million tokens** for Llama 4 models (Maverick and Scout)
+- **262K tokens** for Qwen3 models
+- **163K tokens** for DeepSeek models
+- **Perfect for code repositories** and complex multi-turn conversations
+
+#### Quantization Optimizations
+Models are deployed with advanced quantization techniques (fp4, fp8, fp16) for optimal performance while maintaining quality.
+
+### Migration from Other Providers
+
+Baseten's OpenAI compatibility makes migration straightforward:
+
+**From OpenAI:**
+- Swap `api.openai.com` with `inference.baseten.co/v1`
+- Keep existing request/response formats
+- Benefit from significant cost savings
+
+**From Other Providers:**
+- Use standard OpenAI SDK format
+- Maintain existing prompting strategies
+- Access to newer open-source models
+
+### Tips and Notes
+
+-   **Model Selection:** Choose models based on your specific use case - reasoning models for complex tasks, coding models for development work, and flagship models for general applications.
+-   **Cost Optimization:** Baseten offers some of the most competitive pricing in the market, especially for open-source models.
+-   **Context Windows:** Take advantage of large context windows (up to 1M tokens) for including substantial codebases and documentation.
+-   **Enterprise Ready:** Baseten is designed for production use with enterprise-grade security, compliance, and reliability.
+-   **Dynamic Model Updates:** Cline automatically fetches the latest model list from Baseten, ensuring access to new models as they're released.
+-   **Multi-Cloud Capacity Management (MCM):** Baseten's multi-cloud infrastructure ensures high availability and low latency globally.
+-   **Support:** Baseten provides dedicated support for production deployments and can work with you on dedicated resources as you scale.
+
+### Pricing Information
+
+Current pricing is highly competitive and transparent. For the most up-to-date pricing, visit the [Baseten Model APIs page](https://www.baseten.co/products/model-apis/). Prices typically range from \$0.10-\$6.00 per million tokens, making Baseten significantly more cost-effective than many closed-model alternatives while providing access to state-of-the-art open-source models.

docs/provider-config/sap-aicore.mdx

@@ -7,12 +7,13 @@ SAP AI Core and the generative AI hub help you to integrate LLMs and AI into new
 
 **Website:** [SAP Help Portal](https://help.sap.com/docs/sap-ai-core/sap-ai-core-service-guide/what-is-sap-ai-core)
 
-### Getting a Service Binding
 
 > 💡 **Information**
 >
 > SAP AI Core, and Generative AI Hub, are offerings from SAP BTP.
-> You need an active SAP BTP contract and a existing subaccount with a SAP AI Core instance to perform these steps.
+> You need an active SAP BTP contract and a existing subaccount with a SAP AI Core instance with the `extended` service plan (For more details about SAP AI Core service plans and their capabilities, see the [Service Plans documentation](https://help.sap.com/docs/sap-ai-core/sap-ai-core-service-guide/service-plans)) to perform these steps.
+
+### Getting a Service Binding
 
 1. **Access:** Go to your subaccount via [BTP Cloud Cockpit](cockpit.btp.cloud.sap/cockpit)
 2. **Create a Service Binding:** Go to "Instances and Subscriptions", select your SAP AI Core service instance and click on Service Bindings > Create.
@@ -32,14 +33,30 @@ Refer to the [Generative AI Hub Supported Models page](https://me.sap.com/notes/
 5.  **Enter Base URL:** Add the `.serviceurls.AI_API_URL` field from the service binding into the "AI Core Base URL" field.
 6.  **Enter Auth URL:** Add the `.url` field from the service binding into the "AI Core Auth URL" field.
 7.  **Enter Resource Group:** Add the resource group where you have your model deployments. See [Create a Deployment for a Generative AI Model](https://help.sap.com/docs/sap-ai-core/sap-ai-core-service-guide/create-deployment-for-generative-ai-model-in-sap-ai-core).
-8.  **Select Model:** Choose your desired model from the "Model" dropdown.
+8.  **Configure Orchestration Mode:** If you have an `extended` service plan, the "Orchestration Mode" checkbox will automatically appear.
+9.  **Select Model:** Choose your desired model from the "Model" dropdown.
+
+### Orchestration Mode vs Native API
+
+**Orchestration Mode:**
+- **Simplified usage:** Provides access to all available models without requiring individual deployments using the [Harmonized API](https://help.sap.com/docs/sap-ai-core/sap-ai-core-service-guide/harmonized-api)
+
+**Native API Mode:**
+- **Manual deployments:** Requires manual model deployment and management in your SAP AI Core service instance
 
 ### Tips and Notes
 
--   **Model Selection:** The model dropdown displays models in two separate lists:
-    -   **Deployed Models:** These models are already deployed in your specified resource group and are ready to use immediately.
-    -   **Not Deployed Models:** These models don't have active deployments in your specified resource group. You won't be able to use these models until you create deployments for them in SAP AI Core.
--   **Creating Deployments:** To use a not deployed model, you'll need to create a deployment in your resource group in sap ai core service instance. See [Create a Deployment for a Generative AI Model](https://help.sap.com/docs/sap-ai-core/sap-ai-core-service-guide/create-deployment-for-generative-ai-model-in-sap-ai-core) for instructions.
+- **Service Plan Requirement:** You must have the SAP AI Core `extended` service plan to use LLMs with Cline. Other service plans do not provide access to Generative AI Hub.
+
+- **Orchestration Mode (Recommended):** Keep Orchestration Mode enabled for the simplest setup. It provides automatic access to all available models without requiring manual deployments.
+
+- **Native API Mode:** Only disable Orchestration Mode if you have specific requirements that necessitate direct AI Core API access or need features not supported by the orchestration mode.
+
+- **When using Native API Mode:**
+    -   **Model Selection:** The model dropdown displays models in two separate lists:
+        -   **Deployed Models:** These models are already deployed in your specified resource group and are ready to use immediately.
+        -   **Not Deployed Models:** These models don't have active deployments in your specified resource group. You won't be able to use these models until you create deployments for them in SAP AI Core.
+    -   **Creating Deployments:** To use a model that has not been deployed yet, you'll need to create a deployment in your SAP AI Core service instance. See [Create a Deployment for a Generative AI Model](https://help.sap.com/docs/sap-ai-core/sap-ai-core-service-guide/create-deployment-for-generative-ai-model-in-sap-ai-core) for instructions.
 
 #### Configuring Reasoning Effort for OpenAI Models
 

docs/provider-config/zai.mdx

@@ -49,6 +49,44 @@ All models feature:
 4.  **Enter API Key:** Paste your Z AI API key into the "Z AI API Key" field.
 5.  **Select Model:** Choose your desired model from the "Model" dropdown.
 
+### GLM Coding Plans
+
+Z AI offers subscription plans specifically designed for coding applications. These plans provide cost-effective access to GLM-4.5 models through a prompt-based structure rather than traditional API usage billing.
+
+#### Plan Options
+
+**GLM Coding Lite** - $3/month
+- 120 prompts per 5-hour cycle
+- Access to GLM-4.5 model
+- Works exclusively through coding tools like Cline
+
+**GLM Coding Pro** - $15/month  
+- 600 prompts per 5-hour cycle
+- Access to GLM-4.5 model
+- Works exclusively through coding tools like Cline
+
+Both plans offer promotional pricing for the first month: Lite drops from \$6 to \$3, Pro drops from \$30 to \$15.
+
+<Frame>
+  <img src="https://storage.googleapis.com/cline_public_images/docs/assets/zAI-coding-plan.png" alt="zAI subscription page showing GLM Coding Lite and Pro plans with pricing" />
+</Frame>
+
+#### Setting up GLM Coding Plans
+
+To use the GLM Coding Plans with Cline:
+
+1. **Subscribe:** Go to [https://z.ai/subscribe](https://z.ai/subscribe) and choose your plan.
+
+2. **Create API Key:** After subscribing, log into your zAI dashboard and create an API key for your coding plan.
+
+3. **Configure in Cline:** Open Cline settings, select "Z AI" as your provider, and paste your API key into the "Z AI API Key" field.
+
+<Frame>
+  <img src="https://storage.googleapis.com/cline_public_images/docs/assets/zAI-provider.png" alt="Cline settings with zAI provider selected and API key field highlighted" />
+</Frame>
+
+The setup connects your subscription directly to Cline, giving you access to GLM-4.5's tool-calling capabilities optimized for coding workflows.
+
 ### Z AI's Hybrid Intelligence
 
 Z AI's GLM-4.5 series introduces revolutionary capabilities that set it apart from conventional language models:

esbuild.mjs

@@ -6,7 +6,7 @@ import * as esbuild from "esbuild"
 const __filename = fileURLToPath(import.meta.url)
 const __dirname = path.dirname(__filename)
 
-const production = process.argv.includes("--production")
+const production = process.argv.includes("--production") || process.env["IS_DEBUG_BUILD"] === "false"
 const watch = process.argv.includes("--watch")
 const standalone = process.argv.includes("--standalone")
 const e2eBuild = process.argv.includes("--e2e-build")
@@ -123,15 +123,29 @@ const copyWasmFiles = {
 	},
 }
 
+const buildEnvVars = { "import.meta.url": "_importMetaUrl" }
+if (production) {
+	// IS_DEV is always disable in production builds.
+	buildEnvVars["process.env.IS_DEV"] = "false"
+}
+// Set the environment and telemetry env vars. The API key env vars need to be populated in the GitHub
+// workflows from the secrets.
+if (process.env.CLINE_ENVIRONMENT) {
+	buildEnvVars["process.env.CLINE_ENVIRONMENT"] = JSON.stringify(process.env.CLINE_ENVIRONMENT)
+}
+if (process.env.TELEMETRY_SERVICE_API_KEY) {
+	buildEnvVars["process.env.TELEMETRY_SERVICE_API_KEY"] = JSON.stringify(process.env.TELEMETRY_SERVICE_API_KEY)
+}
+if (process.env.ERROR_SERVICE_API_KEY) {
+	buildEnvVars["process.env.ERROR_SERVICE_API_KEY"] = JSON.stringify(process.env.ERROR_SERVICE_API_KEY)
+}
 // Base configuration shared between extension and standalone builds
 const baseConfig = {
 	bundle: true,
 	minify: production,
 	sourcemap: !production,
 	logLevel: "silent",
-	define: production
-		? { "import.meta.url": "_importMetaUrl", "process.env.IS_DEV": JSON.stringify(!production) }
-		: { "import.meta.url": "_importMetaUrl" },
+	define: buildEnvVars,
 	tsconfig: path.resolve(__dirname, "tsconfig.json"),
 	plugins: [
 		copyWasmFiles,
@@ -160,9 +174,9 @@ const standaloneConfig = {
 	...baseConfig,
 	entryPoints: ["src/standalone/cline-core.ts"],
 	outfile: `${destDir}/cline-core.js`,
-	// These gRPC protos need to load files from the module directory at runtime,
+	// These modules need to load files from the module directory at runtime,
 	// so they cannot be bundled.
-	external: ["vscode", "@grpc/reflection", "grpc-health-check"],
+	external: ["vscode", "@grpc/reflection", "grpc-health-check", "better-sqlite3"],
 }
 
 // E2E build script configuration

evals/diff-edits/types.ts

@@ -1,5 +1,6 @@
 import { Anthropic } from "@anthropic-ai/sdk"
-import { ToolUseName, ToolParamName } from "../../src/core/assistant-message"
+import { ToolParamName } from "../../src/core/assistant-message"
+import { ClineDefaultTool } from "../../src/shared/tools"
 
 export interface InputMessage {
 	role: "user" | "assistant"
@@ -88,7 +89,7 @@ export interface TestResult {
 }
 
 export interface ExtractedToolCall {
-	name: ToolUseName
+	name: ClineDefaultTool
 	input: Partial<Record<ToolParamName, string>>
 }
 

evals/package-lock.json

@@ -9,9 +9,9 @@
       "version": "0.1.0",
       "license": "MIT",
       "dependencies": {
-        "axios": "^1.8.2",
+        "axios": "^1.12.0",
         "better-sqlite3": "^11.10.0",
-        "chalk": "^4.1.2",
+        "chalk": "5.6.2",
         "commander": "^9.4.1",
         "dotenv": "^16.5.0",
         "execa": "^5.1.1",
@@ -200,12 +200,13 @@
       "integrity": "sha512-Oei9OH4tRh0YqU3GxhX79dM/mwVgvbZJaSNaRk+bshkj0S5cfHcgYakreBjrHwatXKbz+IoIdYLxrKim2MjW0Q=="
     },
     "node_modules/axios": {
-      "version": "1.10.0",
-      "resolved": "https://registry.npmjs.org/axios/-/axios-1.10.0.tgz",
-      "integrity": "sha512-/1xYAC4MP/HEG+3duIhFr4ZQXR4sQXOIe+o6sdqzeykGLx6Upp/1p8MHqhINOvGeP7xyNHe7tsiJByc4SSVUxw==",
+      "version": "1.12.2",
+      "resolved": "https://registry.npmjs.org/axios/-/axios-1.12.2.tgz",
+      "integrity": "sha512-vMJzPewAlRyOgxV2dU0Cuz2O8zzzx9VYtbJOaBgXFeLc4IV/Eg50n4LowmehOOR61S8ZMpc2K5Sa7g6A4jfkUw==",
+      "license": "MIT",
       "dependencies": {
         "follow-redirects": "^1.15.6",
-        "form-data": "^4.0.0",
+        "form-data": "^4.0.4",
         "proxy-from-env": "^1.1.0"
       }
     },
@@ -292,15 +293,12 @@
       }
     },
     "node_modules/chalk": {
-      "version": "4.1.2",
-      "resolved": "https://registry.npmjs.org/chalk/-/chalk-4.1.2.tgz",
-      "integrity": "sha512-oKnbhFyRIXpUuez8iBMmyEa4nbj4IOQyuhc/wy9kY7/WVPcwIO9VA668Pu8RkO7+0G76SLROeyw9CpQ061i4mA==",
-      "dependencies": {
-        "ansi-styles": "^4.1.0",
-        "supports-color": "^7.1.0"
-      },
+      "version": "5.6.2",
+      "resolved": "https://registry.npmjs.org/chalk/-/chalk-5.6.2.tgz",
+      "integrity": "sha512-7NzBL0rN6fMUW+f7A6Io4h40qQlG+xGmtMxfbnH/K7TAtt8JQWVQK+6g0UXKMeVJoyV5EkkNsErQ8pVD3bLHbA==",
+      "license": "MIT",
       "engines": {
-        "node": ">=10"
+        "node": "^12.17.0 || ^14.13 || >=16.0.0"
       },
       "funding": {
         "url": "https://github.com/chalk/chalk?sponsor=1"
@@ -607,9 +605,10 @@
       }
     },
     "node_modules/form-data": {
-      "version": "4.0.3",
-      "resolved": "https://registry.npmjs.org/form-data/-/form-data-4.0.3.tgz",
-      "integrity": "sha512-qsITQPfmvMOSAdeyZ+12I1c+CKSstAFAwu+97zrnWAbIr5u8wfsExUzCesVLC8NgHuRUqNN4Zy6UPWUTRGslcA==",
+      "version": "4.0.4",
+      "resolved": "https://registry.npmjs.org/form-data/-/form-data-4.0.4.tgz",
+      "integrity": "sha512-KrGhL9Q4zjj0kiUt5OO4Mr/A/jlI2jDYs5eHBpYHPcBEVSiipAvn2Ko2HnPe20rmcuuvMHNdZFp+4IlGTMF0Ow==",
+      "license": "MIT",
       "dependencies": {
         "asynckit": "^0.4.0",
         "combined-stream": "^1.0.8",
@@ -708,6 +707,7 @@
       "version": "4.0.0",
       "resolved": "https://registry.npmjs.org/has-flag/-/has-flag-4.0.0.tgz",
       "integrity": "sha512-EykJT/Q1KjTWctppgIAgfSO0tKVuZUjhgMr17kqTumMl6Afv3EISleU7qZUzoXDFTAHTDC4NOoG/ZxU3EvlMPQ==",
+      "license": "MIT",
       "engines": {
         "node": ">=8"
       }
@@ -843,6 +843,22 @@
         "url": "https://github.com/sponsors/sindresorhus"
       }
     },
+    "node_modules/log-symbols/node_modules/chalk": {
+      "version": "4.1.2",
+      "resolved": "https://registry.npmjs.org/chalk/-/chalk-4.1.2.tgz",
+      "integrity": "sha512-oKnbhFyRIXpUuez8iBMmyEa4nbj4IOQyuhc/wy9kY7/WVPcwIO9VA668Pu8RkO7+0G76SLROeyw9CpQ061i4mA==",
+      "license": "MIT",
+      "dependencies": {
+        "ansi-styles": "^4.1.0",
+        "supports-color": "^7.1.0"
+      },
+      "engines": {
+        "node": ">=10"
+      },
+      "funding": {
+        "url": "https://github.com/chalk/chalk?sponsor=1"
+      }
+    },
     "node_modules/make-error": {
       "version": "1.3.6",
       "resolved": "https://registry.npmjs.org/make-error/-/make-error-1.3.6.tgz",
@@ -1003,6 +1019,22 @@
         "url": "https://github.com/sponsors/sindresorhus"
       }
     },
+    "node_modules/ora/node_modules/chalk": {
+      "version": "4.1.2",
+      "resolved": "https://registry.npmjs.org/chalk/-/chalk-4.1.2.tgz",
+      "integrity": "sha512-oKnbhFyRIXpUuez8iBMmyEa4nbj4IOQyuhc/wy9kY7/WVPcwIO9VA668Pu8RkO7+0G76SLROeyw9CpQ061i4mA==",
+      "license": "MIT",
+      "dependencies": {
+        "ansi-styles": "^4.1.0",
+        "supports-color": "^7.1.0"
+      },
+      "engines": {
+        "node": ">=10"
+      },
+      "funding": {
+        "url": "https://github.com/chalk/chalk?sponsor=1"
+      }
+    },
     "node_modules/path-key": {
       "version": "3.1.1",
       "resolved": "https://registry.npmjs.org/path-key/-/path-key-3.1.1.tgz",
@@ -1251,6 +1283,7 @@
       "version": "7.2.0",
       "resolved": "https://registry.npmjs.org/supports-color/-/supports-color-7.2.0.tgz",
       "integrity": "sha512-qpCAvRl9stuOHveKsn7HncJRvv501qIacKzQlO/+Lwxc9+0q2wLyv4Dfvt80/DPn2pqOBsJdDiogXGR9+OvwRw==",
+      "license": "MIT",
       "dependencies": {
         "has-flag": "^4.0.0"
       },
@@ -1635,12 +1668,12 @@
       "integrity": "sha512-Oei9OH4tRh0YqU3GxhX79dM/mwVgvbZJaSNaRk+bshkj0S5cfHcgYakreBjrHwatXKbz+IoIdYLxrKim2MjW0Q=="
     },
     "axios": {
-      "version": "1.10.0",
-      "resolved": "https://registry.npmjs.org/axios/-/axios-1.10.0.tgz",
-      "integrity": "sha512-/1xYAC4MP/HEG+3duIhFr4ZQXR4sQXOIe+o6sdqzeykGLx6Upp/1p8MHqhINOvGeP7xyNHe7tsiJByc4SSVUxw==",
+      "version": "1.12.2",
+      "resolved": "https://registry.npmjs.org/axios/-/axios-1.12.2.tgz",
+      "integrity": "sha512-vMJzPewAlRyOgxV2dU0Cuz2O8zzzx9VYtbJOaBgXFeLc4IV/Eg50n4LowmehOOR61S8ZMpc2K5Sa7g6A4jfkUw==",
       "requires": {
         "follow-redirects": "^1.15.6",
-        "form-data": "^4.0.0",
+        "form-data": "^4.0.4",
         "proxy-from-env": "^1.1.0"
       }
     },
@@ -1695,13 +1728,9 @@
       }
     },
     "chalk": {
-      "version": "4.1.2",
-      "resolved": "https://registry.npmjs.org/chalk/-/chalk-4.1.2.tgz",
-      "integrity": "sha512-oKnbhFyRIXpUuez8iBMmyEa4nbj4IOQyuhc/wy9kY7/WVPcwIO9VA668Pu8RkO7+0G76SLROeyw9CpQ061i4mA==",
-      "requires": {
-        "ansi-styles": "^4.1.0",
-        "supports-color": "^7.1.0"
-      }
+      "version": "5.6.2",
+      "resolved": "https://registry.npmjs.org/chalk/-/chalk-5.6.2.tgz",
+      "integrity": "sha512-7NzBL0rN6fMUW+f7A6Io4h40qQlG+xGmtMxfbnH/K7TAtt8JQWVQK+6g0UXKMeVJoyV5EkkNsErQ8pVD3bLHbA=="
     },
     "chownr": {
       "version": "1.1.4",
@@ -1909,9 +1938,9 @@
       "integrity": "sha512-gew4GsXizNgdoRyqmyfMHyAmXsZDk6mHkSxZFCzW9gwlbtOW44CDtYavM+y+72qD/Vq2l550kMF52DT8fOLJqQ=="
     },
     "form-data": {
-      "version": "4.0.3",
-      "resolved": "https://registry.npmjs.org/form-data/-/form-data-4.0.3.tgz",
-      "integrity": "sha512-qsITQPfmvMOSAdeyZ+12I1c+CKSstAFAwu+97zrnWAbIr5u8wfsExUzCesVLC8NgHuRUqNN4Zy6UPWUTRGslcA==",
+      "version": "4.0.4",
+      "resolved": "https://registry.npmjs.org/form-data/-/form-data-4.0.4.tgz",
+      "integrity": "sha512-KrGhL9Q4zjj0kiUt5OO4Mr/A/jlI2jDYs5eHBpYHPcBEVSiipAvn2Ko2HnPe20rmcuuvMHNdZFp+4IlGTMF0Ow==",
       "requires": {
         "asynckit": "^0.4.0",
         "combined-stream": "^1.0.8",
@@ -2054,6 +2083,17 @@
       "requires": {
         "chalk": "^4.1.0",
         "is-unicode-supported": "^0.1.0"
+      },
+      "dependencies": {
+        "chalk": {
+          "version": "4.1.2",
+          "resolved": "https://registry.npmjs.org/chalk/-/chalk-4.1.2.tgz",
+          "integrity": "sha512-oKnbhFyRIXpUuez8iBMmyEa4nbj4IOQyuhc/wy9kY7/WVPcwIO9VA668Pu8RkO7+0G76SLROeyw9CpQ061i4mA==",
+          "requires": {
+            "ansi-styles": "^4.1.0",
+            "supports-color": "^7.1.0"
+          }
+        }
       }
     },
     "make-error": {
@@ -2164,6 +2204,17 @@
         "log-symbols": "^4.1.0",
         "strip-ansi": "^6.0.0",
         "wcwidth": "^1.0.1"
+      },
+      "dependencies": {
+        "chalk": {
+          "version": "4.1.2",
+          "resolved": "https://registry.npmjs.org/chalk/-/chalk-4.1.2.tgz",
+          "integrity": "sha512-oKnbhFyRIXpUuez8iBMmyEa4nbj4IOQyuhc/wy9kY7/WVPcwIO9VA668Pu8RkO7+0G76SLROeyw9CpQ061i4mA==",
+          "requires": {
+            "ansi-styles": "^4.1.0",
+            "supports-color": "^7.1.0"
+          }
+        }
       }
     },
     "path-key": {

evals/package.json

@@ -19,9 +19,9 @@
   "author": "",
   "license": "MIT",
   "dependencies": {
-    "axios": "^1.8.2",
+    "axios": "^1.12.0",
     "better-sqlite3": "^11.10.0",
-    "chalk": "^4.1.2",
+    "chalk": "5.6.2",
     "dotenv": "^16.5.0",
     "commander": "^9.4.1",
     "execa": "^5.1.1",

package.json

@@ -2,7 +2,7 @@
 	"name": "claude-dev",
 	"displayName": "Cline",
 	"description": "Autonomous coding agent right in your IDE, capable of creating/editing files, running commands, using the browser, and more with your permission every step of the way.",
-	"version": "3.26.6",
+	"version": "3.30.3",
 	"icon": "assets/icons/icon.png",
 	"engines": {
 		"vscode": "^1.84.0"
@@ -102,15 +102,8 @@
 			"activitybar": [
 				{
 					"id": "claude-dev-ActivityBar",
-					"title": "Cline (⌘+')",
-					"icon": "assets/icons/icon.svg",
-					"when": "isMac"
-				},
-				{
-					"id": "claude-dev-ActivityBar",
-					"title": "Cline (Ctrl+')",
-					"icon": "assets/icons/icon.svg",
-					"when": "!isMac"
+					"title": "Cline",
+					"icon": "assets/icons/icon.svg"
 				}
 			]
 		},
@@ -207,6 +200,11 @@
 				"command": "cline.openWalkthrough",
 				"title": "Open Walkthrough",
 				"category": "Cline"
+			},
+			{
+				"command": "cline.reconstructTaskHistory",
+				"title": "Reconstruct Task History",
+				"category": "Cline"
 			}
 		],
 		"keybindings": [
@@ -348,7 +346,9 @@
 		"package": "npm run check-types && npm run build:webview && npm run lint && node esbuild.mjs --production",
 		"protos": "node scripts/build-proto.mjs",
 		"postprotos": "biome format src/shared/proto src/core/controller src/hosts/ webview-ui/src/services src/generated --write --no-errors-on-unmatched",
-		"clean": "rimraf dist dist-standalone webview-ui/build src/generated out/",
+		"clean:build": "rimraf dist dist-standalone webview-ui/build src/generated out/",
+		"clean:deps": "rimraf node_modules webview-ui/node_modules",
+		"clean:all": "npm run clean:build && npm run clean:deps",
 		"compile-tests": "node ./scripts/build-tests.js",
 		"watch-tests": "tsc -p . -w --outDir out",
 		"check-types": "npm run protos && npx tsc --noEmit && cd webview-ui && npx tsc -b --noEmit",
@@ -362,15 +362,20 @@
 		"test:integration": "vscode-test",
 		"test:unit": "TS_NODE_PROJECT='./tsconfig.unit-test.json' mocha",
 		"test:coverage": "vscode-test --coverage",
+		"test:sca-server": "npx tsx watch scripts/test-standalone-core-api-server.ts",
+		"test:tp-orchestrator": "npx tsx scripts/testing-platform-orchestrator.ts",
 		"e2e": "playwright test -c playwright.config.ts",
-		"test:e2e": "playwright install && vsce package --no-dependencies --allow-package-secrets sendgrid --out dist/e2e.vsix && node src/test/e2e/utils/build.mjs && playwright test",
-		"test:e2e:optimal": "vsce package --no-dependencies --allow-package-secrets sendgrid --out dist/e2e.vsix && node src/test/e2e/utils/build.mjs && playwright test",
+		"test:e2e:build": "vsce package --allow-package-secrets sendgrid --out dist/e2e.vsix",
+		"test:e2e": "playwright install && npm run test:e2e:build && node src/test/e2e/utils/build.mjs && playwright test",
+		"test:e2e:optimal": "npm run test:e2e:build && node src/test/e2e/utils/build.mjs && playwright test",
+		"test:e2e:ui": "npx tsx scripts/interactive-playwright.ts",
 		"install:all": "npm install && cd webview-ui && npm install",
 		"dev:webview": "cd webview-ui && npm run dev",
 		"build:webview": "cd webview-ui && npm run build",
 		"test:webview": "cd webview-ui && npm run test",
 		"publish:marketplace": "vsce publish --allow-package-secrets sendgrid && ovsx publish",
 		"publish:marketplace:prerelease": "vsce publish --allow-package-secrets sendgrid --pre-release && ovsx publish --pre-release",
+		"publish:marketplace:nightly": "node ./scripts/publish-nightly.mjs",
 		"prepare": "husky",
 		"changeset": "changeset",
 		"version-packages": "changeset version",
@@ -388,6 +393,7 @@
 		"@biomejs/biome": "^2.1.4",
 		"@bufbuild/buf": "^1.54.0",
 		"@changesets/cli": "^2.27.12",
+		"@types/better-sqlite3": "^7.6.13",
 		"@types/chai": "^5.0.1",
 		"@types/clone-deep": "^4.0.4",
 		"@types/diff": "^5.2.1",
@@ -403,19 +409,22 @@
 		"@vscode/test-cli": "^0.0.10",
 		"@vscode/test-electron": "^2.5.2",
 		"@vscode/vsce": "^3.6.0",
+		"c8": "^10.1.3",
 		"chai": "^4.3.10",
-		"chalk": "^5.3.0",
+		"chalk": "5.6.2",
 		"esbuild": "^0.25.0",
 		"grpc-tools": "^1.13.0",
 		"husky": "^9.1.7",
 		"lint-staged": "^16.1.0",
 		"minimatch": "^3.0.3",
 		"npm-run-all": "^4.1.5",
+		"prebuild-install": "^7.1.3",
 		"protoc-gen-ts": "^0.8.7",
 		"proxyquire": "^2.1.3",
 		"rimraf": "^6.0.1",
 		"should": "^13.2.3",
 		"sinon": "^19.0.2",
+		"tree-kill": "^1.2.2",
 		"ts-node": "^10.9.2",
 		"ts-proto": "^2.6.1",
 		"tsconfig-paths": "^4.2.0",
@@ -429,7 +438,7 @@
 		"@bufbuild/protobuf": "^2.2.5",
 		"@cerebras/cerebras_cloud_sdk": "^1.35.0",
 		"@google-cloud/vertexai": "^1.9.3",
-		"@google/genai": "1.0.0",
+		"@google/genai": "^1.11.0",
 		"@grpc/grpc-js": "^1.9.15",
 		"@grpc/reflection": "^1.0.4",
 		"@mistralai/mistralai": "^1.5.0",
@@ -441,12 +450,14 @@
 		"@opentelemetry/sdk-trace-node": "^1.30.1",
 		"@opentelemetry/semantic-conventions": "^1.30.0",
 		"@playwright/test": "^1.53.2",
+		"@sap-ai-sdk/ai-api": "^1.17.0",
+		"@sap-ai-sdk/orchestration": "^1.17.0",
 		"@sentry/browser": "^9.12.0",
 		"@streamparser/json": "^0.0.22",
 		"@types/uuid": "^10.0.0",
 		"@vscode/codicons": "^0.0.36",
 		"archiver": "^7.0.1",
-		"axios": "^1.8.2",
+		"axios": "^1.12.0",
 		"cheerio": "^1.0.0",
 		"chokidar": "^4.0.1",
 		"chrome-launcher": "^1.1.2",
@@ -461,6 +472,7 @@
 		"get-folder-size": "^5.0.0",
 		"globby": "^14.0.2",
 		"grpc-health-check": "^2.0.2",
+		"https-proxy-agent": "^7.0.6",
 		"iconv-lite": "^0.6.3",
 		"ignore": "^7.0.3",
 		"image-size": "^2.0.2",
@@ -477,13 +489,13 @@
 		"p-timeout": "^6.1.4",
 		"p-wait-for": "^5.0.2",
 		"pdf-parse": "^1.1.1",
-		"posthog-node": "^4.8.1",
+		"posthog-node": "^5.8.0",
 		"puppeteer-chromium-resolver": "^23.0.0",
 		"puppeteer-core": "^23.4.0",
 		"reconnecting-eventsource": "^1.6.4",
 		"serialize-error": "^11.0.3",
 		"simple-git": "^3.27.0",
-		"strip-ansi": "^7.1.0",
+		"strip-ansi": "^7.1.2",
 		"tree-sitter-wasms": "^0.1.11",
 		"ts-morph": "^25.0.1",
 		"turndown": "^7.2.0",
@@ -492,5 +504,40 @@
 		"vscode-uri": "^3.1.0",
 		"web-tree-sitter": "^0.22.6",
 		"zod": "^3.24.2"
+	},
+	"c8": {
+		"reporter": [
+			"lcov",
+			"html"
+		],
+		"exclude": [
+			"**/testing-platform/**",
+			"**/webview-ui/**",
+			"**/.vscode-test/**",
+			"**/node_modules/**",
+			"node_modules",
+			"**/dist-standalone/src/**",
+			"**/dist-standalone/vsce-extension/https:/**",
+			"**/dist-standalone/vsce-extension/**",
+			"**/dist-standalone/https:/**",
+			"**/dist-standalone/LIB/src/**",
+			"**/dist-standalone/pdfjs-dist/**",
+			"**/*.d.ts",
+			"**/*.{test,spec}.{js,jsx,ts,tsx,mjs,cjs}",
+			"**/__tests__/**",
+			"**/test/**",
+			"**/tests/**",
+			"**/.nyc_output/**",
+			"**/tests-results/**",
+			"src/test/**",
+			"**/src/xml/**",
+			"**/standalone/**",
+			"**/src/generated/**",
+			"**/evals/cli/dist/**",
+			"**/evals/cli/src/**",
+			"dist"
+		],
+		"all": true,
+		"exclude-after-remap": true
 	}
 }

proto/cline/account.proto

@@ -38,6 +38,9 @@ service AccountService {
   rpc setUserOrganization(UserOrganizationUpdateRequest) returns (Empty);
 
   rpc openrouterAuthClicked(EmptyRequest) returns (Empty);
+
+  // Returns a link the webview can use to redirect back to the user's IDE.
+  rpc getRedirectUrl(EmptyRequest) returns (String);
 }
 
 message AuthStateChangedRequest {

proto/cline/browser.proto

@@ -2,6 +2,7 @@ syntax = "proto3";
 
 package cline;
 import "cline/common.proto";
+import "cline/state.proto";
 option java_package = "bot.cline.proto";
 option java_multiple_files = true;
 
@@ -10,7 +11,6 @@ service BrowserService {
   rpc testBrowserConnection(StringRequest) returns (BrowserConnection);
   rpc discoverBrowser(EmptyRequest) returns (BrowserConnection);
   rpc getDetectedChromePath(EmptyRequest) returns (ChromePath);
-  rpc updateBrowserSettings(UpdateBrowserSettingsRequest) returns (Boolean);
   rpc relaunchChromeDebugMode(EmptyRequest) returns (String);
 }
 
@@ -31,11 +31,6 @@ message ChromePath {
   bool is_bundled = 2;
 }
 
-message Viewport {
-  int32 width = 1;
-  int32 height = 2;
-}
-
 message BrowserSettings {
   Viewport viewport = 1;
   optional string remote_browser_host = 2;

proto/cline/common.proto

@@ -8,19 +8,16 @@ message Metadata {
 }
 
 message EmptyRequest {
-  Metadata metadata = 1;
 }
 
 message Empty {
 }
 
 message StringRequest {
-  Metadata metadata = 1;
   string value = 2;
 }
 
 message StringArrayRequest {
-  Metadata metadata = 1;
   repeated string value = 2;
 }
 
@@ -29,7 +26,6 @@ message String {
 }
 
 message Int64Request {
-  Metadata metadata = 1;
   int64 value = 2;
 }
 
@@ -38,7 +34,6 @@ message Int64 {
 }
 
 message BytesRequest {
-  Metadata metadata = 1;
   bytes value = 2;
 }
 
@@ -47,7 +42,6 @@ message Bytes {
 }
 
 message BooleanRequest {
-  Metadata metadata = 1;
   bool value = 2;
 }
 

proto/cline/dictation.proto

@@ -0,0 +1,41 @@
+syntax = "proto3";
+
+package cline;
+import "cline/common.proto";
+option java_package = "bot.cline.proto";
+option java_multiple_files = true;
+
+service DictationService {
+  rpc startRecording(EmptyRequest) returns (RecordingResult);
+  rpc stopRecording(EmptyRequest) returns (RecordedAudio);
+  rpc cancelRecording(EmptyRequest) returns (RecordingResult);
+  rpc getRecordingStatus(EmptyRequest) returns (RecordingStatus);
+  rpc transcribeAudio(TranscribeAudioRequest) returns (Transcription);
+}
+
+message TranscribeAudioRequest {
+  string audio_base64 = 2;
+  string language = 3;
+}
+
+message RecordingResult {
+  bool success = 1;
+  string error = 2;
+}
+
+message RecordedAudio {
+  bool success = 1;
+  string audio_base64 = 2;
+  string error = 3;
+}
+
+message RecordingStatus {
+  bool is_recording = 1;
+  double duration_seconds = 2;
+  string error = 3;
+}
+
+message Transcription {
+  string text = 1;
+  string error = 2;
+}

proto/cline/models.proto

@@ -32,7 +32,9 @@ service ModelsService {
   // Refreshes and returns Baseten models
   rpc refreshBasetenModels(EmptyRequest) returns (OpenRouterCompatibleModelInfo);
   // Fetches available models from SAP AI Core
-  rpc getSapAiCoreModels(SapAiCoreModelsRequest) returns (StringArray);
+  rpc getSapAiCoreModels(SapAiCoreModelsRequest) returns (SapAiCoreModelsResponse);
+  // Fetches available models from OCA
+  rpc refreshOcaModels(StringRequest) returns (OcaCompatibleModelInfo);
 }
 
 // List of VS Code LM models
@@ -50,20 +52,20 @@ message LanguageModelChatSelector {
 
 // Price tier for tiered pricing models
 message PriceTier {
-  int32 token_limit = 1;  // Upper limit (inclusive) of input tokens for this price
+  int64 token_limit = 1;  // Upper limit (inclusive) of input tokens for this price
   double price = 2;       // Price per million tokens for this tier
 }
 
 // Thinking configuration for models that support thinking/reasoning
 message ThinkingConfig {
-  optional int32 max_budget = 1;                    // Max allowed thinking budget tokens
+  optional int64 max_budget = 1;                    // Max allowed thinking budget tokens
   optional double output_price = 2;                 // Output price per million tokens when budget > 0
   repeated PriceTier output_price_tiers = 3;        // Optional: Tiered output price when budget > 0
 }
 
 // Model tier for tiered pricing structures
 message ModelTier {
-  int32 context_window = 1;
+  int64 context_window = 1;
   optional double input_price = 2;
   optional double output_price = 3;
   optional double cache_writes_price = 4;
@@ -108,12 +110,67 @@ message SapAiCoreModelsRequest {
   string resource_group = 6;
 }
 
+// SAP AI Core model with deployment information
+message SapAiCoreModelDeployment {
+  string model_name = 1;
+  string deployment_id = 2;
+}
+
+
+// Response for SAP AI Core models with orchestration availability
+message SapAiCoreModelsResponse {
+  repeated SapAiCoreModelDeployment deployments = 1;
+  bool orchestration_available = 2;
+}
+
 // Request for updating API configuration
 message UpdateApiConfigurationRequest {
   Metadata metadata = 1;
   ModelsApiConfiguration api_configuration = 2;
 }
 
+ // Model info for OCA (OpenAI-compatible) models exposed by the OCA provider
+message OcaModelInfo {
+  // Maximum completion tokens per request supported by this model
+  optional int64 max_tokens = 1;
+  // Total context window in tokens (input + output)
+  optional int64 context_window = 2;
+  // Whether the model supports image inputs
+  optional bool supports_images = 3;
+  // Whether prompt caching is supported for this model
+  bool supports_prompt_cache = 4;
+  // Price per million input tokens (USD unless otherwise specified by provider)
+  optional double input_price = 5;
+  // Price per million output tokens (USD unless otherwise specified by provider)
+  optional double output_price = 6;
+  // Thinking/reasoning configuration if the model supports it
+  optional ThinkingConfig thinking_config = 7;
+  // Price per million tokens for prompt cache writes
+  optional double cache_writes_price = 9;
+  // Price per million tokens for prompt cache reads
+  optional double cache_reads_price = 10;
+  // Human-readable model description
+  optional string description = 11;
+  // Recommended default temperature for this model
+  optional double temperature = 13;
+  // Optional survey content to display in the UI
+  optional string survey_content = 14;
+  // Identifier for the survey associated with this model
+  optional string survey_id = 15;
+  // Optional banner content (e.g., deprecation or promotion notes)
+  optional string banner = 16;
+  // Canonical model identifier as reported by OCA
+  string model_name = 17;
+}
+
+ // Aggregated OCA model catalog keyed by model identifier
+message OcaCompatibleModelInfo {
+  // key: canonical model id as reported by OCA (e.g., "openai/gpt-4o-mini")
+  // value: OcaModelInfo describing that model
+  map<string, OcaModelInfo> models = 1;
+  optional string error = 2;
+}
+
 // API Provider enumeration
 enum ApiProvider {
   ANTHROPIC = 0;
@@ -150,12 +207,14 @@ enum ApiProvider {
   ZAI = 31;
   VERCEL_AI_GATEWAY = 32;
   QWEN_CODE = 33;
+  DIFY = 34;
+  OCA = 35;
 }
 
 // Model info for OpenAI-compatible models
 message OpenAiCompatibleModelInfo {
-  optional int32 max_tokens = 1;
-  optional int32 context_window = 2;
+  optional int64 max_tokens = 1;
+  optional int64 context_window = 2;
   optional bool supports_images = 3;
   bool supports_prompt_cache = 4;
   optional double input_price = 5;
@@ -172,8 +231,8 @@ message OpenAiCompatibleModelInfo {
 
 // Model info for LiteLLM models
 message LiteLLMModelInfo {
-  optional int32 max_tokens = 1;
-  optional int32 context_window = 2;
+  optional int64 max_tokens = 1;
+  optional int64 context_window = 2;
   optional bool supports_images = 3;
   bool supports_prompt_cache = 4;
   optional double input_price = 5;
@@ -225,8 +284,8 @@ message ModelsApiConfiguration {
   optional string requesty_base_url = 33;
   optional string together_api_key = 34;
   optional string fireworks_api_key = 35;
-  optional int32 fireworks_model_max_completion_tokens = 36;
-  optional int32 fireworks_model_max_tokens = 37;
+  optional int64 fireworks_model_max_completion_tokens = 36;
+  optional int64 fireworks_model_max_tokens = 37;
   optional string qwen_api_key = 38;
   optional string doubao_api_key = 39;
   optional string mistral_api_key = 40;
@@ -238,32 +297,38 @@ message ModelsApiConfiguration {
   optional string xai_api_key = 46;
   optional string sambanova_api_key = 47;
   optional string cerebras_api_key = 48;
-  optional int32 request_timeout_ms = 49;
+  optional int64 request_timeout_ms = 49;
   optional string sap_ai_core_client_id = 50;
   optional string sap_ai_core_client_secret = 51;
   optional string sap_ai_resource_group = 52;
   optional string sap_ai_core_token_url = 53;
   optional string sap_ai_core_base_url = 54;
-  optional string moonshot_api_key = 55;
-  optional string moonshot_api_line = 56;
-  optional string aws_authentication = 57;
-  optional string aws_bedrock_api_key = 58;
-  optional string cline_account_id = 59;
-  optional string groq_api_key = 60;
-  optional string hugging_face_api_key = 61;
-  optional string huawei_cloud_maas_api_key = 62;
-  optional string baseten_api_key = 63;
-  optional string ollama_api_key = 64;
-  optional string zai_api_key = 65;
-  optional string zai_api_line = 66;
-  optional string lm_studio_max_tokens = 67;
-  optional string vercel_ai_gateway_api_key = 68;
-  optional string qwen_code_oauth_path = 69;
+  optional bool sap_ai_core_use_orchestration_mode = 55;
+  optional string moonshot_api_key = 56;
+  optional string moonshot_api_line = 57;
+  optional string aws_authentication = 58;
+  optional string aws_bedrock_api_key = 59;
+  optional string cline_account_id = 60;
+  optional string groq_api_key = 61;
+  optional string hugging_face_api_key = 62;
+  optional string huawei_cloud_maas_api_key = 63;
+  optional string baseten_api_key = 64;
+  optional string ollama_api_key = 65;
+  optional string zai_api_key = 66;
+  optional string zai_api_line = 67;
+  optional string lm_studio_max_tokens = 68;
+  optional string vercel_ai_gateway_api_key = 69;
+  optional string qwen_code_oauth_path = 70;
+  optional string dify_api_key = 71;
+  optional string dify_base_url = 72;
+  optional string oca_base_url = 73;
+  optional string oca_api_key = 74;
+  optional string oca_refresh_token = 75;
 
   // Plan mode configurations
   optional ApiProvider plan_mode_api_provider = 100;
   optional string plan_mode_api_model_id = 101;
-  optional int32 plan_mode_thinking_budget_tokens = 102;
+  optional int64 plan_mode_thinking_budget_tokens = 102;
   optional string plan_mode_reasoning_effort = 103;
   optional LanguageModelChatSelector plan_mode_vs_code_lm_model_selector = 104;
   optional bool plan_mode_aws_bedrock_custom_selected = 105;
@@ -281,21 +346,25 @@ message ModelsApiConfiguration {
   optional string plan_mode_together_model_id = 117;
   optional string plan_mode_fireworks_model_id = 118;
   optional string plan_mode_sap_ai_core_model_id = 119;
-  optional string plan_mode_groq_model_id = 120;
-  optional OpenRouterModelInfo plan_mode_groq_model_info = 121;
-  optional string plan_mode_hugging_face_model_id = 122;
-  optional OpenRouterModelInfo plan_mode_hugging_face_model_info = 123;
-  optional string plan_mode_huawei_cloud_maas_model_id = 124;
-  optional OpenRouterModelInfo plan_mode_huawei_cloud_maas_model_info = 125; 
-  optional string plan_mode_baseten_model_id = 126;
-  optional OpenRouterModelInfo plan_mode_baseten_model_info = 127;
-  optional string plan_mode_vercel_ai_gateway_model_id = 128;
-  optional OpenRouterModelInfo plan_mode_vercel_ai_gateway_model_info = 129;
+  optional string plan_mode_sap_ai_core_deployment_id = 120;
+  optional string plan_mode_groq_model_id = 121;
+  optional OpenRouterModelInfo plan_mode_groq_model_info = 122;
+  optional string plan_mode_hugging_face_model_id = 123;
+  optional OpenRouterModelInfo plan_mode_hugging_face_model_info = 124;
+  optional string plan_mode_huawei_cloud_maas_model_id = 125;
+  optional OpenRouterModelInfo plan_mode_huawei_cloud_maas_model_info = 126; 
+  optional string plan_mode_baseten_model_id = 127;
+  optional OpenRouterModelInfo plan_mode_baseten_model_info = 128;
+  optional string plan_mode_vercel_ai_gateway_model_id = 129;
+  optional OpenRouterModelInfo plan_mode_vercel_ai_gateway_model_info = 130;
+  optional string plan_mode_oca_model_id = 131;
+  optional OcaModelInfo plan_mode_oca_model_info = 132;
+
 
   // Act mode configurations
   optional ApiProvider act_mode_api_provider = 200;
   optional string act_mode_api_model_id = 201;
-  optional int32 act_mode_thinking_budget_tokens = 202;
+  optional int64 act_mode_thinking_budget_tokens = 202;
   optional string act_mode_reasoning_effort = 203;
   optional LanguageModelChatSelector act_mode_vs_code_lm_model_selector = 204;
   optional bool act_mode_aws_bedrock_custom_selected = 205;
@@ -313,16 +382,17 @@ message ModelsApiConfiguration {
   optional string act_mode_together_model_id = 217;
   optional string act_mode_fireworks_model_id = 218;
   optional string act_mode_sap_ai_core_model_id = 219;
-  optional string act_mode_groq_model_id = 220;
-  optional OpenRouterModelInfo act_mode_groq_model_info = 221;
-  optional string act_mode_hugging_face_model_id = 222;
-  optional OpenRouterModelInfo act_mode_hugging_face_model_info = 223;
-  optional string act_mode_huawei_cloud_maas_model_id = 224;
-  optional OpenRouterModelInfo act_mode_huawei_cloud_maas_model_info = 225;
-  optional string act_mode_baseten_model_id = 226;
-  optional OpenRouterModelInfo act_mode_baseten_model_info = 227;
-  optional string act_mode_vercel_ai_gateway_model_id = 228;
-  optional OpenRouterModelInfo act_mode_vercel_ai_gateway_model_info = 229;
-
-  repeated string favorited_model_ids = 300;
+  optional string act_mode_sap_ai_core_deployment_id = 220;
+  optional string act_mode_groq_model_id = 221;
+  optional OpenRouterModelInfo act_mode_groq_model_info = 222;
+  optional string act_mode_hugging_face_model_id = 223;
+  optional OpenRouterModelInfo act_mode_hugging_face_model_info = 224;
+  optional string act_mode_huawei_cloud_maas_model_id = 225;
+  optional OpenRouterModelInfo act_mode_huawei_cloud_maas_model_info = 226;
+  optional string act_mode_baseten_model_id = 227;
+  optional OpenRouterModelInfo act_mode_baseten_model_info = 228;
+  optional string act_mode_vercel_ai_gateway_model_id = 229;
+  optional OpenRouterModelInfo act_mode_vercel_ai_gateway_model_info = 230;
+  optional string act_mode_oca_model_id = 231;
+  optional OcaModelInfo act_mode_oca_model_info = 232;
 }

proto/cline/oca_account.proto

@@ -0,0 +1,36 @@
+syntax = "proto3";
+
+package cline;
+import "cline/common.proto";
+option java_package = "bot.cline.proto";
+option java_multiple_files = true;
+
+// Service for account-related operations
+service OcaAccountService {
+  // Handles the user clicking the login link in the UI.
+  // Generates a secure nonce for state validation, stores it in secrets,
+  // and opens the authentication URL in the external browser.
+  rpc ocaAccountLoginClicked(EmptyRequest) returns (String);
+    
+  // Handles the user clicking the logout button in the UI.
+  // Clears API keys and user state.
+  rpc ocaAccountLogoutClicked(EmptyRequest) returns (Empty);
+
+  // Subscribe to auth status update events (when authentication state changes)
+  rpc ocaSubscribeToAuthStatusUpdate(EmptyRequest) 
+      returns (stream OcaAuthState);
+
+}
+
+
+message OcaAuthState {
+  optional OcaUserInfo user = 1;
+  optional string api_key = 2;
+}
+
+// User's information
+message OcaUserInfo {
+  string uid = 1;
+  optional string display_name = 2;
+  optional string email = 3;
+}

proto/cline/state.proto

@@ -1,6 +1,7 @@
 syntax = "proto3";
 package cline;
 import "cline/common.proto";
+import "cline/models.proto";
 option java_package = "bot.cline.proto";
 option java_multiple_files = true;
 
@@ -8,7 +9,6 @@ service StateService {
   rpc getLatestState(EmptyRequest) returns (State);
   rpc updateTerminalConnectionTimeout(UpdateTerminalConnectionTimeoutRequest) returns (UpdateTerminalConnectionTimeoutResponse);
   rpc updateTerminalReuseEnabled(BooleanRequest) returns (Empty);
-  rpc updateDefaultTerminalProfile(StringRequest) returns (TerminalProfileUpdateResponse);
   rpc getAvailableTerminalProfiles(EmptyRequest) returns (TerminalProfiles);
   rpc subscribeToState(EmptyRequest) returns (stream State);
   rpc toggleFavoriteModel(StringRequest) returns (Empty);
@@ -19,7 +19,11 @@ service StateService {
   rpc updateTelemetrySetting(TelemetrySettingRequest) returns (Empty);
   rpc setWelcomeViewCompleted(BooleanRequest) returns (Empty);
 }
-
+message DictationSettings {
+  bool feature_enabled = 1;
+  bool dictation_enabled = 2;
+  string dictation_language = 3;
+}
 message State {
   string state_json = 1;
 }
@@ -56,6 +60,7 @@ enum OpenaiReasoningEffort {
   LOW = 0;
   MEDIUM = 1;
   HIGH = 2;
+  MINIMAL = 3;
 }
 
 enum McpDisplayMode {
@@ -106,6 +111,16 @@ message TelemetrySettingRequest {
   TelemetrySettingEnum setting = 2;
 }
 
+// Browser settings for UpdateSettingsRequest
+message BrowserSettingsUpdate {
+  optional Viewport viewport = 1;
+  optional string remote_browser_host = 2;
+  optional bool remote_browser_enabled = 3;
+  optional string chrome_executable_path = 4;
+  optional bool disable_tool_use = 5;
+  optional string custom_args = 6;
+}
+
 // Message for updating settings
 message UpdateSettingsRequest {
   Metadata metadata = 1;
@@ -126,6 +141,12 @@ message UpdateSettingsRequest {
   optional FocusChainSettings focus_chain_settings = 17;
   optional bool use_auto_condense = 18;
   optional string custom_prompt = 19;
+  optional BrowserSettingsUpdate browser_settings = 20;
+  optional string default_terminal_profile = 21;
+  optional bool yolo_mode_toggled = 22;
+  optional DictationSettings dictation_settings = 23;
+  optional int32 auto_condense_threshold = 24;
+  optional bool multi_root_enabled = 25;
 }
 
 // Complete API Configuration message
@@ -137,10 +158,10 @@ message ApiConfiguration {
   optional string lite_llm_base_url = 4;
   optional string lite_llm_api_key = 5;
   optional bool lite_llm_use_prompt_cache = 6;
-  optional string openai_headers = 7; // JSON string
+  map<string, string> open_ai_headers = 7;
   optional string anthropic_base_url = 8;
-  optional string openrouter_api_key = 9;
-  optional string openrouter_provider_sorting = 10;
+  optional string open_router_api_key = 9;
+  optional string open_router_provider_sorting = 10;
   optional string aws_access_key = 11;
   optional string aws_secret_key = 12;
   optional string aws_session_token = 13;
@@ -153,14 +174,14 @@ message ApiConfiguration {
   optional string claude_code_path = 20;
   optional string vertex_project_id = 21;
   optional string vertex_region = 22;
-  optional string openai_base_url = 23;
-  optional string openai_api_key = 24;
+  optional string open_ai_base_url = 23;
+  optional string open_ai_api_key = 24;
   optional string ollama_base_url = 25;
   optional string ollama_api_options_ctx_num = 26;
   optional string lm_studio_base_url = 27;
   optional string gemini_api_key = 28;
   optional string gemini_base_url = 29;
-  optional string openai_native_api_key = 30;
+  optional string open_ai_native_api_key = 30;
   optional string deep_seek_api_key = 31;
   optional string requesty_api_key = 32;
   optional string requesty_base_url = 33;
@@ -194,61 +215,67 @@ message ApiConfiguration {
   optional string lm_studio_max_tokens = 61;
   optional string vercel_ai_gateway_api_key = 62;
   optional string qwen_code_oauth_path = 63;
+  optional string dify_api_key = 64;
+  optional string dify_base_url = 65;
+  optional string oca_base_url = 66;
+  optional string oca_api_key = 67;
+  optional string oca_refresh_token = 68;
   
   // Plan mode configurations
-  optional string plan_mode_api_provider = 100;
+  optional ApiProvider plan_mode_api_provider = 100;
   optional string plan_mode_api_model_id = 101;
   optional int32 plan_mode_thinking_budget_tokens = 102;
   optional string plan_mode_reasoning_effort = 103;
-  optional string plan_mode_vscode_lm_model_selector = 104; // JSON string
+  optional LanguageModelChatSelector plan_mode_vs_code_lm_model_selector = 104;
   optional bool plan_mode_aws_bedrock_custom_selected = 105;
   optional string plan_mode_aws_bedrock_custom_model_base_id = 106;
-  optional string plan_mode_openrouter_model_id = 107;
-  optional string plan_mode_openrouter_model_info = 108; // JSON string
-  optional string plan_mode_openai_model_id = 109;
-  optional string plan_mode_openai_model_info = 110; // JSON string
+  optional string plan_mode_open_router_model_id = 107;
+  optional OpenRouterModelInfo plan_mode_open_router_model_info = 108;
+  optional string plan_mode_open_ai_model_id = 109;
+  optional OpenAiCompatibleModelInfo plan_mode_open_ai_model_info = 110;
   optional string plan_mode_ollama_model_id = 111;
   optional string plan_mode_lm_studio_model_id = 112;
   optional string plan_mode_lite_llm_model_id = 113;
-  optional string plan_mode_lite_llm_model_info = 114; // JSON string
+  optional LiteLLMModelInfo plan_mode_lite_llm_model_info = 114;
   optional string plan_mode_requesty_model_id = 115;
-  optional string plan_mode_requesty_model_info = 116; // JSON string
+  optional OpenRouterModelInfo plan_mode_requesty_model_info = 116;
   optional string plan_mode_together_model_id = 117;
   optional string plan_mode_fireworks_model_id = 118;
   optional string plan_mode_sap_ai_core_model_id = 119;
   optional string plan_mode_huawei_cloud_maas_model_id = 120;
-  optional string plan_mode_huawei_cloud_maas_model_info = 121;
+  optional OpenRouterModelInfo plan_mode_huawei_cloud_maas_model_info = 121;
   optional string plan_mode_vercel_ai_gateway_model_id = 122;
-  optional string plan_mode_vercel_ai_gateway_model_info = 123;
+  optional OpenRouterModelInfo plan_mode_vercel_ai_gateway_model_info = 123;
+  optional string plan_mode_oca_model_id = 124;
+  optional OcaModelInfo plan_mode_oca_model_info = 125;
   
   // Act mode configurations
-  optional string act_mode_api_provider = 200;
+  optional ApiProvider act_mode_api_provider = 200;
   optional string act_mode_api_model_id = 201;
   optional int32 act_mode_thinking_budget_tokens = 202;
   optional string act_mode_reasoning_effort = 203;
-  optional string act_mode_vscode_lm_model_selector = 204; // JSON string
+  optional LanguageModelChatSelector act_mode_vs_code_lm_model_selector = 204;
   optional bool act_mode_aws_bedrock_custom_selected = 205;
   optional string act_mode_aws_bedrock_custom_model_base_id = 206;
-  optional string act_mode_openrouter_model_id = 207;
-  optional string act_mode_openrouter_model_info = 208; // JSON string
-  optional string act_mode_openai_model_id = 209;
-  optional string act_mode_openai_model_info = 210; // JSON string
+  optional string act_mode_open_router_model_id = 207;
+  optional OpenRouterModelInfo act_mode_open_router_model_info = 208;
+  optional string act_mode_open_ai_model_id = 209;
+  optional OpenAiCompatibleModelInfo act_mode_open_ai_model_info = 210;
   optional string act_mode_ollama_model_id = 211;
   optional string act_mode_lm_studio_model_id = 212;
   optional string act_mode_lite_llm_model_id = 213;
-  optional string act_mode_lite_llm_model_info = 214; // JSON string
+  optional LiteLLMModelInfo act_mode_lite_llm_model_info = 214;
   optional string act_mode_requesty_model_id = 215;
-  optional string act_mode_requesty_model_info = 216; // JSON string
+  optional OpenRouterModelInfo act_mode_requesty_model_info = 216;
   optional string act_mode_together_model_id = 217;
   optional string act_mode_fireworks_model_id = 218;
   optional string act_mode_sap_ai_core_model_id = 219;
   optional string act_mode_huawei_cloud_maas_model_id = 220;
-  optional string act_mode_huawei_cloud_maas_model_info = 221;
+  optional OpenRouterModelInfo act_mode_huawei_cloud_maas_model_info = 221;
   optional string act_mode_vercel_ai_gateway_model_id = 222;
-  optional string act_mode_vercel_ai_gateway_model_info = 223;
-  
-  // Favorited model IDs
-  repeated string favorited_model_ids = 300;
+  optional OpenRouterModelInfo act_mode_vercel_ai_gateway_model_info = 223;
+  optional string act_mode_oca_model_id = 224;
+  optional OcaModelInfo act_mode_oca_model_info = 225;
 
   // Extension fields for Bedrock Api Keys
   optional string aws_authentication = 301;
@@ -266,6 +293,11 @@ message FocusChainSettings {
   int32 remind_cline_interval = 2;
 }
 
+message Viewport {
+  int32 width = 1;
+  int32 height = 2;
+}
+
 message UpdateTerminalConnectionTimeoutResponse {
   optional int32 timeout_ms = 1;
 }

proto/host/env.proto

@@ -17,7 +17,20 @@ service EnvService {
   // Returns a stable machine identifier for telemetry distinctId purposes.
   rpc getMachineId(cline.EmptyRequest) returns (cline.String);
 
+  // Returns the name and version of the host IDE or environment.
   rpc getHostVersion(cline.EmptyRequest) returns (GetHostVersionResponse);
+
+  // Returns a URI that will redirect to the host environment.
+  // e.g. vscode://saoudrizwan.claude-dev, idea://, pycharm://, etc. 
+  // If the host does not support URIs it should return empty.
+  rpc getIdeRedirectUri(cline.EmptyRequest) returns (cline.String);
+
+  // Returns the telemetry settings of the host environment. This may return UNSUPPORTED
+  // if the host does not specify telemetry settings for the plugin.
+  rpc getTelemetrySettings(cline.EmptyRequest) returns (GetTelemetrySettingsResponse);
+
+  // Returns events when the telemetry settings change.
+  rpc subscribeToTelemetrySettings(cline.EmptyRequest) returns (stream TelemetrySettingsEvent);
 }
 
 message GetHostVersionResponse {
@@ -26,3 +39,16 @@ message GetHostVersionResponse {
   // The version of the host platform, e.g. 1.103.0
   optional string version = 2;
 }
+
+enum Setting {
+  UNSUPPORTED = 0; // This host does not support this setting.
+  ENABLED = 1;
+  DISABLED = 2;
+}
+message GetTelemetrySettingsResponse {
+  Setting is_enabled = 1;
+}
+
+message TelemetrySettingsEvent {
+  Setting is_enabled = 1; 
+}

proto/host/watch.proto

@@ -1,36 +0,0 @@
-syntax = "proto3";
-
-package host;
-option java_package = "bot.cline.host.proto";
-option java_multiple_files = true;
-
-import "cline/common.proto";
-
-/** 
- * The watch service is only here as example of a streaming rpc in the host bridge. 
- * This being replaced with a native JS file watcher.
- */
-// WatchService provides methods for watching files in the IDE
-service WatchService {
-  // Subscribe to file changes
-  rpc subscribeToFile(SubscribeToFileRequest) returns (stream FileChangeEvent);
-}
-
-// Request to subscribe to file changes
-message SubscribeToFileRequest {
-  cline.Metadata metadata = 1;
-  string path = 2;
-}
-
-// Event representing a file change
-message FileChangeEvent {
-  enum ChangeType {
-    CREATED = 0;
-    CHANGED = 1;
-    DELETED = 2;
-  }
-  
-  string path = 1;
-  ChangeType type = 2;
-  string content = 3; // Optional content of the file after change
-}

proto/host/window.proto

@@ -4,8 +4,6 @@ package host;
 option java_package = "bot.cline.host.proto";
 option java_multiple_files = true;
 
-import "cline/common.proto";
-
 // Provides methods for working with IDE windows and editors.
 service WindowService {
   // Opens a text document in the IDE editor and returns editor information.
@@ -40,7 +38,6 @@ service WindowService {
 }
 
 message ShowTextDocumentRequest {
-  cline.Metadata metadata = 1;
   string path = 2;
   optional ShowTextDocumentOptions options = 3;
 }
@@ -59,7 +56,6 @@ message TextEditorInfo {
 }
 
 message ShowOpenDialogueRequest {
-  cline.Metadata metadata = 1;
   optional bool can_select_many = 2;
   optional string open_label = 3;
   optional ShowOpenDialogueFilterOption filters = 4;

proto/host/workspace.proto

@@ -18,14 +18,15 @@ service WorkspaceService {
 
   // Get diagnostics from the workspace.
   rpc getDiagnostics(GetDiagnosticsRequest) returns (GetDiagnosticsResponse);
-  // Returns workspace items (files/folders) matching a query for mention autocomplete
-  rpc searchWorkspaceItems(SearchWorkspaceItemsRequest) returns (SearchWorkspaceItemsResponse);
 
   // Makes the problems panel/pane visible in the IDE and focuses it.
   rpc openProblemsPanel(OpenProblemsPanelRequest) returns (OpenProblemsPanelResponse);
 
   // Opens the IDE file explorer panel and selects a file or directory.
   rpc openInFileExplorerPanel(OpenInFileExplorerPanelRequest) returns (OpenInFileExplorerPanelResponse); 
+
+  // Opens and focuses the Cline sidebar panel in the host IDE.
+  rpc openClineSidebarPanel(OpenClineSidebarPanelRequest) returns (OpenClineSidebarPanelResponse);
 }
 
 message GetWorkspacePathsRequest {
@@ -86,3 +87,7 @@ message OpenInFileExplorerPanelRequest {
   string path = 1;
 }
 message OpenInFileExplorerPanelResponse {}
+
+// Request/response for opening the Cline sidebar
+message OpenClineSidebarPanelRequest {}
+message OpenClineSidebarPanelResponse {}

scripts/build-proto.mjs

@@ -10,7 +10,6 @@ import * as path from "path"
 import { rmrf } from "./file-utils.mjs"
 import { main as generateHostBridgeClient } from "./generate-host-bridge-client.mjs"
 import { main as generateProtoBusSetup } from "./generate-protobus-setup.mjs"
-import { loadProtoDescriptorSet } from "./proto-utils.mjs"
 
 const require = createRequire(import.meta.url)
 const PROTOC = path.join(require.resolve("grpc-tools"), "../bin/protoc")
@@ -31,14 +30,13 @@ const TS_PROTO_OPTIONS = [
 	"esModuleInterop=true",
 	"outputServices=generic-definitions", // output generic ServiceDefinitions
 	"outputIndex=true", // output an index file for each package which exports all protos in the package.
-	"useOptionals=messages", // Message fields are optional, scalars are not.
+	"useOptionals=none", // scalar and message fields are required unless they are marked as optional.
 	"useDate=false", // Timestamp fields will not be automatically converted to Date.
 ]
 
 async function main() {
 	await cleanup()
 	await compileProtos()
-	await checkProtos()
 	await generateProtoBusSetup()
 	await generateHostBridgeClient()
 }
@@ -59,7 +57,7 @@ async function compileProtos() {
 
 	tsProtoc(TS_OUT_DIR, protoFiles, TS_PROTO_OPTIONS)
 	// grpc-js is used to generate service impls for the ProtoBus service.
-	tsProtoc(GRPC_JS_OUT_DIR, protoFiles, ["outputServices=grpc-js,outputClientImpl=false", ...TS_PROTO_OPTIONS])
+	tsProtoc(GRPC_JS_OUT_DIR, protoFiles, ["outputServices=grpc-js", ...TS_PROTO_OPTIONS])
 	// nice-js is used for the Host Bridge client impls because it uses promises.
 	tsProtoc(NICE_JS_OUT_DIR, protoFiles, ["outputServices=nice-grpc,useExactTypes=false", ...TS_PROTO_OPTIONS])
 
@@ -187,51 +185,6 @@ function checkAppleSiliconCompatibility() {
 	}
 }
 
-const int64TypeNames = ["TYPE_INT64", "TYPE_UINT64", "TYPE_SINT64", "TYPE_FIXED64", "TYPE_SFIXED64"]
-
-async function checkProtos() {
-	const proto = await loadProtoDescriptorSet()
-	const int64Fields = []
-
-	for (const [packageName, packageDef] of Object.entries(proto)) {
-		for (const [messageName, def] of Object.entries(packageDef)) {
-			// Skip service definitions
-			if (def && typeof def === "object" && "service" in def) {
-				continue
-			}
-			// Check message fields
-			if (def && def.type && def.type.field) {
-				for (const field of def.type.field) {
-					if (int64TypeNames.includes(field.type)) {
-						const name = `${packageName}.${messageName}.${field.name}`
-						int64Fields.push({
-							name: name,
-							type: field.type,
-						})
-					}
-				}
-			}
-		}
-	}
-
-	if (int64Fields.length > 0) {
-		console.log(chalk.yellow(`\nWarning: Found ${int64Fields.length} fields using 64-bit integer types`))
-		for (const field of int64Fields) {
-			const typeNames = {
-				TYPE_INT64: "int64",
-				TYPE_UINT64: "uint64",
-				TYPE_SINT64: "sint64",
-				TYPE_FIXED64: "fixed64",
-				TYPE_SFIXED64: "sfixed64",
-			}
-			log_verbose(chalk.yellow(`  - ${field.name} (${typeNames[field.type]})`))
-		}
-		log_verbose(chalk.yellow("\nWARNING: 64-bit integer fields detected in proto definitions"))
-		log_verbose(chalk.yellow("JavaScript cannot safely represent integers larger than 2^53-1 (Number.MAX_SAFE_INTEGER)."))
-		log_verbose(chalk.yellow("Consider using string representation for large numbers or implementing BigInt support.\n"))
-	}
-}
-
 function log_verbose(s) {
 	if (process.argv.includes("-v") || process.argv.includes("--verbose")) {
 		console.log(s)

scripts/interactive-playwright.ts

@@ -0,0 +1,112 @@
+#!/usr/bin/env npx tsx
+
+/**
+ * Interactive Playwright launcher for the Cline VS Code extension.
+ *
+ * Overview:
+ *  - Starts the mock Cline API server (from the e2e test fixtures).
+ *  - Downloads a stable build of VS Code (via @vscode/test-electron).
+ *  - Creates a temporary VS Code user profile directory.
+ *  - Installs and links the Cline extension (from dist/e2e.vsix and the dev path).
+ *  - Opens a test workspace and automatically reveals the Cline sidebar.
+ *  - Records **all gRPC calls** during the session for later inspection.
+ *  - Keeps VS Code running for manual interactive testing until the window is closed or Ctrl+C is pressed.
+ *  - Cleans up all resources (mock server, temp profile, Electron process) on exit.
+ *
+ * Usage:
+ *   1. (Optional) Build and install the e2e extension:
+ *        npm run test:e2e:build
+ *
+ *   2. From the repo root, start the interactive session:
+ *        npm run test:playwright:interactive
+ *
+ *   3. VS Code will launch with the Cline extension loaded and gRPC recording enabled.
+ *
+ *   4. Interact with the extension manually.
+ *
+ *   5. Close the VS Code window or press Ctrl+C to end the session and trigger cleanup.
+ */
+
+import { downloadAndUnzipVSCode, SilentReporter } from "@vscode/test-electron"
+import { mkdtempSync } from "fs"
+import os from "os"
+import path from "path"
+import { _electron } from "playwright"
+import { ClineApiServerMock } from "../src/test/e2e/fixtures/server"
+import { E2ETestHelper } from "../src/test/e2e/utils/helpers"
+
+async function main() {
+	await ClineApiServerMock.startGlobalServer()
+
+	const userDataDir = mkdtempSync(path.join(os.tmpdir(), "vsce-interactive"))
+	const executablePath = await downloadAndUnzipVSCode("stable", undefined, new SilentReporter())
+
+	// launch VSCode
+	const app = await _electron.launch({
+		executablePath,
+		env: {
+			...process.env,
+			TEMP_PROFILE: "true",
+			E2E_TEST: "true",
+			CLINE_ENVIRONMENT: "local",
+			GRPC_RECORDER_ENABLED: "true",
+			GRPC_RECORDER_TESTS_FILTERS_ENABLED: "true",
+		},
+		args: [
+			"--no-sandbox",
+			"--disable-updates",
+			"--disable-workspace-trust",
+			"--disable-extensions",
+			"--skip-welcome",
+			"--skip-release-notes",
+			`--user-data-dir=${userDataDir}`,
+			`--install-extension=${path.join(E2ETestHelper.CODEBASE_ROOT_DIR, "dist", "e2e.vsix")}`,
+			`--extensionDevelopmentPath=${E2ETestHelper.CODEBASE_ROOT_DIR}`,
+			path.join(E2ETestHelper.E2E_TESTS_DIR, "fixtures", "workspace"),
+		],
+	})
+
+	const page = await app.firstWindow()
+
+	await E2ETestHelper.openClineSidebar(page)
+
+	console.log("VSCode with Cline extension is now running!")
+	console.log(`Temporary data directory on: ${userDataDir}`)
+	console.log("You can manually interact with the extension.")
+	console.log("Press Ctrl+C to close when done.")
+
+	async function teardown() {
+		console.log("Cleaning up resources...")
+		try {
+			await app?.close()
+			await ClineApiServerMock.stopGlobalServer?.()
+			await E2ETestHelper.rmForRetries(userDataDir, { recursive: true })
+		} catch (e) {
+			console.log(`We could teardown interactive playwright properly, error:${e}`)
+		}
+		console.log("Finished cleaning up resources...")
+	}
+
+	process.on("SIGINT", async () => {
+		await teardown()
+		process.exit(0)
+	})
+
+	process.on("SIGTERM", async () => {
+		await teardown()
+		process.exit(0)
+	})
+
+	const win = await app.firstWindow()
+	win.on("close", async () => {
+		console.log("VS Code window closed.")
+		await teardown()
+		process.exit(0)
+	})
+	process.stdin.resume()
+}
+
+main().catch((err) => {
+	console.error("Failed to start:", err)
+	process.exit(1)
+})

scripts/package-standalone.mjs

@@ -6,40 +6,102 @@ import fs from "fs"
 import { cp } from "fs/promises"
 import { glob } from "glob"
 import minimatch from "minimatch"
+import os from "os"
 import path from "path"
+import { rmrf } from "./file-utils.mjs"
 
 const BUILD_DIR = "dist-standalone"
+const BINARIES_DIR = `${BUILD_DIR}/binaries`
 const RUNTIME_DEPS_DIR = "standalone/runtime-files"
+const IS_DEBUG_BUILD = process.env.IS_DEBUG_BUILD === "true"
+
+// This should match the node version packaged with the JetBrains plugin.
+const TARGET_NODE_VERSION = "22.15.0"
+const TARGET_PLATFORMS = [
+	{ platform: "win32", arch: "x64", targetDir: "win-x64" },
+	{ platform: "darwin", arch: "x64", targetDir: "darwin-x64" },
+	{ platform: "darwin", arch: "arm64", targetDir: "darwin-arm64" },
+	{ platform: "linux", arch: "x64", targetDir: "linux-x64" },
+]
+const SUPPORTED_BINARY_MODULES = ["better-sqlite3"]
+
+const UNIVERSAL_BUILD = !process.argv.includes("-s")
+const IS_VERBOSE = process.argv.includes("-v") || process.argv.includes("--verbose")
 
 async function main() {
 	await installNodeDependencies()
+	if (UNIVERSAL_BUILD) {
+		console.log("Building universal package for all platforms...")
+		await packageAllBinaryDeps()
+	} else {
+		console.log(`Building package for ${os.platform()}-${os.arch()}...`)
+	}
 	await zipDistribution()
 }
 
 async function installNodeDependencies() {
+	// Clean modules from any previous builds
+	await rmrf(path.join(BUILD_DIR, "node_modules"))
+	await rmrf(path.join(BINARIES_DIR))
+
 	await cpr(RUNTIME_DEPS_DIR, BUILD_DIR)
 
 	console.log("Running npm install in distribution directory...")
-	const cwd = process.cwd()
-	process.chdir(BUILD_DIR)
+	execSync("npm install", { stdio: "inherit", cwd: BUILD_DIR })
 
-	try {
-		execSync("npm install", { stdio: "inherit" })
-		// Move the vscode directory into node_modules.
-		// It can't be installed using npm because it will create a symlink which cannot be unzipped correctly on windows.
-		fs.renameSync("vscode", path.join("node_modules", "vscode"))
-	} catch (error) {
-		console.error("Error during setup:", error)
+	// Move the vscode directory into node_modules.
+	// It can't be installed using npm because it will create a symlink which cannot be unzipped correctly on windows.
+	fs.renameSync(`${BUILD_DIR}/vscode`, `${BUILD_DIR}/node_modules/vscode`)
+}
+
+/**
+ * Downloads prebuilt binaries for each platform for the modules that include binaries. It uses `npx prebuild-install`
+ * to download the binary.
+ *
+ * The modules are downloaded to dist-standalone/binaries/{os}-{platform}/.
+ * When cline-core is installed, the installer should use the correct module for the current platform.
+ */
+async function packageAllBinaryDeps() {
+	// Check for native .node modules.
+	const allNativeModules = await glob("**/*.node", { cwd: path.join(BUILD_DIR, "node_modules"), nodir: true })
+	const isAllowed = (path) => SUPPORTED_BINARY_MODULES.some((allowed) => path.includes(allowed))
+	const blocked = allNativeModules.filter((x) => !isAllowed(x))
+
+	if (blocked.length > 0) {
+		console.error(`Error: Native node modules cannot be included in the standalone distribution:\n\n${blocked.join("\n")}`)
+		console.error(
+			"\nThese modules must support prebuilt-install and be added to the supported list in scripts/package-standalone.mjs",
+		)
 		process.exit(1)
-	} finally {
-		process.chdir(cwd)
 	}
 
-	// Check for native .node modules.
-	const nativeModules = await glob("**/*.node", { cwd: BUILD_DIR, nodir: true })
-	if (nativeModules.length > 0) {
-		console.error("Native node modules cannot be included in the standalone distribution:\n", nativeModules.join("\n"))
-		process.exit(1)
+	for (const module of SUPPORTED_BINARY_MODULES) {
+		console.log(`Installing binaries for ${module}...`)
+		const src = path.join(BUILD_DIR, "node_modules", module)
+		if (!fs.existsSync(src)) {
+			console.warn(`Warning: Trying to install binaries for the module '${module}', but it is not being used by cline.`)
+			continue
+		}
+
+		for (const { platform, arch, targetDir } of TARGET_PLATFORMS) {
+			const binaryDir = `${BINARIES_DIR}/${targetDir}/node_modules`
+			fs.mkdirSync(binaryDir, { recursive: true })
+
+			// Copy the module from the build dir
+			const dest = path.join(binaryDir, module)
+			await cpr(src, dest)
+
+			// Download the binary libs
+			const v = IS_VERBOSE ? "--verbose" : ""
+			const cmd = `npx prebuild-install --platform=${platform} --arch=${arch} --target=${TARGET_NODE_VERSION} ${v}`
+			log_verbose(`${module}: ${cmd}`)
+			execSync(cmd, { cwd: dest, stdio: "inherit" })
+			log_verbose("")
+		}
+		// Remove the original module with the host platform binaries installed directly into node_modules.
+		log_verbose(`Cleaning up host version of ${module}`)
+		await rmrf(src)
+		log_verbose("")
 	}
 }
 
@@ -47,10 +109,13 @@ async function zipDistribution() {
 	// Zip the build directory (excluding any pre-existing output zip).
 	const zipPath = path.join(BUILD_DIR, "standalone.zip")
 	const output = fs.createWriteStream(zipPath)
-	const archive = archiver("zip", { zlib: { level: 3 } })
+	const startTime = Date.now()
+	const archive = archiver("zip", { zlib: { level: 6 } })
 
 	output.on("close", () => {
-		console.log(`Created ${zipPath} (${(archive.pointer() / 1024 / 1024).toFixed(1)} MB)`)
+		const endTime = Date.now()
+		const duration = (endTime - startTime) / 1000
+		console.log(`Created ${zipPath} (${(archive.pointer() / 1024 / 1024).toFixed(1)} MB) in ${duration.toFixed(2)} seconds`)
 	})
 	archive.on("warning", (err) => {
 		console.warn(`Warning: ${err}`)
@@ -73,7 +138,7 @@ async function zipDistribution() {
 	// Add the whole cline directory under "extension", except the for the ignored files.
 	archive.directory(process.cwd(), "extension", (entry) => {
 		if (isIgnored(entry.name)) {
-			log_verbose("Ignoring", entry.name)
+			//log_verbose("Ignoring", entry.name)
 			return false
 		}
 		return entry
@@ -145,7 +210,7 @@ function createIsIgnored(standaloneIgnores) {
 	let allIgnore = [...defaultIgnore, ...expandedIgnore, ...standaloneIgnores]
 
 	// Map files need to be included in the debug build. Remove .map ignores when IS_DEBUG_BUILD is set
-	if (process.env.IS_DEBUG_BUILD) {
+	if (IS_DEBUG_BUILD) {
 		allIgnore = allIgnore.filter((pattern) => !pattern.endsWith(".map"))
 		console.log("Debug build: Including .map files in package")
 	}
@@ -167,6 +232,7 @@ function createIsIgnored(standaloneIgnores) {
 
 /* cp -r */
 async function cpr(source, dest) {
+	log_verbose(`Copying ${source} -> ${dest}`)
 	await cp(source, dest, {
 		recursive: true,
 		preserveTimestamps: true,
@@ -175,7 +241,7 @@ async function cpr(source, dest) {
 }
 
 function log_verbose(...args) {
-	if (process.argv.includes("-v") || process.argv.includes("--verbose")) {
+	if (IS_VERBOSE) {
 		console.log(...args)
 	}
 }

scripts/publish-nightly.mjs

@@ -0,0 +1,377 @@
+#!/usr/bin/env node
+
+/**
+ * Nightly publish script for VS Code extension
+ * Converts package.json to testing version, packages, publishes, and restores
+ *
+ * This script:
+ * 1. Backs up the original package.json
+ * 2. Updates package.json with:
+ *    - New version (major.minor.timestamp format)
+ *    - Changes name to "cline-nightly"
+ *    - Changes displayName to "Cline (Nightly)"
+ * 3. Packages the extension as a .vsix file
+ * 4. Publishes to VS Code Marketplace (if VSCE_PAT is set)
+ * 5. Publishes to OpenVSX Registry (if OVSX_PAT is set)
+ * 6. Restores the original package.json
+ *
+ * Usage:
+ *   npm run publish:marketplace:nightly
+ *   npm run publish:marketplace:nightly -- --dry-run
+ *
+ * Environment variables:
+ *   VSCE_PAT  - Personal Access Token for VS Code Marketplace
+ *   OVSX_PAT  - Personal Access Token for OpenVSX Registry
+ *
+ * Dependencies:
+ *   - vsce (VS Code Extension Manager)
+ *   - ovsx (OpenVSX CLI)
+ */
+
+import { execFileSync, execSync } from "node:child_process"
+import fs from "node:fs"
+import path from "node:path"
+import { fileURLToPath } from "node:url"
+
+// Get __dirname equivalent in ES modules
+const __filename = fileURLToPath(import.meta.url)
+const __dirname = path.dirname(__filename)
+
+// ANSI color codes for console output
+const colors = {
+	reset: "\x1b[0m",
+	red: "\x1b[31m",
+	green: "\x1b[32m",
+	yellow: "\x1b[33m",
+}
+
+// Logging utilities
+const log = {
+	info: (msg) => console.log(`${colors.green}[INFO]${colors.reset} ${msg}`),
+	warn: (msg) => console.log(`${colors.yellow}[WARN]${colors.reset} ${msg}`),
+	error: (msg) => console.error(`${colors.red}[ERROR]${colors.reset} ${msg}`),
+}
+
+// Configuration
+const config = {
+	// The name and display name for the nightly version
+	nightlyName: "cline-nightly",
+	nightlyDisplayName: "Cline (Nightly)",
+	projectRoot: path.join(__dirname, ".."),
+	get packageJsonPath() {
+		return path.join(this.projectRoot, "package.json")
+	},
+	get packageBackupPath() {
+		return path.join(this.projectRoot, "package.json.backup")
+	},
+	get distDir() {
+		return path.join(this.projectRoot, "dist")
+	},
+	get vsixPath() {
+		return path.join(this.distDir, "cline-nightly.vsix")
+	},
+}
+
+// Utility class for managing the publish process
+class NightlyPublisher {
+	constructor() {
+		this.originalPackageJson = null
+		this.hasBackup = false
+	}
+
+	/**
+	 * Check if required dependencies are installed
+	 */
+	checkDependencies() {
+		const dependencies = [
+			{ name: "vsce", check: "vsce --version" },
+			{ name: "npx", check: "npx --version" },
+		]
+
+		const missing = []
+
+		for (const dep of dependencies) {
+			try {
+				execSync(dep.check, { stdio: "ignore" })
+			} catch {
+				missing.push(dep.name)
+			}
+		}
+
+		if (missing.length > 0) {
+			throw new Error(
+				`Missing required dependencies: ${missing.join(", ")}. Please install them before running this script.`,
+			)
+		}
+
+		log.info("All dependencies are installed")
+	}
+
+	/**
+	 * Check if a command exists
+	 */
+	commandExists(command) {
+		try {
+			execSync(`which ${command}`, { stdio: "ignore" })
+			return true
+		} catch {
+			return false
+		}
+	}
+
+	/**
+	 * Create backup of package.json
+	 */
+	backupPackageJson() {
+		if (!fs.existsSync(config.packageJsonPath)) {
+			throw new Error(`package.json not found at ${config.packageJsonPath}`)
+		}
+
+		log.info("Backing up original package.json")
+		this.originalPackageJson = fs.readFileSync(config.packageJsonPath, "utf-8")
+		fs.writeFileSync(config.packageBackupPath, this.originalPackageJson)
+		this.hasBackup = true
+	}
+
+	/**
+	 * Restore original package.json
+	 */
+	restorePackageJson() {
+		if (this.hasBackup && fs.existsSync(config.packageBackupPath)) {
+			log.info("Restoring original package.json")
+			fs.writeFileSync(config.packageJsonPath, this.originalPackageJson)
+			fs.unlinkSync(config.packageBackupPath)
+			this.hasBackup = false
+		}
+	}
+
+	/**
+	 * Generate new version with timestamp
+	 * Format: major.minor.timestamp
+	 */
+	generateVersion(currentVersion) {
+		// Extract major.minor from current version (e.g., "3.27.1" -> "3.27")
+		const versionParts = currentVersion.split(".")
+		if (versionParts.length < 2) {
+			throw new Error(`Invalid version format: ${currentVersion}`)
+		}
+
+		const major = versionParts[0]
+		const minor = versionParts[1]
+		const timestamp = Math.floor(Date.now() / 1000)
+
+		return `${major}.${minor}.${timestamp}`
+	}
+
+	/**
+	 * Update package.json with nightly configuration
+	 */
+	updatePackageJson() {
+		// Replace any occurrences cline. or claude-dev with nightly name
+		const rawContent = fs.readFileSync(config.packageJsonPath, "utf-8")
+		const content = rawContent.replaceAll("claude-dev", config.nightlyName).replaceAll('"cline.', `"${config.nightlyName}.`)
+
+		const pkg = JSON.parse(content)
+		const currentVersion = pkg.version
+
+		if (!currentVersion) {
+			throw new Error("Could not read version from package.json")
+		}
+
+		log.info(`Current version: ${currentVersion}`)
+
+		const newVersion = this.generateVersion(currentVersion)
+		log.info(`New version: ${newVersion}`)
+
+		// Update package.json fields
+		pkg.version = newVersion
+		pkg.name = config.nightlyName
+		pkg.displayName = config.nightlyDisplayName
+		pkg.contributes.viewsContainers.activitybar.title = config.nightlyDisplayName
+
+		// Save updated package.json
+		log.info("Updating package.json for nightly build")
+		fs.writeFileSync(config.packageJsonPath, JSON.stringify(pkg, null, "\t"))
+
+		return newVersion
+	}
+
+	/**
+	 * Package the extension
+	 */
+	packageExtension() {
+		// Ensure dist directory exists
+		if (!fs.existsSync(config.distDir)) {
+			fs.mkdirSync(config.distDir, { recursive: true })
+		}
+
+		log.info("Packaging extension")
+
+		const args = ["package", "--pre-release", "--no-update-package-json", "--no-git-tag-version", "--out", config.vsixPath]
+
+		try {
+			execFileSync("vsce", args, {
+				stdio: "inherit",
+				cwd: config.projectRoot,
+			})
+			log.info(`Package created: ${config.vsixPath}`)
+		} catch (error) {
+			throw new Error(`Failed to package extension: ${error.message}`)
+		}
+	}
+
+	/**
+	 * Publish to VS Code Marketplace
+	 */
+	publishToVSCodeMarketplace() {
+		const token = process.env.VSCE_PAT
+
+		if (!token) {
+			log.warn("VSCE_PAT not set, skipping VS Code Marketplace publish")
+			return false
+		}
+
+		log.info("Publishing to VS Code Marketplace")
+
+		const args = ["publish", "--pre-release", "--no-git-tag-version", "--packagePath", config.vsixPath]
+
+		try {
+			execFileSync("vsce", args, {
+				env: { ...process.env, VSCE_PAT: token },
+				stdio: "inherit",
+				cwd: config.projectRoot,
+			})
+			log.info("Successfully published to VS Code Marketplace")
+			return true
+		} catch (error) {
+			throw new Error(`Failed to publish to VS Code Marketplace: ${error.message}`)
+		}
+	}
+
+	/**
+	 * Publish to OpenVSX Registry
+	 */
+	publishToOpenVSX() {
+		const token = process.env.OVSX_PAT
+
+		if (!token) {
+			log.warn("OVSX_PAT not set, skipping OpenVSX Registry publish")
+			return false
+		}
+
+		log.info("Publishing to OpenVSX Registry")
+
+		const args = ["ovsx", "publish", "--pre-release", "--packagePath", config.vsixPath, "--pat", token]
+
+		try {
+			execFileSync("npx", args, {
+				stdio: "inherit",
+				cwd: config.projectRoot,
+			})
+			log.info("Successfully published to OpenVSX Registry")
+			return true
+		} catch (error) {
+			throw new Error(`Failed to publish to OpenVSX Registry: ${error.message}`)
+		}
+	}
+
+	/**
+	 * Main execution flow
+	 */
+	async run(isDryRun = false) {
+		try {
+			log.info(`Starting nightly publish process${isDryRun ? " (dry run)" : ""}`)
+
+			// Step 1: Check dependencies
+			this.checkDependencies()
+
+			// Step 2: Backup package.json
+			this.backupPackageJson()
+
+			// Step 3: Update package.json
+			const newVersion = this.updatePackageJson()
+
+			// Step 4: Package extension
+			this.packageExtension()
+
+			// Step 5: Publish to marketplaces (skip if dry run)
+			let vsCodePublished = false
+			let openVSXPublished = false
+
+			if (isDryRun) {
+				log.info("Dry run mode: Skipping marketplace publishing")
+			} else {
+				vsCodePublished = this.publishToVSCodeMarketplace()
+				openVSXPublished = this.publishToOpenVSX()
+			}
+
+			// Summary
+			log.info(`Nightly publish process completed successfully${isDryRun ? " (dry run)" : ""}`)
+			log.info(`Package created for v${newVersion}: ${config.vsixPath}`)
+
+			if (!isDryRun && !vsCodePublished && !openVSXPublished) {
+				log.warn("Extension was packaged but not published to any marketplace")
+				log.warn("Set VSCE_PAT and/or OVSX_PAT environment variables to enable publishing")
+			}
+		} catch (error) {
+			log.error(`Publish failed: ${error.message}`)
+			process.exit(1)
+		} finally {
+			// Always restore package.json
+			this.restorePackageJson()
+		}
+	}
+}
+
+// Handle cleanup on process exit
+const publisher = new NightlyPublisher()
+
+process.on("exit", () => {
+	publisher.restorePackageJson()
+})
+
+process.on("SIGINT", () => {
+	log.info("\nInterrupted, cleaning up...")
+	publisher.restorePackageJson()
+	process.exit(130)
+})
+
+process.on("SIGTERM", () => {
+	log.info("\nTerminated, cleaning up...")
+	publisher.restorePackageJson()
+	process.exit(143)
+})
+
+// Parse command line arguments
+const args = process.argv.slice(2)
+const isDryRun = args.includes("--dry-run") || args.includes("-n")
+const showHelp = args.includes("--help") || args.includes("-h")
+
+if (showHelp) {
+	console.log(`
+Nightly publish script for VS Code extension
+
+Usage:
+  npm run publish:marketplace:nightly [options]
+
+Options:
+  --dry-run, -n    Run without actually publishing (package only)
+  --help, -h       Show this help message
+
+Environment variables:
+  VSCE_PAT         Personal Access Token for VS Code Marketplace
+  OVSX_PAT         Personal Access Token for OpenVSX Registry
+
+Examples:
+  npm run publish:marketplace:nightly                    # Full publish
+  npm run publish:marketplace:nightly  -- --dry-run      # Package only
+  VSCE_PAT="token" npm run publish:marketplace:nightly   # Publish to VS Code only
+`)
+	process.exit(0)
+}
+
+// Run the publisher
+publisher.run(isDryRun).catch((error) => {
+	log.error(error.message)
+	process.exit(1)
+})

scripts/runclinecore.sh

@@ -9,7 +9,7 @@ if [[ "${1:-}" == "-h" ]]; then
 fi
 
 CORE_DIR=~/.cline/core
-INSTALL_DIR=$CORE_DIR/0.0.1
+INSTALL_DIR=$CORE_DIR/dev-instance/
 LOG_FILE=~/.cline/cline-core-service.log
 
 ZIP_FILE=standalone.zip
@@ -25,4 +25,6 @@ unp $ZIP_FILE > /dev/null
 
 pkill -f cline-core.js || true
 
+echo pwd: $(pwd)
+set -x
 NODE_PATH=./node_modules DEV_WORKSPACE_FOLDER=/tmp/ node cline-core.js 2>&1 | tee $LOG_FILE

scripts/test-hostbridge-server.ts

@@ -3,7 +3,11 @@ import * as grpc from "@grpc/grpc-js"
 import { ReflectionService } from "@grpc/reflection"
 import * as health from "grpc-health-check"
 import * as os from "os"
-import { host } from "src/generated/grpc-js/index"
+import { type DiffServiceServer, DiffServiceService } from "../src/generated/grpc-js/host/diff"
+import { type EnvServiceServer, EnvServiceService } from "../src/generated/grpc-js/host/env"
+import { type TestingServiceServer, TestingServiceService } from "../src/generated/grpc-js/host/testing"
+import { type WindowServiceServer, WindowServiceService } from "../src/generated/grpc-js/host/window"
+import { type WorkspaceServiceServer, WorkspaceServiceService } from "../src/generated/grpc-js/host/workspace"
 import { getPackageDefinition } from "./proto-utils.mjs"
 
 export async function startTestHostBridgeServer() {
@@ -14,11 +18,11 @@ export async function startTestHostBridgeServer() {
 	healthImpl.addToServer(server)
 
 	// Add host bridge services using the mock implementations
-	server.addService(host.WorkspaceServiceService, createMockService<host.WorkspaceServiceServer>("WorkspaceService"))
-	server.addService(host.WindowServiceService, createMockService<host.WindowServiceServer>("WindowService"))
-	server.addService(host.EnvServiceService, createMockService<host.EnvServiceServer>("EnvService"))
-	server.addService(host.DiffServiceService, createMockService<host.DiffServiceServer>("DiffService"))
-	server.addService(host.WatchServiceService, createMockService<host.WatchServiceServer>("WatchService"))
+	server.addService(WorkspaceServiceService, createMockService<WorkspaceServiceServer>("WorkspaceService"))
+	server.addService(WindowServiceService, createMockService<WindowServiceServer>("WindowService"))
+	server.addService(EnvServiceService, createMockService<EnvServiceServer>("EnvService"))
+	server.addService(DiffServiceService, createMockService<DiffServiceServer>("DiffService"))
+	server.addService(TestingServiceService, createMockService<TestingServiceServer>("TestingService"))
 
 	// Load package definition for reflection service
 	const packageDefinition = await getPackageDefinition()
@@ -58,8 +62,9 @@ function createMockService<T extends grpc.UntypedServiceImplementation>(serviceN
 				// Special cases that need specific return values
 				switch (prop) {
 					case "getWorkspacePaths":
+						const workspaceDir = process.env.TEST_HOSTBRIDGE_WORKSPACE_DIR || "/test-workspace"
 						callback(null, {
-							paths: ["/test-workspace"],
+							paths: [workspaceDir],
 						})
 						return
 
@@ -69,6 +74,12 @@ function createMockService<T extends grpc.UntypedServiceImplementation>(serviceN
 						})
 						return
 
+					case "getTelemetrySettings":
+						callback(null, {
+							isEnabled: 2, // Setting.DISABLED
+						})
+						return
+
 					case "clipboardReadText":
 						callback(null, {
 							value: "",
@@ -115,8 +126,8 @@ function createMockService<T extends grpc.UntypedServiceImplementation>(serviceN
 						})
 						return
 
-					// For streaming methods (like subscribeToFile)
-					case "subscribeToFile":
+					// For streaming methods (like subscribeToTelemetrySettings)
+					case "subscribeToTelemetrySettings":
 						// Just end the stream immediately
 						call.end()
 						return

scripts/test-standalone-core-api-server.ts

@@ -0,0 +1,183 @@
+#!/usr/bin/env npx tsx
+
+/**
+ * Simple Cline gRPC Server
+ *
+ * This script provides a minimal way to run the Cline core gRPC service
+ * without requiring the full installation, while automatically mocking all external services. Simply run:
+ *
+ *   # One-time setup (generates protobuf files)
+ *	 npm run compile-standalone
+ *   npm run test:sca-server
+ *
+ * The following components are started automatically:
+ *   1. HostBridge test server
+ *   2. ClineApiServerMock (mock implementation of the Cline API)
+ *   3. AuthServiceMock (activated if E2E_TEST="true")
+ *
+ * Environment Variables for Customization:
+ *   PROJECT_ROOT - Override project root directory (default: parent of scripts dir)
+ *   CLINE_DIST_DIR - Override distribution directory (default: PROJECT_ROOT/dist-standalone)
+ *   CLINE_CORE_FILE - Override core file name (default: cline-core.js)
+ *   PROTOBUS_PORT - gRPC server port (default: 26040)
+ *   HOSTBRIDGE_PORT - HostBridge server port (default: 26041)
+ *   WORKSPACE_DIR - Working directory (default: current directory)
+ *   E2E_TEST - Enable E2E test mode (default: true)
+ *   CLINE_ENVIRONMENT - Environment setting (default: local)
+ *
+ * Ideal for local development, testing, or lightweight E2E scenarios.
+ */
+
+import * as fs from "node:fs"
+import { mkdtempSync, rmSync } from "node:fs"
+import * as os from "node:os"
+import { ChildProcess, execSync, spawn } from "child_process"
+import * as path from "path"
+import { ClineApiServerMock } from "../src/test/e2e/fixtures/server/index"
+
+const PROTOBUS_PORT = process.env.PROTOBUS_PORT || "26040"
+const HOSTBRIDGE_PORT = process.env.HOSTBRIDGE_PORT || "26041"
+const WORKSPACE_DIR = process.env.WORKSPACE_DIR || process.cwd()
+const E2E_TEST = process.env.E2E_TEST || "true"
+const CLINE_ENVIRONMENT = process.env.CLINE_ENVIRONMENT || "local"
+const USE_C8 = process.env.USE_C8 === "true"
+
+// Locate the standalone build directory and core file with flexible path resolution
+const projectRoot = process.env.PROJECT_ROOT || path.resolve(__dirname, "..")
+const distDir = process.env.CLINE_DIST_DIR || path.join(projectRoot, "dist-standalone")
+const clineCoreFile = process.env.CLINE_CORE_FILE || "cline-core.js"
+const coreFile = path.join(distDir, clineCoreFile)
+
+const childProcesses: ChildProcess[] = []
+
+async function main(): Promise<void> {
+	console.log("Starting Simple Cline gRPC Server...")
+	console.log(`Project Root: ${projectRoot}`)
+	console.log(`Workspace: ${WORKSPACE_DIR}`)
+	console.log(`ProtoBus Port: ${PROTOBUS_PORT}`)
+	console.log(`HostBridge Port: ${HOSTBRIDGE_PORT}`)
+
+	console.log(`Looking for standalone build at: ${coreFile}`)
+
+	if (!fs.existsSync(coreFile)) {
+		console.error(`Standalone build not found at: ${coreFile}`)
+		console.error("Available environment variables for customization:")
+		console.error("  PROJECT_ROOT - Override project root directory")
+		console.error("  CLINE_DIST_DIR - Override distribution directory")
+		console.error("  CLINE_CORE_FILE - Override core file name")
+		console.error("")
+		console.error("To build the standalone version, run: npm run compile-standalone")
+		process.exit(1)
+	}
+
+	try {
+		await ClineApiServerMock.startGlobalServer()
+		console.log("Cline API Server started in-process")
+	} catch (error) {
+		console.error("Failed to start Cline API Server:", error)
+		process.exit(1)
+	}
+
+	const extensionsDir = path.join(distDir, "vsce-extension")
+	const userDataDir = mkdtempSync(path.join(os.tmpdir(), "vsce"))
+	const clineTestWorkspace = mkdtempSync(path.join(os.tmpdir(), "cline-test-workspace-"))
+
+	console.log("Starting HostBridge test server...")
+	const hostbridge: ChildProcess = spawn("npx", ["tsx", path.join(__dirname, "test-hostbridge-server.ts")], {
+		stdio: "pipe",
+		env: {
+			...process.env,
+			TEST_HOSTBRIDGE_WORKSPACE_DIR: clineTestWorkspace,
+			HOST_BRIDGE_ADDRESS: `127.0.0.1:${HOSTBRIDGE_PORT}`,
+		},
+	})
+	childProcesses.push(hostbridge)
+
+	console.log(`Temp user data dir: ${userDataDir}`)
+	console.log(`Temp extensions dir: ${extensionsDir}`)
+	// Extract standalone.zip if needed
+	const standaloneZipPath = path.join(distDir, "standalone.zip")
+	if (!fs.existsSync(standaloneZipPath)) {
+		console.error(`standalone.zip not found at: ${standaloneZipPath}`)
+		process.exit(1)
+	}
+
+	console.log("Extracting standalone.zip to extensions directory...")
+	try {
+		if (!fs.existsSync(extensionsDir)) {
+			execSync(`unzip -q "${standaloneZipPath}" -d "${extensionsDir}"`, { stdio: "inherit" })
+		}
+		console.log(`Successfully extracted standalone.zip to: ${extensionsDir}`)
+	} catch (error) {
+		console.error("Failed to extract standalone.zip:", error)
+		process.exit(1)
+	}
+
+	const covDir = path.join(projectRoot, `coverage/coverage-core-${PROTOBUS_PORT}`)
+
+	const baseArgs = ["--enable-source-maps", path.join(distDir, "cline-core.js")]
+
+	const spawnArgs = USE_C8 ? ["c8", "--report-dir", covDir, "node", ...baseArgs] : ["node", ...baseArgs]
+
+	console.log(`Starting Cline Core Service... (useC8=${USE_C8})`)
+
+	const coreService: ChildProcess = spawn("npx", spawnArgs, {
+		cwd: projectRoot,
+		env: {
+			...process.env,
+			NODE_PATH: "./node_modules",
+			DEV_WORKSPACE_FOLDER: WORKSPACE_DIR,
+			PROTOBUS_ADDRESS: `127.0.0.1:${PROTOBUS_PORT}`,
+			HOST_BRIDGE_ADDRESS: `localhost:${HOSTBRIDGE_PORT}`,
+			E2E_TEST,
+			CLINE_ENVIRONMENT,
+			CLINE_DIR: userDataDir,
+			INSTALL_DIR: extensionsDir,
+		},
+		stdio: "inherit",
+	})
+	childProcesses.push(coreService)
+
+	const shutdown = async () => {
+		console.log("\nShutting down services...")
+
+		while (childProcesses.length > 0) {
+			const child = childProcesses.pop()
+			if (child && !child.killed) child.kill("SIGINT")
+		}
+
+		await ClineApiServerMock.stopGlobalServer()
+
+		try {
+			rmSync(userDataDir, { recursive: true, force: true })
+			rmSync(clineTestWorkspace, { recursive: true, force: true })
+			console.log("Cleaned up temporary directories")
+		} catch (err) {
+			console.warn("Failed to cleanup temp directories:", err)
+		}
+
+		process.exit(0)
+	}
+
+	process.on("SIGINT", shutdown)
+	process.on("SIGTERM", shutdown)
+
+	coreService.on("exit", (code) => {
+		console.log(`Core service exited with code ${code}`)
+		shutdown()
+	})
+	hostbridge.on("exit", (code) => {
+		console.log(`HostBridge exited with code ${code}`)
+		shutdown()
+	})
+
+	console.log(`Cline gRPC Server is running on 127.0.0.1:${PROTOBUS_PORT}`)
+	console.log("Press Ctrl+C to stop")
+}
+
+if (require.main === module) {
+	main().catch((err) => {
+		console.error("Failed to start simple Cline server:", err)
+		process.exit(1)
+	})
+}

scripts/testing-platform-orchestrator.ts

@@ -0,0 +1,229 @@
+#!/usr/bin/env npx tsx
+/**
+ * Test Orchestrator
+ *
+ * Automates server lifecycle for running spec files against the standalone server.
+ *
+ * Prerequisites:
+ *   Build standalone first: `npm run compile-standalone`
+ *
+ * Usage:
+ *   - Single file:   `npm run test:tp-orchestrator path/to/spec.json`
+ *   - All specs dir: `npm run test:tp-orchestrator tests/specs`
+ *
+ * Flags:
+ *   --server-logs        Show server logs (hidden by default)
+ *   --count=<number>     Repeat execution N times (default: 1)
+ *   --fix     			  Automatically update spec files with actual responses
+ *   --coverage     	  Generate integration test coverage information
+ *
+ */
+
+import { ChildProcess, spawn } from "child_process"
+import fs from "fs"
+import minimist from "minimist"
+import net from "net"
+import path from "path"
+import kill from "tree-kill"
+
+let showServerLogs = false
+let fix = false
+let coverage = false
+const WAIT_SERVER_DEFAULT_TIMEOUT = 15000
+const usedPorts = new Set<number>()
+
+/**
+ * Find an available TCP port within the given range [min, max].
+ *
+ * - Ports are allocated sequentially (starting at `min`) rather than randomly,
+ *   which avoids accidental reuse when running hundreds of tests in a row.
+ * - Each successfully allocated port is tracked in `usedPorts` to guarantee
+ *   it is never handed out again within the lifetime of this orchestrator.
+ * - Before returning, the function binds a temporary server to the port to
+ *   verify that the OS really considers it available, then immediately closes it.
+ *
+ * This approach makes the orchestrator much more robust on CI (e.g. GitHub Actions),
+ * where a just-terminated server may leave its socket in TIME_WAIT and cause
+ * flakiness if the same port is reallocated too soon.
+ */
+async function getAvailablePort(min = 20000, max = 49151): Promise<number> {
+	return new Promise((resolve, _) => {
+		const tryPort = (candidate?: number) => {
+			const port = candidate ?? Math.floor(Math.random() * (max - min + 1)) + min
+			if (usedPorts.has(port)) {
+				// already allocated in this run
+				return tryPort()
+			}
+			const server = net.createServer()
+			server.once("error", () => tryPort())
+			server.once("listening", () => {
+				server.close(() => {
+					usedPorts.add(port) // mark reserved
+					resolve(port)
+				})
+			})
+			server.listen(port, "127.0.0.1")
+		}
+		tryPort()
+	})
+}
+
+// Poll until a given TCP port on a host is accepting connections.
+async function waitForPort(port: number, host = "127.0.0.1", timeout = 10000): Promise<void> {
+	const start = Date.now()
+	const waitForPortSleepMs = 100
+	while (Date.now() - start < timeout) {
+		await new Promise((res) => setTimeout(res, waitForPortSleepMs))
+		try {
+			await new Promise<void>((resolve, reject) => {
+				const socket = net.connect(port, host, () => {
+					socket.destroy()
+					resolve()
+				})
+				socket.on("error", reject)
+			})
+			return
+		} catch {
+			// try again
+		}
+	}
+	throw new Error(`Timeout waiting for ${host}:${port}`)
+}
+
+async function startServer(): Promise<{ server: ChildProcess; grpcPort: string }> {
+	const grpcPort = (await getAvailablePort()).toString()
+	const hostbridgePort = (await getAvailablePort()).toString()
+
+	const server = spawn("npx", ["tsx", "scripts/test-standalone-core-api-server.ts"], {
+		stdio: showServerLogs ? "inherit" : "pipe",
+		env: {
+			...process.env,
+			PROTOBUS_PORT: grpcPort,
+			HOSTBRIDGE_PORT: hostbridgePort,
+			USE_C8: coverage ? "true" : "false",
+		},
+	})
+
+	// Wait for either the server to become ready or fail on spawn error
+	await Promise.race([
+		waitForPort(Number(grpcPort), "127.0.0.1", WAIT_SERVER_DEFAULT_TIMEOUT),
+		new Promise((_, reject) => server.once("error", reject)),
+	])
+
+	return { server, grpcPort }
+}
+
+function stopServer(server: ChildProcess): Promise<void> {
+	return new Promise((resolve) => {
+		if (!server.pid) return resolve()
+
+		kill(server.pid, "SIGINT", (err) => {
+			if (err) console.warn("Failed to kill server process:", err)
+			server.once("exit", () => resolve())
+		})
+	})
+}
+
+function runTestingPlatform(specFile: string, grpcPort: string): Promise<void> {
+	return new Promise((resolve, reject) => {
+		const testProcess = spawn("npx", ["ts-node", "index.ts", specFile, ...(fix ? ["--fix"] : [])], {
+			cwd: path.join(process.cwd(), "testing-platform"),
+			stdio: "inherit",
+			env: {
+				...process.env,
+				STANDALONE_GRPC_SERVER_PORT: grpcPort,
+			},
+		})
+
+		testProcess.once("error", reject)
+		testProcess.once("exit", (code) => {
+			code === 0 ? resolve() : reject(new Error(`Exit code ${code}`))
+		})
+	})
+}
+
+async function runSpec(specFile: string): Promise<void> {
+	const { server, grpcPort } = await startServer()
+	try {
+		await runTestingPlatform(specFile, grpcPort)
+		console.log(`✅ ${path.basename(specFile)} passed`)
+	} finally {
+		await stopServer(server)
+	}
+}
+
+function collectSpecFiles(inputPath: string): string[] {
+	const fullPath = path.resolve(inputPath)
+	if (!fs.existsSync(fullPath)) throw new Error(`Path does not exist: ${fullPath}`)
+
+	const stat = fs.statSync(fullPath)
+	if (stat.isDirectory()) {
+		return fs
+			.readdirSync(fullPath)
+			.filter((f) => f.endsWith(".json"))
+			.map((f) => path.join(fullPath, f))
+	}
+	if (fullPath.endsWith(".json")) return [fullPath]
+	throw new Error("Spec path must be a JSON file or a folder containing JSON files")
+}
+
+async function runAll(inputPath: string, count: number) {
+	const specFiles = collectSpecFiles(inputPath)
+	if (specFiles.length === 0) {
+		console.warn(`⚠️ No spec files found in ${inputPath}`)
+		return
+	}
+
+	let success = 0
+	let failure = 0
+	const totalStart = Date.now()
+
+	for (let i = 0; i < count; i++) {
+		console.log(`\n🔁 Run #${i + 1} of ${count}`)
+		for (const specFile of specFiles) {
+			try {
+				await runSpec(specFile)
+				success++
+			} catch (err) {
+				console.error(`❌ run #${i + 1}: ${path.basename(specFile)} failed:`, (err as Error).message)
+				failure++
+			}
+		}
+
+		if (failure > 0) process.exitCode = 1
+	}
+
+	console.log(`✅ Passed: ${success}`)
+	if (failure > 0) {
+		console.log(
+			`❌ Test Failed: ${failure}. Please review the integration tests in the /tests/specs directory and try rerunning them with the --fix flag.`,
+		)
+	}
+	console.log(`📋 Total specs: ${specFiles.length} Total runs: ${specFiles.length * count}`)
+	console.log(`🏁 All runs completed in ${((Date.now() - totalStart) / 1000).toFixed(2)}s`)
+}
+
+async function main() {
+	const args = minimist(process.argv.slice(2), { default: { count: 1 } })
+	const inputPath = args._[0]
+	const count = Number(args.count)
+	showServerLogs = Boolean(args["server-logs"])
+	fix = Boolean(args["fix"])
+	coverage = Boolean(args["coverage"])
+
+	if (!inputPath) {
+		console.error(
+			"Usage: npx tsx scripts/testing-platform-orchestrator.ts <spec-file-or-folder> [--count=N] [--server-logs] [--fix] [--coverage]",
+		)
+		process.exit(1)
+	}
+
+	await runAll(inputPath, count)
+}
+
+if (require.main === module) {
+	main().catch((err) => {
+		console.error("❌ Fatal error:", err)
+		process.exit(1)
+	})
+}

src/api/providers/dify.ts

@@ -0,0 +1,293 @@
+import { Anthropic } from "@anthropic-ai/sdk"
+import { ModelInfo } from "@shared/api"
+import { ApiHandler } from "../../core/api/index"
+import { ApiStream } from "../../core/api/transform/stream"
+
+interface DifyHandlerOptions {
+	difyApiKey?: string
+	difyBaseUrl?: string
+}
+
+export class DifyHandler implements ApiHandler {
+	private options: DifyHandlerOptions
+	private baseUrl: string
+	private apiKey: string
+	private conversationId: string | null = null
+
+	constructor(options: DifyHandlerOptions) {
+		this.options = options
+		this.apiKey = options.difyApiKey || ""
+		this.baseUrl = options.difyBaseUrl || ""
+
+		console.log("[DIFY DEBUG] Constructor called with:", {
+			hasApiKey: !!this.apiKey,
+			baseUrl: this.baseUrl,
+		})
+
+		if (!this.apiKey) {
+			throw new Error("Dify API key is required")
+		}
+		if (!this.baseUrl) {
+			throw new Error("Dify base URL is required")
+		}
+	}
+
+	async *createMessage(systemPrompt: string, messages: Anthropic.Messages.MessageParam[]): ApiStream {
+		console.log("[DIFY DEBUG] createMessage called with:", {
+			systemPromptLength: systemPrompt?.length || 0,
+			messagesCount: messages?.length || 0,
+		})
+
+		// Convert messages to Dify format
+		const query = this.convertMessagesToQuery(systemPrompt, messages)
+		const requestBody = {
+			inputs: {},
+			query: query,
+			response_mode: "streaming",
+			conversation_id: this.conversationId || "",
+			user: "cline-user", // A unique user identifier
+			files: [],
+		}
+
+		const fullUrl = `${this.baseUrl}/chat-messages`
+		console.log("[DIFY DEBUG] Making request to:", fullUrl)
+		console.log("[DIFY DEBUG] Request body:", JSON.stringify(requestBody, null, 2))
+		console.log("[DIFY DEBUG] Current process environment variables (for proxy debugging):", process.env)
+
+		let response: Response
+		try {
+			response = await fetch(fullUrl, {
+				method: "POST",
+				headers: {
+					Authorization: `Bearer ${this.apiKey}`,
+					"Content-Type": "application/json",
+				},
+				body: JSON.stringify(requestBody),
+			})
+		} catch (error: any) {
+			console.error("[DIFY DEBUG] Network error during fetch:", error)
+			// Log more detailed error information if available (e.g., from undici)
+			const cause = error.cause ? ` | Cause: ${error.cause}` : ""
+			throw new Error(`Dify API network error: ${error.message}${cause}`)
+		}
+
+		console.log("[DIFY DEBUG] Response status:", response.status)
+		const headersObj: Record<string, string> = {}
+		response.headers.forEach((value, key) => {
+			headersObj[key] = value
+		})
+		console.log("[DIFY DEBUG] Response headers:", headersObj)
+
+		if (!response.ok) {
+			const errorText = await response.text()
+			console.error("[DIFY DEBUG] Error response:", errorText)
+			throw new Error(`Dify API error: ${response.status} ${response.statusText} - ${errorText}`)
+		}
+
+		if (!response.body) {
+			throw new Error("No response body from Dify API")
+		}
+
+		const reader = response.body.getReader()
+		const decoder = new TextDecoder()
+		let buffer = ""
+		let fullText = ""
+
+		console.log("[DIFY DEBUG] Starting to read streaming response...")
+
+		try {
+			while (true) {
+				const { done, value } = await reader.read()
+				if (done) {
+					console.log("[DIFY DEBUG] Stream ended naturally")
+					break
+				}
+
+				const chunk = decoder.decode(value, { stream: true })
+				console.log("[DIFY DEBUG] Raw chunk received:", JSON.stringify(chunk))
+
+				buffer += chunk
+				const lines = buffer.split("\n")
+
+				// Keep the last incomplete line in the buffer
+				buffer = lines.pop() || ""
+
+				for (const line of lines) {
+					console.log("[DIFY DEBUG] Processing line:", JSON.stringify(line))
+
+					if (line.startsWith("data: ")) {
+						const data = line.slice(6).trim()
+						console.log("[DIFY DEBUG] Extracted data:", JSON.stringify(data))
+
+						if (data === "[DONE]") {
+							console.log("[DIFY DEBUG] Received [DONE] signal")
+							return // Explicitly return on [DONE]
+						}
+
+						if (data === "") {
+							console.log("[DIFY DEBUG] Empty data line, skipping")
+							continue
+						}
+
+						try {
+							const parsed = JSON.parse(data)
+							console.log("[DIFY DEBUG] Parsed JSON:", parsed)
+
+							// Capture conversation_id as soon as it's available
+							if (parsed.conversation_id && !this.conversationId) {
+								this.conversationId = parsed.conversation_id
+								console.log("[DIFY DEBUG] Captured conversation_id:", this.conversationId)
+							}
+
+							// Handle different Dify event types based on actual Dify API
+							if (parsed.event === "message") {
+								console.log("[DIFY DEBUG] Message event, answer:", parsed.answer)
+								// Dify sends the full text in each "answer" chunk, so we replace.
+								if (typeof parsed.answer === "string") {
+									fullText = parsed.answer
+									console.log("[DIFY DEBUG] Updated fullText length:", fullText.length)
+									yield {
+										type: "text",
+										text: fullText,
+									}
+								}
+							} else if (parsed.event === "message_replace") {
+								console.log("[DIFY DEBUG] Replace message event:", parsed)
+								if (parsed.answer) {
+									fullText = parsed.answer // Replace instead of append
+									console.log("[DIFY DEBUG] Replaced fullText length:", fullText.length)
+									yield {
+										type: "text",
+										text: fullText,
+									}
+								}
+							} else if (parsed.event === "message_end") {
+								console.log("[DIFY DEBUG] Message end event", parsed)
+								// Message completed. Yield final text if we have any.
+								if (fullText) {
+									yield {
+										type: "text",
+										text: fullText,
+									}
+								}
+								// Yield usage data if available
+								if (parsed.usage) {
+									yield {
+										type: "usage",
+										inputTokens: parsed.usage.prompt_tokens || 0,
+										outputTokens: parsed.usage.completion_tokens || parsed.usage.total_tokens || 0,
+										totalCost: parsed.usage.total_price || 0,
+									}
+								}
+								return // End of stream
+							} else if (parsed.event === "error") {
+								console.error("[DIFY DEBUG] Error event:", parsed)
+								throw new Error(`Dify API error: ${parsed.message || "Unknown error"}`)
+							} else if (parsed.event === "workflow_started" || parsed.event === "workflow_finished") {
+								console.log("[DIFY DEBUG] Workflow event:", parsed.event)
+								// These are informational events, continue processing
+							} else if (parsed.event === "node_started" || parsed.event === "node_finished") {
+								console.log("[DIFY DEBUG] Node event:", parsed.event, parsed.data)
+								// These are informational events, continue processing
+							} else if (parsed.event === "ping") {
+								console.log("[DIFY DEBUG] Ping event received, keeping connection alive.")
+								// Ping event, do nothing
+							} else {
+								console.log("[DIFY DEBUG] Unknown event type:", parsed.event, "Full object:", parsed)
+								// Try to extract text from other possible fields
+								if (parsed.text) {
+									fullText += parsed.text
+									yield {
+										type: "text",
+										text: fullText,
+									}
+								} else if (parsed.content) {
+									fullText += parsed.content
+									yield {
+										type: "text",
+										text: fullText,
+									}
+								}
+							}
+						} catch (e) {
+							console.warn("[DIFY DEBUG] Failed to parse JSON:", data, "Error:", e)
+						}
+					} else if (line.trim() !== "") {
+						console.log(
+							"[DIFY DEBUG] Non-data line (not starting with 'data:'), trying to parse as direct JSON:",
+							JSON.stringify(line),
+						)
+						// Try to parse as direct JSON (fallback for non-SSE responses, though Dify uses SSE)
+						try {
+							const parsed = JSON.parse(line.trim())
+							console.log("[DIFY DEBUG] Parsed direct JSON:", parsed)
+
+							// Handle the same event types as above
+							if (parsed.event === "message" && parsed.answer) {
+								fullText += parsed.answer
+								yield {
+									type: "text",
+									text: fullText,
+								}
+							} else if (parsed.event === "message_end") {
+								if (fullText) {
+									yield {
+										type: "text",
+										text: fullText,
+									}
+								}
+								return
+							} else if (parsed.event === "error") {
+								console.error("[DIFY DEBUG] Direct JSON Error event:", parsed)
+								throw new Error(`Dify API error: ${parsed.message || "Unknown error"}`)
+							}
+						} catch (e) {
+							// Not JSON, continue
+							console.log("[DIFY DEBUG] Line is not direct JSON, continuing")
+						}
+					}
+				}
+			}
+		} finally {
+			reader.releaseLock()
+			console.log("[DIFY DEBUG] Stream reader released")
+		}
+	}
+
+	private convertMessagesToQuery(systemPrompt: string, messages: Anthropic.Messages.MessageParam[]): string {
+		// Dify's context is managed by `conversation_id`. The `query` should be the last user message.
+		// The system prompt is typically configured in the Dify App itself.
+		const lastUserMessage = messages.filter((m) => m.role === "user").pop()
+
+		if (!lastUserMessage) {
+			return "" // Should not happen in normal flow
+		}
+
+		const userQuery = Array.isArray(lastUserMessage.content)
+			? lastUserMessage.content.map((c) => ("text" in c ? c.text : "")).join("\n")
+			: (lastUserMessage.content as string)
+
+		// Only prepend the system prompt if it's the very first message of a new conversation.
+		if (!this.conversationId && systemPrompt) {
+			console.log("[DIFY DEBUG] Prepending system prompt for new conversation.")
+			return `${systemPrompt}\n\n---\n\n${userQuery}`
+		}
+
+		return userQuery
+	}
+
+	getModel(): { id: string; info: ModelInfo } {
+		return {
+			id: "dify-workflow",
+			info: {
+				maxTokens: 8192,
+				contextWindow: 128000,
+				supportsImages: true,
+				supportsPromptCache: false,
+				inputPrice: 0,
+				outputPrice: 0,
+				description: "Dify workflow - model selection is configured in your Dify application",
+			},
+		}
+	}
+}

src/common.ts

@@ -1,19 +1,24 @@
 import * as vscode from "vscode"
 import {
 	migrateCustomInstructionsToGlobalRules,
+	migrateTaskHistoryToFile,
 	migrateWelcomeViewCompleted,
 	migrateWorkspaceToGlobalStorage,
 } from "./core/storage/state-migrations"
 import { WebviewProvider } from "./core/webview"
 import { Logger } from "./services/logging/Logger"
-import { PostHogClientProvider } from "./services/posthog/PostHogClientProvider"
-import { EmptyRequest } from "./shared/proto/cline/common"
 import { WebviewProviderType } from "./shared/webview/types"
 import "./utils/path" // necessary to have access to String.prototype.toPosix
 
 import { HostProvider } from "@/hosts/host-provider"
 import { FileContextTracker } from "./core/context/context-tracking/FileContextTracker"
-import { telemetryService } from "./services/posthog/PostHogClientProvider"
+import { ExtensionRegistryInfo } from "./registry"
+import { audioRecordingService } from "./services/dictation/AudioRecordingService"
+import { ErrorService } from "./services/error"
+import { featureFlagsService } from "./services/feature-flags"
+import { initializeDistinctId } from "./services/logging/distinctId"
+import { PostHogClientProvider } from "./services/posthog/PostHogClientProvider"
+import { telemetryService } from "./services/telemetry"
 import { ShowMessageType } from "./shared/proto/host/window"
 import { getLatestAnnouncementId } from "./utils/announcements"
 /**
@@ -23,18 +28,15 @@ import { getLatestAnnouncementId } from "./utils/announcements"
  * @returns The webview provider
  */
 export async function initialize(context: vscode.ExtensionContext): Promise<WebviewProvider> {
+	// Set the distinct ID for logging and telemetry
+	await initializeDistinctId(context)
+
 	// Initialize PostHog client provider
-	let distinctId = context.globalState.get<string>("cline.distinctId")
-	if (!distinctId) {
-		try {
-			const response = await HostProvider.env.getMachineId(EmptyRequest.create({}))
-			distinctId = response.value
-		} catch (e) {
-			Logger.warn(`Failed to get machine ID: ${e instanceof Error ? e.message : String(e)}`)
-			// PostHogProvider will fall back to uuid
-		}
-	}
-	PostHogClientProvider.getInstance(distinctId)
+	PostHogClientProvider.getInstance()
+
+	// Setup the external services
+	await ErrorService.initialize()
+	await featureFlagsService.poll()
 
 	// Migrate custom instructions to global Cline rules (one-time cleanup)
 	await migrateCustomInstructionsToGlobalRules(context)
@@ -45,6 +47,9 @@ export async function initialize(context: vscode.ExtensionContext): Promise<Webv
 	// Migrate workspace storage values back to global storage (reverting previous migration)
 	await migrateWorkspaceToGlobalStorage(context)
 
+	// Ensure taskHistory.json exists and migrate legacy state (runs once)
+	await migrateTaskHistoryToFile(context)
+
 	// Clean up orphaned file context warnings (startup cleanup)
 	await FileContextTracker.cleanupOrphanedWarnings(context)
 
@@ -59,7 +64,7 @@ export async function initialize(context: vscode.ExtensionContext): Promise<Webv
 
 async function showVersionUpdateAnnouncement(context: vscode.ExtensionContext) {
 	// Version checking for autoupdate notification
-	const currentVersion = context.extension.packageJSON.version
+	const currentVersion = ExtensionRegistryInfo.version
 	const previousVersion = context.globalState.get<string>("clineVersion")
 	// Perform post-update actions if necessary
 	try {
@@ -68,14 +73,14 @@ async function showVersionUpdateAnnouncement(context: vscode.ExtensionContext) {
 
 			// Use the same condition as announcements: focus when there's a new announcement to show
 			const lastShownAnnouncementId = context.globalState.get<string>("lastShownAnnouncementId")
-			const latestAnnouncementId = getLatestAnnouncementId(context)
+			const latestAnnouncementId = getLatestAnnouncementId()
 
 			if (lastShownAnnouncementId !== latestAnnouncementId) {
 				// Focus Cline when there's a new announcement to show (major/minor updates or fresh installs)
 				const message = previousVersion
 					? `Cline has been updated to v${currentVersion}`
 					: `Welcome to Cline v${currentVersion}`
-				await vscode.commands.executeCommand("claude-dev.SidebarProvider.focus")
+				await HostProvider.workspace.openClineSidebarPanel({})
 				await new Promise((resolve) => setTimeout(resolve, 200))
 				HostProvider.window.showMessage({
 					type: ShowMessageType.INFORMATION,
@@ -95,8 +100,13 @@ async function showVersionUpdateAnnouncement(context: vscode.ExtensionContext) {
  * Performs cleanup when Cline is deactivated that is common to all platforms.
  */
 export async function tearDown(): Promise<void> {
-	PostHogClientProvider.getInstance().dispose()
+	// Clean up audio recording service to ensure no orphaned processes
+	audioRecordingService.cleanup()
 
+	PostHogClientProvider.getInstance().dispose()
+	telemetryService.dispose()
+	ErrorService.get().dispose()
+	featureFlagsService.dispose()
 	// Dispose all webview instances
 	await WebviewProvider.disposeAllInstances()
 }

src/config.ts

@@ -4,7 +4,7 @@ export enum Environment {
 	local = "local",
 }
 
-interface EnvironmentConfig {
+export interface EnvironmentConfig {
 	appBaseUrl: string
 	apiBaseUrl: string
 	mcpBaseUrl: string

src/core/api/index.ts

@@ -9,6 +9,7 @@ import { CerebrasHandler } from "./providers/cerebras"
 import { ClaudeCodeHandler } from "./providers/claude-code"
 import { ClineHandler } from "./providers/cline"
 import { DeepSeekHandler } from "./providers/deepseek"
+import { DifyHandler } from "./providers/dify"
 import { DoubaoHandler } from "./providers/doubao"
 import { FireworksHandler } from "./providers/fireworks"
 import { GeminiHandler } from "./providers/gemini"
@@ -20,6 +21,7 @@ import { LmStudioHandler } from "./providers/lmstudio"
 import { MistralHandler } from "./providers/mistral"
 import { MoonshotHandler } from "./providers/moonshot"
 import { NebiusHandler } from "./providers/nebius"
+import { OcaHandler } from "./providers/oca"
 import { OllamaHandler } from "./providers/ollama"
 import { OpenAiHandler } from "./providers/openai"
 import { OpenAiNativeHandler } from "./providers/openai-native"
@@ -53,9 +55,10 @@ export interface ApiHandlerModel {
 }
 
 export interface ApiProviderInfo {
-	modelId: string
 	providerId: string
-	customPrompt?: string
+	model: ApiHandlerModel
+	customPrompt?: string // "compact"
+	autoCondenseThreshold?: number // 0-1 range
 }
 
 export interface SingleCompletionHandler {
@@ -209,6 +212,7 @@ function createHandlerForProvider(
 			})
 		case "qwen-code":
 			return new QwenCodeHandler({
+				onRetryAttempt: options.onRetryAttempt,
 				qwenCodeOauthPath: options.qwenCodeOauthPath,
 				apiModelId: mode === "plan" ? options.planModeApiModelId : options.actModeApiModelId,
 			})
@@ -311,6 +315,7 @@ function createHandlerForProvider(
 			})
 		case "baseten":
 			return new BasetenHandler({
+				onRetryAttempt: options.onRetryAttempt,
 				basetenApiKey: options.basetenApiKey,
 				basetenModelId: mode === "plan" ? options.planModeBasetenModelId : options.actModeBasetenModelId,
 				basetenModelInfo: mode === "plan" ? options.planModeBasetenModelInfo : options.actModeBasetenModelInfo,
@@ -328,6 +333,8 @@ function createHandlerForProvider(
 				thinkingBudgetTokens:
 					mode === "plan" ? options.planModeThinkingBudgetTokens : options.actModeThinkingBudgetTokens,
 				reasoningEffort: mode === "plan" ? options.planModeReasoningEffort : options.actModeReasoningEffort,
+				deploymentId: mode === "plan" ? options.planModeSapAiCoreDeploymentId : options.actModeSapAiCoreDeploymentId,
+				sapAiCoreUseOrchestrationMode: options.sapAiCoreUseOrchestrationMode,
 			})
 		case "claude-code":
 			return new ClaudeCodeHandler({
@@ -339,14 +346,21 @@ function createHandlerForProvider(
 			})
 		case "huawei-cloud-maas":
 			return new HuaweiCloudMaaSHandler({
+				onRetryAttempt: options.onRetryAttempt,
 				huaweiCloudMaasApiKey: options.huaweiCloudMaasApiKey,
 				huaweiCloudMaasModelId:
 					mode === "plan" ? options.planModeHuaweiCloudMaasModelId : options.actModeHuaweiCloudMaasModelId,
 				huaweiCloudMaasModelInfo:
 					mode === "plan" ? options.planModeHuaweiCloudMaasModelInfo : options.actModeHuaweiCloudMaasModelInfo,
 			})
+		case "dify": // Add Dify.ai handler
+			return new DifyHandler({
+				difyApiKey: options.difyApiKey,
+				difyBaseUrl: options.difyBaseUrl,
+			})
 		case "vercel-ai-gateway":
 			return new VercelAIGatewayHandler({
+				onRetryAttempt: options.onRetryAttempt,
 				vercelAiGatewayApiKey: options.vercelAiGatewayApiKey,
 				vercelAiGatewayModelId:
 					mode === "plan" ? options.planModeVercelAiGatewayModelId : options.actModeVercelAiGatewayModelId,
@@ -355,10 +369,24 @@ function createHandlerForProvider(
 			})
 		case "zai":
 			return new ZAiHandler({
+				onRetryAttempt: options.onRetryAttempt,
 				zaiApiLine: options.zaiApiLine,
 				zaiApiKey: options.zaiApiKey,
 				apiModelId: mode === "plan" ? options.planModeApiModelId : options.actModeApiModelId,
 			})
+		case "oca":
+			return new OcaHandler({
+				ocaBaseUrl: options.ocaBaseUrl,
+				ocaModelId: mode === "plan" ? options.planModeOcaModelId : options.actModeOcaModelId,
+				ocaModelInfo: mode === "plan" ? options.planModeOcaModelInfo : options.actModeOcaModelInfo,
+				thinkingBudgetTokens:
+					mode === "plan" ? options.planModeThinkingBudgetTokens : options.actModeThinkingBudgetTokens,
+				ocaUsePromptCache:
+					mode === "plan"
+						? options.planModeOcaModelInfo?.supportsPromptCache
+						: options.actModeOcaModelInfo?.supportsPromptCache,
+				taskId: options.ulid,
+			})
 		default:
 			return new AnthropicHandler({
 				onRetryAttempt: options.onRetryAttempt,
@@ -384,7 +412,7 @@ export function buildApiHandler(configuration: ApiConfiguration, mode: Mode): Ap
 			const handler = createHandlerForProvider(apiProvider, options, mode)
 
 			const modelInfo = handler.getModel().info
-			if (modelInfo.maxTokens && thinkingBudgetTokens > modelInfo.maxTokens) {
+			if (modelInfo?.maxTokens && modelInfo.maxTokens > 0 && thinkingBudgetTokens > modelInfo.maxTokens) {
 				const clippedValue = modelInfo.maxTokens - 1
 				if (mode === "plan") {
 					options.planModeThinkingBudgetTokens = clippedValue

src/core/api/providers/__tests__/bedrock.test.ts

@@ -1,7 +1,7 @@
 import "should"
 import { ConverseStreamCommand } from "@aws-sdk/client-bedrock-runtime"
-import { ApiHandlerOptions } from "@shared/api"
 import { Readable } from "stream"
+import type { AwsBedrockHandlerOptions } from "../bedrock"
 import { AwsBedrockHandler } from "../bedrock"
 
 describe("AwsBedrockHandler", () => {
@@ -202,8 +202,8 @@ describe("AwsBedrockHandler", () => {
 		})
 	})
 
-	const mockOptions: ApiHandlerOptions = {
-		actModeApiModelId: "anthropic.claude-3-7-sonnet-20250219-v1:0",
+	const mockOptions: AwsBedrockHandlerOptions = {
+		apiModelId: "anthropic.claude-3-7-sonnet-20250219-v1:0",
 		awsRegion: "us-east-1",
 		awsAccessKey: "test-key",
 		awsSecretKey: "test-secret",
@@ -214,9 +214,9 @@ describe("AwsBedrockHandler", () => {
 		awsBedrockUsePromptCache: false,
 		awsUseCrossRegionInference: false,
 		awsBedrockEndpoint: "",
-		actModeAwsBedrockCustomSelected: false,
-		actModeAwsBedrockCustomModelBaseId: undefined,
-		actModeThinkingBudgetTokens: 1600,
+		awsBedrockCustomSelected: false,
+		awsBedrockCustomModelBaseId: undefined,
+		thinkingBudgetTokens: 1600,
 	}
 
 	const mockModelInfo = {

src/core/api/providers/__tests__/claude-code.test.ts

@@ -0,0 +1,248 @@
+import { Anthropic } from "@anthropic-ai/sdk"
+import { afterEach, beforeEach, describe, it } from "mocha"
+import sinon from "sinon"
+import "should"
+import { ClaudeCodeHandler } from "@core/api/providers/claude-code"
+
+describe("ClaudeCodeHandler", () => {
+	let handler: ClaudeCodeHandler
+	let sandbox: sinon.SinonSandbox
+
+	beforeEach(() => {
+		sandbox = sinon.createSandbox()
+		handler = new ClaudeCodeHandler({
+			claudeCodePath: "/mock/path",
+			apiModelId: "claude-3-5-sonnet-20241022",
+		})
+	})
+
+	afterEach(() => {
+		sandbox.restore()
+	})
+
+	describe("token counting", () => {
+		it("should correctly handle token usage from assistant messages", async () => {
+			// The 'input_tokens' field represents the TOTAL number of input tokens used.
+			// See https://docs.anthropic.com/en/api/messages#usage-object
+
+			// Mock the runClaudeCode function
+			const runClaudeCodeModule = await import("@/integrations/claude-code/run")
+			const runClaudeCodeStub = sandbox.stub(runClaudeCodeModule, "runClaudeCode")
+
+			// Create a proper async generator mock for the Claude Code response
+			async function* mockGenerator() {
+				// First yield the system init
+				yield {
+					type: "system",
+					subtype: "init",
+					apiKeySource: "api",
+				}
+
+				// Yield assistant message with usage data
+				// Example: If base input is 70 tokens, cache read is 20, and cache creation is 10,
+				// then input_tokens from Anthropic API will be 100 (70 + 20 + 10)
+				yield {
+					type: "assistant",
+					message: {
+						content: [
+							{
+								type: "text",
+								text: "Test response",
+							},
+						],
+						usage: {
+							input_tokens: 100, // Total including cache (per Anthropic docs)
+							output_tokens: 50,
+							cache_read_input_tokens: 20, // Already included in input_tokens
+							cache_creation_input_tokens: 10, // Already included in input_tokens
+						},
+						stop_reason: "end_turn",
+					},
+				}
+
+				// Yield result with cost
+				yield {
+					type: "result",
+					result: {},
+					total_cost_usd: 0.005,
+				}
+			}
+
+			runClaudeCodeStub.returns(mockGenerator() as any)
+
+			const systemPrompt = "You are a helpful assistant."
+			const messages: Anthropic.Messages.MessageParam[] = [{ role: "user", content: "Hello" }]
+
+			const usageData: any[] = []
+
+			// Collect the results
+			for await (const chunk of handler.createMessage(systemPrompt, messages)) {
+				if (chunk.type === "usage") {
+					usageData.push({
+						inputTokens: chunk.inputTokens,
+						outputTokens: chunk.outputTokens,
+						cacheReadTokens: chunk.cacheReadTokens,
+						cacheWriteTokens: chunk.cacheWriteTokens,
+						totalCost: chunk.totalCost,
+					})
+				}
+			}
+
+			// Verify token counting follows Anthropic API specification
+			usageData.should.have.length(1)
+			usageData[0].should.deepEqual({
+				inputTokens: 100, // Total including cache tokens (per Anthropic API docs)
+				outputTokens: 50,
+				cacheReadTokens: 20, // Tracked separately for reporting
+				cacheWriteTokens: 10, // Tracked separately for reporting
+				totalCost: 0.005,
+			})
+
+			// CRITICAL ASSERTION: Verify that input_tokens is NOT inflated by re-adding cache tokens
+			// The bug would have caused inputTokens to be incorrectly calculated as 130 (100 + 20 + 10)
+			// The fix ensures it remains 100, as per Anthropic's specification
+			usageData[0].inputTokens.should.equal(100) // Correct: matches API response
+			usageData[0].inputTokens.should.not.equal(130) // Would be wrong: double-counting cache tokens
+		})
+
+		it("should handle missing usage fields with nullish coalescing", async () => {
+			// Mock the runClaudeCode function
+			const runClaudeCodeModule = await import("@/integrations/claude-code/run")
+			const runClaudeCodeStub = sandbox.stub(runClaudeCodeModule, "runClaudeCode")
+
+			// Create a proper async generator mock with missing/undefined usage fields
+			async function* mockGenerator() {
+				yield {
+					type: "assistant",
+					message: {
+						content: [
+							{
+								type: "text",
+								text: "Test response",
+							},
+						],
+						usage: {
+							input_tokens: 100,
+							output_tokens: 50,
+							// cache fields are undefined/missing
+						},
+						stop_reason: "end_turn",
+					},
+				}
+
+				yield {
+					type: "result",
+					result: {},
+					total_cost_usd: 0.005,
+				}
+			}
+
+			runClaudeCodeStub.returns(mockGenerator() as any)
+
+			const systemPrompt = "You are a helpful assistant."
+			const messages: Anthropic.Messages.MessageParam[] = [{ role: "user", content: "Hello" }]
+
+			const usageData: any[] = []
+
+			// Collect the results
+			for await (const chunk of handler.createMessage(systemPrompt, messages)) {
+				if (chunk.type === "usage") {
+					usageData.push({
+						inputTokens: chunk.inputTokens,
+						outputTokens: chunk.outputTokens,
+						cacheReadTokens: chunk.cacheReadTokens,
+						cacheWriteTokens: chunk.cacheWriteTokens,
+					})
+				}
+			}
+
+			// Verify that undefined cache tokens default to 0
+			usageData.should.have.length(1)
+			usageData[0].should.deepEqual({
+				inputTokens: 100,
+				outputTokens: 50,
+				cacheReadTokens: 0, // Should default to 0
+				cacheWriteTokens: 0, // Should default to 0
+			})
+		})
+
+		it("should handle completely missing usage object", async () => {
+			// Mock the runClaudeCode function
+			const runClaudeCodeModule = await import("@/integrations/claude-code/run")
+			const runClaudeCodeStub = sandbox.stub(runClaudeCodeModule, "runClaudeCode")
+
+			// Create a proper async generator mock with missing usage object
+			async function* mockGenerator() {
+				yield {
+					type: "assistant",
+					message: {
+						content: [
+							{
+								type: "text",
+								text: "Test response",
+							},
+						],
+						// usage is undefined
+						usage: undefined,
+						stop_reason: "end_turn",
+					},
+				}
+
+				// Need to yield a result chunk to trigger usage data emission
+				yield {
+					type: "result",
+					result: {},
+					total_cost_usd: 0,
+				}
+			}
+
+			runClaudeCodeStub.returns(mockGenerator() as any)
+
+			const systemPrompt = "You are a helpful assistant."
+			const messages: Anthropic.Messages.MessageParam[] = [{ role: "user", content: "Hello" }]
+
+			const usageData: any[] = []
+
+			// Collect the results
+			for await (const chunk of handler.createMessage(systemPrompt, messages)) {
+				if (chunk.type === "usage") {
+					usageData.push({
+						inputTokens: chunk.inputTokens,
+						outputTokens: chunk.outputTokens,
+						cacheReadTokens: chunk.cacheReadTokens,
+						cacheWriteTokens: chunk.cacheWriteTokens,
+					})
+				}
+			}
+
+			// All token counts should default to 0 when usage is undefined
+			usageData.should.have.length(1)
+			usageData[0].should.deepEqual({
+				inputTokens: 0,
+				outputTokens: 0,
+				cacheReadTokens: 0,
+				cacheWriteTokens: 0,
+			})
+		})
+	})
+
+	describe("getModel", () => {
+		it("should return the correct model when specified", () => {
+			const handler = new ClaudeCodeHandler({
+				apiModelId: "claude-3-5-sonnet-20241022",
+			})
+
+			const model = handler.getModel()
+			model.id.should.equal("claude-3-5-sonnet-20241022")
+		})
+
+		it("should return default model when not specified", () => {
+			const handler = new ClaudeCodeHandler({})
+
+			const model = handler.getModel()
+			// The default model should be set
+			model.id.should.be.type("string")
+			model.info.should.be.type("object")
+		})
+	})
+})

src/core/api/providers/__tests__/litellm.test.ts

@@ -0,0 +1,243 @@
+import Anthropic from "@anthropic-ai/sdk"
+import { LiteLlmHandler, type LiteLlmModelInfoResponse } from "@core/api/providers/litellm"
+import { convertToOpenAiMessages } from "@core/api/transform/openai-format"
+import { expect } from "chai"
+import sinon from "sinon"
+
+const fakeClient = {
+	chat: {
+		completions: {
+			create: sinon.stub(),
+		},
+	},
+	baseURL: "fake",
+}
+
+describe("LiteLlmHandler", () => {
+	const originalFetch = global.fetch
+	const mockFetch = sinon.stub()
+
+	const mockModelFetch = (modelInfo: LiteLlmModelInfoResponse["data"][number]) => {
+		mockFetch.resolves({
+			ok: true,
+			json: () =>
+				Promise.resolve({
+					data: [modelInfo],
+				}),
+		})
+	}
+
+	let handler: LiteLlmHandler
+
+	const mockHandlerChat = () => {
+		sinon.stub(handler, "ensureClient" as any).returns(fakeClient)
+	}
+
+	const initializeHandler = (model: string) => {
+		handler = new LiteLlmHandler({
+			liteLlmApiKey: "test-api-key",
+			liteLlmBaseUrl: "http://localhost:4000",
+			liteLlmUsePromptCache: true,
+			liteLlmModelId: model,
+		})
+
+		mockHandlerChat()
+	}
+
+	beforeEach(() => {
+		global.fetch = mockFetch
+
+		// Configure the stub to return a stream that closes immediately with usage data
+		fakeClient.chat.completions.create.resolves(
+			createAsyncIterable([
+				{
+					choices: [{ delta: { content: "test response" } }],
+				},
+				{
+					choices: [{}],
+					usage: {
+						prompt_tokens: 100,
+						completion_tokens: 50,
+						cache_creation_input_tokens: 20,
+						cache_read_input_tokens: 10,
+					},
+				},
+			]),
+		)
+	})
+
+	afterEach(() => {
+		sinon.reset()
+
+		global.fetch = originalFetch
+	})
+
+	const createAsyncIterable = (data: any[] = []) => {
+		return {
+			[Symbol.asyncIterator]: async function* () {
+				yield* data
+			},
+		}
+	}
+
+	describe("prompt cache", () => {
+		const setModelData = (model: string, supportsPromptCaching: boolean) => {
+			mockModelFetch({
+				model_name: model,
+				litellm_params: {
+					model,
+				},
+				model_info: {
+					supports_prompt_caching: supportsPromptCaching,
+					input_cost_per_token: 0.01,
+					output_cost_per_token: 0.02,
+				},
+			})
+		}
+
+		describe("when the model doesn't support prompt caching", () => {
+			const model = "openai/gpt-5"
+
+			beforeEach(() => {
+				initializeHandler(model)
+				setModelData(model, false)
+			})
+
+			it("sends the system prompt and messages with the openai format", async () => {
+				const systemPrompt = "Test System Prompt"
+				const messages: Anthropic.Messages.MessageParam[] = [
+					{
+						role: "user",
+						content: "first message",
+					},
+					{
+						role: "assistant",
+						content: "first response",
+					},
+					{
+						role: "user",
+						content: [
+							{
+								type: "text",
+								text: "test",
+							},
+							{
+								type: "text",
+								text: "second message",
+							},
+						],
+					},
+				]
+
+				for await (const _ of handler.createMessage(systemPrompt, messages)) {
+				}
+
+				sinon.assert.calledOnce(fakeClient.chat.completions.create)
+
+				const callArgs = fakeClient.chat.completions.create.getCall(0).args[0]
+
+				const systemPromptMessage = callArgs.messages.shift()
+				expect(systemPromptMessage).to.deep.equal({
+					role: "system",
+					content: systemPrompt,
+				})
+
+				expect(callArgs.messages).to.deep.equal(convertToOpenAiMessages(messages))
+			})
+		})
+
+		describe("when the model supports prompt caching", () => {
+			const model = "anthropic/claude-sonnet-4-20250514"
+
+			beforeEach(() => {
+				initializeHandler(model)
+
+				setModelData(model, true)
+			})
+
+			it("inserts the cache control in the system prompt and the last two user messages", async () => {
+				const systemPrompt = "Test System Prompt"
+				const messages: Anthropic.Messages.MessageParam[] = [
+					{
+						role: "user",
+						content: "first message",
+					},
+					{
+						role: "assistant",
+						content: "first response",
+					},
+					{
+						role: "user",
+						content: [
+							{
+								type: "text",
+								text: "test",
+							},
+							{
+								type: "text",
+								text: "second message",
+							},
+						],
+					},
+				]
+
+				for await (const _ of handler.createMessage(systemPrompt, messages)) {
+				}
+
+				sinon.assert.calledOnce(fakeClient.chat.completions.create)
+
+				const callArgs = fakeClient.chat.completions.create.getCall(0).args[0]
+
+				expect(callArgs.messages[0]).to.deep.equal({
+					role: "system",
+					content: [
+						{
+							text: systemPrompt,
+							type: "text",
+							cache_control: {
+								type: "ephemeral",
+							},
+						},
+					],
+				})
+
+				const sentMessages = callArgs.messages
+				expect(sentMessages.length).to.equal(4)
+
+				const firstUserMessage = sentMessages[1]
+
+				expect(firstUserMessage).to.deep.equal({
+					role: "user",
+					content: [
+						{
+							type: "text",
+							text: "first message",
+							cache_control: {
+								type: "ephemeral",
+							},
+						},
+					],
+				})
+
+				const lastUserMessage = sentMessages[3]
+				expect(lastUserMessage.content[0]).to.deep.equal({
+					type: "text",
+					text: "test",
+				})
+
+				const lastContentBlock = lastUserMessage.content[lastUserMessage.content.length - 1]
+				expect(lastContentBlock).to.deep.equal({
+					type: "text",
+					text: "second message",
+					cache_control: {
+						type: "ephemeral",
+					},
+				})
+
+				expect(callArgs.model).to.be.a("string")
+				expect(callArgs.stream).to.equal(true)
+				expect(callArgs.stream_options).to.deep.equal({ include_usage: true })
+			})
+		})
+	})
+})

src/core/api/providers/baseten.ts

@@ -2,12 +2,12 @@ import { Anthropic } from "@anthropic-ai/sdk"
 import { BasetenModelId, basetenDefaultModelId, basetenModels, ModelInfo } from "@shared/api"
 import { calculateApiCostOpenAI } from "@utils/cost"
 import OpenAI from "openai"
-import { ApiHandler } from "../"
+import { ApiHandler, CommonApiHandlerOptions } from "../"
 import { withRetry } from "../retry"
 import { convertToOpenAiMessages } from "../transform/openai-format"
 import { ApiStream } from "../transform/stream"
 
-interface BasetenHandlerOptions {
+interface BasetenHandlerOptions extends CommonApiHandlerOptions {
 	basetenApiKey?: string
 	basetenModelId?: string
 	basetenModelInfo?: ModelInfo
@@ -146,20 +146,15 @@ export class BasetenHandler implements ApiHandler {
 		}
 	}
 
-	/**
-	 * Checks if the current model supports vision/images
-	 */
-	supportsImages(): boolean {
-		const model = this.getModel()
-		return model.info.supportsImages === true
-	}
-
 	/**
 	 * Checks if the current model supports tools
 	 */
 	supportsTools(): boolean {
-		const _model = this.getModel()
-		// Baseten models support tools via OpenAI-compatible API
-		return true
+		const model = this.getModel()
+		const modelInfo = model.info as any
+
+		// Use dynamic API data when available, fallback to true since all current Baseten models support tools
+		// (as of 2025-09-16 - could change if Baseten add non-tool models in future, currently no plans to do so)
+		return modelInfo.supportedFeatures ? modelInfo.supportedFeatures.includes("tools") : true
 	}
 }

src/core/api/providers/bedrock.ts

@@ -16,7 +16,7 @@ import { withRetry } from "../retry"
 import { convertToR1Format } from "../transform/r1-format"
 import { ApiStream } from "../transform/stream"
 
-interface AwsBedrockHandlerOptions extends CommonApiHandlerOptions {
+export interface AwsBedrockHandlerOptions extends CommonApiHandlerOptions {
 	apiModelId?: string
 	awsAccessKey?: string
 	awsSecretKey?: string
@@ -30,7 +30,7 @@ interface AwsBedrockHandlerOptions extends CommonApiHandlerOptions {
 	awsProfile?: string
 	awsBedrockEndpoint?: string
 	awsBedrockCustomSelected?: boolean
-	awsBedrockCustomModelBaseId?: BedrockModelId
+	awsBedrockCustomModelBaseId?: string
 	thinkingBudgetTokens?: number
 }
 
@@ -170,7 +170,7 @@ export class AwsBedrockHandler implements ApiHandler {
 			if (baseModel && baseModel in bedrockModels) {
 				return {
 					id: modelId,
-					info: bedrockModels[baseModel],
+					info: bedrockModels[baseModel as BedrockModelId],
 				}
 			}
 			// For custom models without valid base model in bedrock model list, use default model's capabilities

src/core/api/providers/claude-code.ts

@@ -2,7 +2,7 @@ import type { Anthropic } from "@anthropic-ai/sdk"
 import { filterMessagesForClaudeCode } from "@/integrations/claude-code/message-filter"
 import { runClaudeCode } from "@/integrations/claude-code/run"
 import { ClaudeCodeModelId, claudeCodeDefaultModelId, claudeCodeModels } from "@/shared/api"
-import { CommonApiHandlerOptions, type ApiHandler } from ".."
+import { type ApiHandler, CommonApiHandlerOptions } from ".."
 import { withRetry } from "../retry"
 import { type ApiStream, ApiStreamUsageChunk } from "../transform/stream"
 
@@ -118,10 +118,14 @@ export class ClaudeCodeHandler implements ApiHandler {
 					}
 				}
 
-				usage.inputTokens += message.usage.input_tokens
-				usage.outputTokens += message.usage.output_tokens
-				usage.cacheReadTokens = (usage.cacheReadTokens || 0) + (message.usage.cache_read_input_tokens || 0)
-				usage.cacheWriteTokens = (usage.cacheWriteTokens || 0) + (message.usage.cache_creation_input_tokens || 0)
+				// According to Anthropic's API documentation:
+				// https://docs.anthropic.com/en/api/messages#usage-object
+				// The `input_tokens` field already includes both `cache_read_input_tokens` and `cache_creation_input_tokens`.
+				// Therefore, we should not add cache tokens to the input_tokens count again, as this would result in double-counting.
+				usage.inputTokens = message.usage?.input_tokens ?? 0
+				usage.outputTokens = message.usage?.output_tokens ?? 0
+				usage.cacheReadTokens = message.usage?.cache_read_input_tokens ?? 0
+				usage.cacheWriteTokens = message.usage?.cache_creation_input_tokens ?? 0
 
 				continue
 			}

src/core/api/providers/cline.ts

@@ -32,6 +32,7 @@ export class ClineHandler implements ApiHandler {
 	private client: OpenAI | undefined
 	private readonly _baseUrl = clineEnvConfig.apiBaseUrl
 	lastGenerationId?: string
+	private lastRequestId?: string
 
 	constructor(options: ClineHandlerOptions) {
 		this.options = options
@@ -54,6 +55,31 @@ export class ClineHandler implements ApiHandler {
 						"X-Task-ID": this.options.ulid || "",
 						"X-Cline-Version": extensionVersion,
 					},
+					// Capture real HTTP request ID from initial streaming response headers
+					fetch: async (...args: Parameters<typeof fetch>): Promise<Awaited<ReturnType<typeof fetch>>> => {
+						const [input, init] = args
+						const resp = await fetch(input, init)
+						try {
+							let urlStr = ""
+							if (typeof input === "string") {
+								urlStr = input
+							} else if (input instanceof URL) {
+								urlStr = input.toString()
+							} else if (typeof (input as { url?: unknown }).url === "string") {
+								urlStr = (input as { url: string }).url
+							}
+							// Only record for chat completions (the primary streaming request)
+							if (urlStr.includes("/chat/completions")) {
+								const rid = resp.headers.get("x-request-id") || resp.headers.get("request-id")
+								if (rid) {
+									this.lastRequestId = rid
+								}
+							}
+						} catch {
+							// ignore header capture errors
+						}
+						return resp
+					},
 				})
 			} catch (error: any) {
 				throw new Error(`Error creating Cline client: ${error.message}`)
@@ -70,6 +96,7 @@ export class ClineHandler implements ApiHandler {
 			const client = await this.ensureClient()
 
 			this.lastGenerationId = undefined
+			this.lastRequestId = undefined
 
 			let didOutputUsage: boolean = false
 
@@ -92,6 +119,7 @@ export class ClineHandler implements ApiHandler {
 					const metadataStr = error.metadata ? `\nMetadata: ${JSON.stringify(error.metadata, null, 2)}` : ""
 					throw new Error(`Cline API Error ${error.code}: ${error.message}${metadataStr}`)
 				}
+
 				if (!this.lastGenerationId && chunk.id) {
 					this.lastGenerationId = chunk.id
 				}
@@ -134,11 +162,11 @@ export class ClineHandler implements ApiHandler {
 					// @ts-ignore-next-line
 					let totalCost = (chunk.usage.cost || 0) + (chunk.usage.cost_details?.upstream_inference_cost || 0)
 
-					if (this.getModel().id === "cline/sonic") {
+					if (this.getModel().id === "cline/code-supernova") {
 						totalCost = 0
-					} 
+					}
 
-					if  (this.getModel().id === "x-ai/grok-code-fast-1") {
+					if (this.getModel().id === "x-ai/grok-code-fast-1") {
 						totalCost = 0
 					}
 
@@ -172,14 +200,13 @@ export class ClineHandler implements ApiHandler {
 	async getApiStreamUsage(): Promise<ApiStreamUsageChunk | undefined> {
 		if (this.lastGenerationId) {
 			try {
-				// TODO: replace this with firebase auth
-				// TODO: use global API Host
 				const clineAccountAuthToken = await this._authService.getAuthToken()
 				if (!clineAccountAuthToken) {
 					throw new Error(CLINE_ACCOUNT_AUTH_ERROR_MESSAGE)
 				}
 				const response = await axios.get(`${this.clineAccountService.baseUrl}/generation?id=${this.lastGenerationId}`, {
 					headers: {
+						// Align with backend auth expectations
 						Authorization: `Bearer ${clineAccountAuthToken}`,
 					},
 					timeout: 15_000, // this request hangs sometimes
@@ -203,6 +230,11 @@ export class ClineHandler implements ApiHandler {
 		return undefined
 	}
 
+	// Expose the last HTTP request ID captured from response headers (X-Request-ID)
+	getLastRequestId(): string | undefined {
+		return this.lastRequestId
+	}
+
 	getModel(): { id: string; info: ModelInfo } {
 		const modelId = this.options.openRouterModelId
 		const modelInfo = this.options.openRouterModelInfo

src/core/api/providers/dify.ts

@@ -0,0 +1,660 @@
+import { Anthropic } from "@anthropic-ai/sdk"
+import { ModelInfo } from "../../../shared/api"
+import { ApiHandler } from "../index"
+import { ApiStream } from "../transform/stream"
+
+interface DifyHandlerOptions {
+	difyApiKey?: string
+	difyBaseUrl?: string
+}
+
+// Dify API Response Types
+export interface DifyFileResponse {
+	id: string
+	name: string
+	size: number
+	extension: string
+	mime_type: string
+	created_by: string
+	created_at: number
+}
+
+export interface DifyMessage {
+	id: string
+	conversation_id: string
+	inputs: Record<string, any>
+	query: string
+	message_files: Array<{
+		id: string
+		type: string
+		url: string
+		belongs_to: string
+	}>
+	answer: string
+	created_at: number
+	feedback?: {
+		rating: string
+	}
+	retriever_resources?: any[]
+}
+
+interface DifyHistoryResponse {
+	data: DifyMessage[]
+	has_more: boolean
+	limit: number
+}
+
+interface DifyConversation {
+	id: string
+	name: string
+	inputs: Record<string, any>
+	status: string
+	introduction: string
+	created_at: number
+	updated_at: number
+}
+
+interface DifyConversationsResponse {
+	data: DifyConversation[]
+	has_more: boolean
+	limit: number
+}
+
+interface DifyConversationResponse {
+	id: string
+	name: string
+	inputs: Record<string, any>
+	status: string
+	introduction: string
+	created_at: number
+	updated_at: number
+}
+
+export class DifyHandler implements ApiHandler {
+	private options: DifyHandlerOptions
+	private baseUrl: string
+	private apiKey: string
+	private conversationId: string | null = null
+	private currentTaskId: string | null = null
+	private abortController: AbortController | null = null
+
+	constructor(options: DifyHandlerOptions) {
+		this.options = options
+		this.apiKey = options.difyApiKey || ""
+		this.baseUrl = options.difyBaseUrl || ""
+
+		console.log("[DIFY DEBUG] Constructor called with:", {
+			hasApiKey: !!this.apiKey,
+			baseUrl: this.baseUrl,
+		})
+
+		if (!this.apiKey) {
+			throw new Error("Dify API key is required")
+		}
+		if (!this.baseUrl) {
+			throw new Error("Dify base URL is required")
+		}
+	}
+
+	async *createMessage(systemPrompt: string, messages: Anthropic.Messages.MessageParam[]): ApiStream {
+		console.log("[DIFY DEBUG] createMessage called with:", {
+			systemPromptLength: systemPrompt?.length || 0,
+			messagesCount: messages?.length || 0,
+		})
+
+		// Convert messages to Dify format
+		const query = this.convertMessagesToQuery(systemPrompt, messages)
+		const requestBody = {
+			inputs: {},
+			query: query,
+			response_mode: "streaming",
+			conversation_id: this.conversationId || "",
+			user: "cline-user", // A unique user identifier
+			files: [],
+		}
+
+		const fullUrl = `${this.baseUrl}/chat-messages`
+		console.log("[DIFY DEBUG] Making request to:", fullUrl)
+		console.log("[DIFY DEBUG] Request body:", JSON.stringify(requestBody, null, 2))
+
+		let response: Response
+		try {
+			response = await fetch(fullUrl, {
+				method: "POST",
+				headers: {
+					Authorization: `Bearer ${this.apiKey}`,
+					"Content-Type": "application/json",
+				},
+				body: JSON.stringify(requestBody),
+			})
+		} catch (error: any) {
+			console.error("[DIFY DEBUG] Network error during fetch:", error)
+			const cause = error.cause ? ` | Cause: ${error.cause}` : ""
+			throw new Error(`Dify API network error: ${error.message}${cause}`)
+		}
+
+		console.log("[DIFY DEBUG] Response status:", response.status)
+		const headersObj: Record<string, string> = {}
+		response.headers.forEach((value, key) => {
+			headersObj[key] = value
+		})
+		console.log("[DIFY DEBUG] Response headers:", headersObj)
+
+		if (!response.ok) {
+			const errorText = await response.text()
+			console.error("[DIFY DEBUG] Error response:", errorText)
+			throw new Error(`Dify API error: ${response.status} ${response.statusText} - ${errorText}`)
+		}
+
+		if (!response.body) {
+			throw new Error("No response body from Dify API")
+		}
+
+		const reader = response.body.getReader()
+		const decoder = new TextDecoder()
+		let buffer = ""
+		let fullText = ""
+		let hasYieldedContent = false
+		const processedEvents: string[] = []
+		let lastEventTime = Date.now()
+
+		console.log("[DIFY DEBUG] Starting to read streaming response...")
+
+		try {
+			while (true) {
+				const { done, value } = await reader.read()
+				if (done) {
+					console.log("[DIFY DEBUG] Stream ended naturally")
+					console.log(
+						"[DIFY DEBUG] Final state - hasYieldedContent:",
+						hasYieldedContent,
+						"fullText length:",
+						fullText.length,
+						"processedEvents:",
+						processedEvents,
+					)
+					break
+				}
+
+				const chunk = decoder.decode(value, { stream: true })
+				console.log("[DIFY DEBUG] Raw chunk received:", JSON.stringify(chunk))
+
+				buffer += chunk
+				const lines = buffer.split("\n")
+
+				// Keep the last incomplete line in the buffer
+				buffer = lines.pop() || ""
+
+				for (const line of lines) {
+					console.log("[DIFY DEBUG] Processing line:", JSON.stringify(line))
+
+					if (line.startsWith("data: ")) {
+						const data = line.slice(6).trim()
+						console.log("[DIFY DEBUG] Extracted data:", JSON.stringify(data))
+
+						if (data === "[DONE]") {
+							console.log("[DIFY DEBUG] Received [DONE] signal")
+							break
+						}
+
+						if (data === "") {
+							console.log("[DIFY DEBUG] Empty data line, skipping")
+							continue
+						}
+
+						try {
+							const parsed = JSON.parse(data)
+							console.log("[DIFY DEBUG] Parsed JSON:", parsed)
+							processedEvents.push(parsed.event || "unknown")
+							lastEventTime = Date.now()
+
+							// Capture conversation_id as soon as it's available
+							if (parsed.conversation_id && !this.conversationId) {
+								this.conversationId = parsed.conversation_id
+								console.log("[DIFY DEBUG] Captured conversation_id:", this.conversationId)
+							}
+
+							// Handle different Dify event types based on actual Dify API
+							if (parsed.event === "message") {
+								console.log("[DIFY DEBUG] Message event, answer:", parsed.answer)
+								// Dify sends the full text in each "answer" chunk, so we replace.
+								if (typeof parsed.answer === "string") {
+									fullText = parsed.answer
+									console.log("[DIFY DEBUG] Updated fullText length:", fullText.length)
+									yield {
+										type: "text",
+										text: fullText,
+									}
+									hasYieldedContent = true
+								}
+							} else if (parsed.event === "message_replace") {
+								console.log("[DIFY DEBUG] Replace message event:", parsed)
+								if (parsed.answer) {
+									fullText = parsed.answer // Replace instead of append
+									console.log("[DIFY DEBUG] Replaced fullText length:", fullText.length)
+									yield {
+										type: "text",
+										text: fullText,
+									}
+									hasYieldedContent = true
+								}
+							} else if (parsed.event === "message_end") {
+								console.log("[DIFY DEBUG] Message end event", parsed)
+								// Message completed. Yield final text if we have any.
+								if (fullText) {
+									yield {
+										type: "text",
+										text: fullText,
+									}
+									hasYieldedContent = true
+								}
+								// Yield usage data if available
+								if (parsed.usage) {
+									yield {
+										type: "usage",
+										inputTokens: parsed.usage.prompt_tokens || 0,
+										outputTokens: parsed.usage.completion_tokens || parsed.usage.total_tokens || 0,
+										totalCost: parsed.usage.total_price || 0,
+									}
+								}
+								return // End of stream
+							} else if (parsed.event === "error") {
+								console.error("[DIFY DEBUG] Error event:", parsed)
+								throw new Error(`Dify API error: ${parsed.message || "Unknown error"}`)
+							} else if (parsed.event === "workflow_started" || parsed.event === "workflow_finished") {
+								console.log("[DIFY DEBUG] Workflow event:", parsed.event)
+								// These are informational events, continue processing
+							} else if (parsed.event === "node_started" || parsed.event === "node_finished") {
+								console.log("[DIFY DEBUG] Node event:", parsed.event, parsed.data)
+								// These are informational events, continue processing
+							} else if (parsed.event === "ping") {
+								console.log("[DIFY DEBUG] Ping event received, keeping connection alive.")
+								// Ping event, do nothing
+							} else {
+								console.log("[DIFY DEBUG] Unknown event type:", parsed.event, "Full object:", parsed)
+								// Try to extract text from other possible fields
+								if (parsed.text) {
+									fullText += parsed.text
+									yield {
+										type: "text",
+										text: fullText,
+									}
+									hasYieldedContent = true
+								} else if (parsed.content) {
+									fullText += parsed.content
+									yield {
+										type: "text",
+										text: fullText,
+									}
+									hasYieldedContent = true
+								} else if (parsed.answer) {
+									// Fallback: some events might have answer field even if not "message" type
+									fullText += parsed.answer
+									yield {
+										type: "text",
+										text: fullText,
+									}
+									hasYieldedContent = true
+								}
+							}
+						} catch (e) {
+							console.warn("[DIFY DEBUG] Failed to parse JSON:", data, "Error:", e)
+						}
+					} else if (line.trim() !== "") {
+						console.log(
+							"[DIFY DEBUG] Non-data line (not starting with 'data:'), trying to parse as direct JSON:",
+							JSON.stringify(line),
+						)
+						// Try to parse as direct JSON (fallback for non-SSE responses, though Dify uses SSE)
+						try {
+							const parsed = JSON.parse(line.trim())
+							console.log("[DIFY DEBUG] Parsed direct JSON:", parsed)
+							processedEvents.push(parsed.event || "direct-json")
+
+							// Handle the same event types as above
+							if (parsed.event === "message" && parsed.answer) {
+								fullText += parsed.answer
+								yield {
+									type: "text",
+									text: fullText,
+								}
+								hasYieldedContent = true
+							} else if (parsed.event === "message_end") {
+								if (fullText) {
+									yield {
+										type: "text",
+										text: fullText,
+									}
+									hasYieldedContent = true
+								}
+								return
+							} else if (parsed.event === "error") {
+								console.error("[DIFY DEBUG] Direct JSON Error event:", parsed)
+								throw new Error(`Dify API error: ${parsed.message || "Unknown error"}`)
+							} else if (parsed.answer || parsed.text || parsed.content) {
+								// Fallback for any content in direct JSON
+								const content = parsed.answer || parsed.text || parsed.content
+								fullText += content
+								yield {
+									type: "text",
+									text: fullText,
+								}
+								hasYieldedContent = true
+							}
+						} catch (e) {
+							// Not JSON, continue
+							console.log("[DIFY DEBUG] Line is not direct JSON, continuing")
+						}
+					}
+				}
+			}
+
+			// Final check - if we haven't yielded any content, provide diagnostic information
+			if (!hasYieldedContent) {
+				const diagnosticInfo = {
+					processedEvents,
+					finalFullTextLength: fullText.length,
+					finalFullText: fullText,
+					streamDuration: Date.now() - lastEventTime,
+					conversationId: this.conversationId,
+				}
+				console.error("[DIFY DEBUG] No content was yielded! Diagnostic info:", diagnosticInfo)
+
+				// If we have any accumulated text at all, yield it as a fallback
+				if (fullText.trim()) {
+					console.log("[DIFY DEBUG] Yielding accumulated text as fallback:", fullText)
+					yield {
+						type: "text",
+						text: fullText,
+					}
+				} else {
+					// Provide a more informative error
+					throw new Error(
+						`Dify API did not provide any assistant messages. ` +
+							`Events processed: [${processedEvents.join(", ")}]. ` +
+							`Check your Dify application configuration and ensure it's properly set up to return responses. ` +
+							`API URL: ${fullUrl}. Conversation ID: ${this.conversationId || "none"}.`,
+					)
+				}
+			}
+		} finally {
+			reader.releaseLock()
+			console.log("[DIFY DEBUG] Stream reader released")
+		}
+	}
+
+	private convertMessagesToQuery(systemPrompt: string, messages: Anthropic.Messages.MessageParam[]): string {
+		// Dify's context is managed by `conversation_id`. The `query` should be the last user message.
+		// The system prompt is typically configured in the Dify App itself.
+		const lastUserMessage = messages.filter((m) => m.role === "user").pop()
+
+		if (!lastUserMessage) {
+			return "" // Should not happen in normal flow
+		}
+
+		const userQuery = Array.isArray(lastUserMessage.content)
+			? lastUserMessage.content.map((c) => ("text" in c ? c.text : "")).join("\n")
+			: (lastUserMessage.content as string)
+
+		// Only prepend the system prompt if it's the very first message of a new conversation.
+		if (!this.conversationId && systemPrompt) {
+			console.log("[DIFY DEBUG] Prepending system prompt for new conversation.")
+			return `${systemPrompt}\n\n---\n\n${userQuery}`
+		}
+
+		return userQuery
+	}
+
+	getModel(): { id: string; info: ModelInfo } {
+		return {
+			id: "dify-workflow",
+			info: {
+				maxTokens: 8192,
+				contextWindow: 128000,
+				supportsImages: true,
+				supportsPromptCache: false,
+				inputPrice: 0,
+				outputPrice: 0,
+				description: "Dify workflow - model selection is configured in your Dify application",
+			},
+		}
+	}
+
+	// Additional Dify API Methods
+
+	/**
+	 * Upload a file for use in conversations
+	 * @param file File buffer to upload
+	 * @param filename Name of the file
+	 * @param user User identifier (defaults to "cline-user")
+	 * @returns Promise with file upload response
+	 */
+	async uploadFile(file: Buffer, filename: string, user: string = "cline-user"): Promise<DifyFileResponse> {
+		const formData = new FormData()
+		formData.append("file", new Blob([new Uint8Array(file)]), filename)
+		formData.append("user", user)
+
+		const response = await fetch(`${this.baseUrl}/files/upload`, {
+			method: "POST",
+			headers: {
+				Authorization: `Bearer ${this.apiKey}`,
+			},
+			body: formData,
+		})
+
+		if (!response.ok) {
+			const errorText = await response.text()
+			throw new Error(`Dify file upload error: ${response.status} ${response.statusText} - ${errorText}`)
+		}
+
+		return response.json()
+	}
+
+	/**
+	 * Stop generation for a specific task
+	 * @param taskId Task ID from streaming response
+	 * @param user User identifier (defaults to "cline-user")
+	 * @returns Promise that resolves when generation is stopped
+	 */
+	async stopGeneration(taskId: string, user: string = "cline-user"): Promise<void> {
+		const response = await fetch(`${this.baseUrl}/chat-messages/${taskId}/stop`, {
+			method: "POST",
+			headers: {
+				Authorization: `Bearer ${this.apiKey}`,
+				"Content-Type": "application/json",
+			},
+			body: JSON.stringify({ user }),
+		})
+
+		if (!response.ok) {
+			const errorText = await response.text()
+			throw new Error(`Dify stop generation error: ${response.status} ${response.statusText} - ${errorText}`)
+		}
+	}
+
+	/**
+	 * Get conversation history messages with pagination
+	 * @param conversationId Conversation ID
+	 * @param user User identifier (defaults to "cline-user")
+	 * @param firstId First message ID for pagination (optional)
+	 * @param limit Number of messages to return (default: 20)
+	 * @returns Promise with conversation history
+	 */
+	async getConversationHistory(
+		conversationId: string,
+		user: string = "cline-user",
+		firstId?: string,
+		limit: number = 20,
+	): Promise<DifyHistoryResponse> {
+		const params = new URLSearchParams({ user, limit: limit.toString() })
+		if (firstId) {
+			params.append("first_id", firstId)
+		}
+
+		const response = await fetch(`${this.baseUrl}/conversations/${conversationId}/messages?${params}`, {
+			headers: {
+				Authorization: `Bearer ${this.apiKey}`,
+			},
+		})
+
+		if (!response.ok) {
+			const errorText = await response.text()
+			throw new Error(`Dify get conversation history error: ${response.status} ${response.statusText} - ${errorText}`)
+		}
+
+		return response.json()
+	}
+
+	/**
+	 * Get list of conversations for a user
+	 * @param user User identifier (defaults to "cline-user")
+	 * @param lastId Last conversation ID for pagination (optional)
+	 * @param limit Number of conversations to return (default: 20)
+	 * @param sortBy Sort field (default: "-updated_at")
+	 * @returns Promise with conversations list
+	 */
+	async getConversations(
+		user: string = "cline-user",
+		lastId?: string,
+		limit: number = 20,
+		sortBy: string = "-updated_at",
+	): Promise<DifyConversationsResponse> {
+		const params = new URLSearchParams({
+			user,
+			limit: limit.toString(),
+			sort_by: sortBy,
+		})
+		if (lastId) {
+			params.append("last_id", lastId)
+		}
+
+		const response = await fetch(`${this.baseUrl}/conversations?${params}`, {
+			headers: {
+				Authorization: `Bearer ${this.apiKey}`,
+			},
+		})
+
+		if (!response.ok) {
+			const errorText = await response.text()
+			throw new Error(`Dify get conversations error: ${response.status} ${response.statusText} - ${errorText}`)
+		}
+
+		return response.json()
+	}
+
+	/**
+	 * Delete a conversation
+	 * @param conversationId Conversation ID to delete
+	 * @param user User identifier (defaults to "cline-user")
+	 * @returns Promise that resolves when conversation is deleted
+	 */
+	async deleteConversation(conversationId: string, user: string = "cline-user"): Promise<void> {
+		const response = await fetch(`${this.baseUrl}/conversations/${conversationId}`, {
+			method: "DELETE",
+			headers: {
+				Authorization: `Bearer ${this.apiKey}`,
+				"Content-Type": "application/json",
+			},
+			body: JSON.stringify({ user }),
+		})
+
+		if (!response.ok) {
+			const errorText = await response.text()
+			throw new Error(`Dify delete conversation error: ${response.status} ${response.statusText} - ${errorText}`)
+		}
+	}
+
+	/**
+	 * Rename a conversation
+	 * @param conversationId Conversation ID to rename
+	 * @param user User identifier (defaults to "cline-user")
+	 * @param name New conversation name (optional if auto_generate is true)
+	 * @param autoGenerate Whether to auto-generate the name (default: false)
+	 * @returns Promise with updated conversation details
+	 */
+	async renameConversation(
+		conversationId: string,
+		user: string = "cline-user",
+		name?: string,
+		autoGenerate: boolean = false,
+	): Promise<DifyConversationResponse> {
+		const body: any = { user, auto_generate: autoGenerate }
+		if (name) {
+			body.name = name
+		}
+
+		const response = await fetch(`${this.baseUrl}/conversations/${conversationId}/name`, {
+			method: "POST",
+			headers: {
+				Authorization: `Bearer ${this.apiKey}`,
+				"Content-Type": "application/json",
+			},
+			body: JSON.stringify(body),
+		})
+
+		if (!response.ok) {
+			const errorText = await response.text()
+			throw new Error(`Dify rename conversation error: ${response.status} ${response.statusText} - ${errorText}`)
+		}
+
+		return response.json()
+	}
+
+	/**
+	 * Submit feedback for a message
+	 * @param messageId Message ID to provide feedback for
+	 * @param rating Rating: "like" or "dislike"
+	 * @param content Optional feedback content
+	 * @param user User identifier (defaults to "cline-user")
+	 * @returns Promise that resolves when feedback is submitted
+	 */
+	async submitMessageFeedback(
+		messageId: string,
+		rating: "like" | "dislike",
+		content?: string,
+		user: string = "cline-user",
+	): Promise<void> {
+		const body: any = { rating, user }
+		if (content) {
+			body.content = content
+		}
+
+		const response = await fetch(`${this.baseUrl}/messages/${messageId}/feedbacks`, {
+			method: "POST",
+			headers: {
+				Authorization: `Bearer ${this.apiKey}`,
+				"Content-Type": "application/json",
+			},
+			body: JSON.stringify(body),
+		})
+
+		if (!response.ok) {
+			const errorText = await response.text()
+			throw new Error(`Dify submit feedback error: ${response.status} ${response.statusText} - ${errorText}`)
+		}
+	}
+
+	/**
+	 * Get current conversation ID
+	 * @returns Current conversation ID or null
+	 */
+	getCurrentConversationId(): string | null {
+		return this.conversationId
+	}
+
+	/**
+	 * Set conversation ID for continuing existing conversations
+	 * @param conversationId Conversation ID to set
+	 */
+	setConversationId(conversationId: string): void {
+		this.conversationId = conversationId
+	}
+
+	/**
+	 * Reset conversation ID to start a new conversation
+	 */
+	resetConversation(): void {
+		this.conversationId = null
+		this.currentTaskId = null
+	}
+}

src/core/api/providers/gemini.ts

@@ -1,16 +1,18 @@
 import type { Anthropic } from "@anthropic-ai/sdk"
 // Restore GenerateContentConfig import and add GenerateContentResponseUsageMetadata
-import { type GenerateContentConfig, type GenerateContentResponseUsageMetadata, GoogleGenAI, Part } from "@google/genai"
-import { telemetryService } from "@services/posthog/PostHogClientProvider"
+import { ApiError, type GenerateContentConfig, type GenerateContentResponseUsageMetadata, GoogleGenAI, Part } from "@google/genai"
 import { GeminiModelId, geminiDefaultModelId, geminiModels, ModelInfo } from "@shared/api"
+import { telemetryService } from "@/services/telemetry"
 import { ApiHandler, CommonApiHandlerOptions } from "../"
-import { withRetry } from "../retry"
+import { RetriableError, withRetry } from "../retry"
 import { convertAnthropicMessageToGemini } from "../transform/gemini-format"
 import { ApiStream } from "../transform/stream"
 
 // Define a default TTL for the cache (e.g., 15 minutes in seconds)
 const _DEFAULT_CACHE_TTL_SECONDS = 900
 
+const rateLimitPatterns = [/got status: 429/i, /429 Too Many Requests/i, /rate limit exceeded/i, /too many requests/i]
+
 interface GeminiHandlerOptions extends CommonApiHandlerOptions {
 	isVertex?: boolean
 	vertexProjectId?: string
@@ -222,24 +224,40 @@ export class GeminiHandler implements ApiHandler {
 			if (error instanceof Error) {
 				apiError = error.message
 
-				// Gemini doesn't include status codes in their errors
-				// https://github.com/googleapis/js-genai/blob/61f7f27b866c74333ca6331883882489bcb708b9/src/_api_client.ts#L569
-				const rateLimitPatterns = [
-					/got status: 429/i,
-					/429 Too Many Requests/i,
-					/rate limit exceeded/i,
-					/too many requests/i,
-				]
+				if (error instanceof ApiError) {
+					if (error.status === 429) {
+						// The API includes more details in the message
+						// https://github.com/googleapis/js-genai/blob/v1.11.0/src/_api_client.ts#L758
+						const response = this.attemptParse(error.message)
 
-				const isRateLimit =
-					error.name === "ClientError" && rateLimitPatterns.some((pattern) => pattern.test(error.message))
+						if (response && response.error) {
+							const responseBody = this.attemptParse(response.error.message)
 
-				if (isRateLimit) {
-					const rateLimitError = Object.assign(new Error(error.message), {
-						...error,
-						status: 429,
-					})
-					throw rateLimitError
+							if (responseBody.error) {
+								const detail = responseBody.error.details?.find(
+									(d: any) => d["@type"] === "type.googleapis.com/google.rpc.RetryInfo",
+								)
+
+								const detailedError = new RetriableError(
+									apiError,
+									this.parseRetryDelay(detail?.retryDelay) || undefined,
+									{
+										cause: error,
+									},
+								)
+								throw detailedError
+							}
+						}
+
+						throw new RetriableError(apiError, undefined, { cause: error })
+					}
+
+					// Fallback in case Gemini throws a rate limit error without a 429 status code
+					// https://github.com/cline/cline/pull/5205#discussion_r2311761559
+					const isRateLimit = rateLimitPatterns.some((pattern) => pattern.test(error.message))
+					if (isRateLimit) {
+						throw new RetriableError(apiError, undefined, { cause: error })
+					}
 				}
 			} else {
 				apiError = String(error)
@@ -420,4 +438,35 @@ export class GeminiHandler implements ApiHandler {
 
 		return Math.ceil(totalChars / 4)
 	}
+
+	private parseRetryDelay(retryAfter?: string): number {
+		if (!retryAfter) {
+			return 0
+		}
+
+		const unit = retryAfter.at(-1)
+		const value = parseInt(retryAfter, 10)
+
+		if (Number.isNaN(value)) {
+			return 0
+		}
+
+		if (unit === "s") {
+			return value
+		} else if (unit === "m") {
+			return value * 60 // Convert minutes to seconds
+		} else if (unit === "h") {
+			return value * 60 * 60 // Convert hours to seconds
+		}
+
+		return value
+	}
+
+	private attemptParse(str: string) {
+		try {
+			return JSON.parse(str)
+		} catch (_) {
+			return null
+		}
+	}
 }

src/core/api/providers/huawei-cloud-maas.ts

@@ -1,12 +1,12 @@
 import { Anthropic } from "@anthropic-ai/sdk"
 import { HuaweiCloudMaasModelId, huaweiCloudMaasDefaultModelId, huaweiCloudMaasModels, ModelInfo } from "@shared/api"
 import OpenAI from "openai"
-import { ApiHandler } from ".."
+import { ApiHandler, CommonApiHandlerOptions } from ".."
 import { withRetry } from "../retry"
 import { convertToOpenAiMessages } from "../transform/openai-format"
 import { ApiStream } from "../transform/stream"
 
-interface HuaweiCloudMaaSHandlerOptions {
+interface HuaweiCloudMaaSHandlerOptions extends CommonApiHandlerOptions {
 	huaweiCloudMaasApiKey?: string
 	huaweiCloudMaasModelId?: string
 	huaweiCloudMaasModelInfo?: ModelInfo

src/core/api/providers/litellm.ts

@@ -1,6 +1,7 @@
 import { Anthropic } from "@anthropic-ai/sdk"
 import { LiteLLMModelInfo, liteLlmDefaultModelId, liteLlmModelInfoSaneDefaults } from "@shared/api"
 import OpenAI from "openai"
+import { isAnthropicModelId } from "@/utils/model-utils"
 import { ApiHandler, CommonApiHandlerOptions } from ".."
 import { withRetry } from "../retry"
 import { convertToOpenAiMessages } from "../transform/openai-format"
@@ -16,7 +17,7 @@ interface LiteLlmHandlerOptions extends CommonApiHandlerOptions {
 	ulid?: string
 }
 
-interface LiteLlmModelInfoResponse {
+export interface LiteLlmModelInfoResponse {
 	data: Array<{
 		model_name: string
 		litellm_params: {
@@ -28,6 +29,7 @@ interface LiteLlmModelInfoResponse {
 			output_cost_per_token: number
 			cache_creation_input_token_cost?: number
 			cache_read_input_token_cost?: number
+			supports_prompt_caching?: boolean
 			[key: string]: any
 		}
 	}>
@@ -61,7 +63,17 @@ export class LiteLlmHandler implements ApiHandler {
 		return this.client
 	}
 
-	private async fetchModelInfo(): Promise<LiteLlmModelInfoResponse | undefined> {
+	private async modelInfo(publicModelName: string): Promise<LiteLlmModelInfoResponse["data"][number] | undefined> {
+		const modelInfo = await this.fetchModelsInfo()
+
+		if (!modelInfo?.data) {
+			return undefined
+		}
+
+		return modelInfo.data.find((model) => model.model_name === publicModelName)
+	}
+
+	private async fetchModelsInfo(): Promise<LiteLlmModelInfoResponse | undefined> {
 		// Check if cache is still valid
 		const now = Date.now()
 		if (this.modelInfoCache && now - this.modelInfoCacheTimestamp < this.modelInfoCacheTTL) {
@@ -121,19 +133,14 @@ export class LiteLlmHandler implements ApiHandler {
 		cacheReadCostPerToken?: number
 	}> {
 		try {
-			const modelInfo = await this.fetchModelInfo()
+			const matchingModel = await this.modelInfo(publicModelName)
 
-			if (modelInfo?.data) {
-				// Find the model by public name
-				const matchingModel = modelInfo.data.find((model) => model.model_name === publicModelName)
-
-				if (matchingModel?.model_info) {
-					return {
-						inputCostPerToken: matchingModel.model_info.input_cost_per_token || 0,
-						outputCostPerToken: matchingModel.model_info.output_cost_per_token || 0,
-						cacheCreationCostPerToken: matchingModel.model_info.cache_creation_input_token_cost,
-						cacheReadCostPerToken: matchingModel.model_info.cache_read_input_token_cost,
-					}
+			if (matchingModel) {
+				return {
+					inputCostPerToken: matchingModel.model_info.input_cost_per_token || 0,
+					outputCostPerToken: matchingModel.model_info.output_cost_per_token || 0,
+					cacheCreationCostPerToken: matchingModel.model_info.cache_creation_input_token_cost,
+					cacheReadCostPerToken: matchingModel.model_info.cache_read_input_token_cost,
 				}
 			}
 		} catch (error) {
@@ -177,7 +184,7 @@ export class LiteLlmHandler implements ApiHandler {
 	async *createMessage(systemPrompt: string, messages: Anthropic.Messages.MessageParam[]): ApiStream {
 		const client = this.ensureClient()
 		const formattedMessages = convertToOpenAiMessages(messages)
-		const systemMessage: OpenAI.Chat.ChatCompletionSystemMessageParam = {
+		const systemMessage: OpenAI.Chat.ChatCompletionSystemMessageParam | Anthropic.Messages.TextBlockParam = {
 			role: "system",
 			content: systemPrompt,
 		}
@@ -191,17 +198,26 @@ export class LiteLlmHandler implements ApiHandler {
 
 		let temperature: number | undefined = this.options.liteLlmModelInfo?.temperature ?? 0
 
-		if (isOminiModel && reasoningOn) {
-			temperature = undefined // Thinking mode doesn't support temperature
+		if ((isOminiModel || isAnthropicModelId(modelId)) && reasoningOn) {
+			temperature = undefined // OAI omni and Anthropic extended thinking mode doesn't support temperature
 		}
 
-		// Define cache control object if prompt caching is enabled
-		const cacheControl = this.options.liteLlmUsePromptCache ? { cache_control: { type: "ephemeral" } } : undefined
+		const modelInfo = await this.modelInfo(modelId)
+		const cacheControl =
+			this.options.liteLlmUsePromptCache && Boolean(modelInfo?.model_info.supports_prompt_caching)
+				? { cache_control: { type: "ephemeral" } }
+				: undefined
 
-		// Add cache_control to system message if enabled
-		const enhancedSystemMessage = {
-			...systemMessage,
-			...(cacheControl && cacheControl),
+		if (cacheControl) {
+			// Add cache_control to system message if enabled
+			// https://docs.litellm.ai/docs/providers/anthropic#caching---large-context-caching
+			systemMessage.content = [
+				{
+					text: systemPrompt,
+					type: "text",
+					...cacheControl,
+				},
+			] as Anthropic.Messages.TextBlockParam[]
 		}
 
 		// Find the last two user messages to apply caching
@@ -213,19 +229,49 @@ export class LiteLlmHandler implements ApiHandler {
 		const secondLastUserMsgIndex = userMsgIndices[userMsgIndices.length - 2] ?? -1
 
 		// Apply cache_control to the last two user messages if enabled
-		const enhancedMessages = formattedMessages.map((message, index) => {
-			if ((index === lastUserMsgIndex || index === secondLastUserMsgIndex) && cacheControl) {
-				return {
-					...message,
-					...cacheControl,
+		// https://docs.litellm.ai/docs/providers/anthropic#caching---large-context-caching
+		const enhancedMessages: OpenAI.Chat.ChatCompletionMessageParam[] = formattedMessages.map(
+			(message, index): OpenAI.Chat.ChatCompletionMessageParam => {
+				if ((index === lastUserMsgIndex || index === secondLastUserMsgIndex) && cacheControl) {
+					// Handle both string and array content types
+					if (typeof message.content === "string") {
+						return {
+							...message,
+							content: [
+								{
+									type: "text",
+									text: message.content,
+									...cacheControl,
+								},
+							] as any,
+						}
+					} else if (Array.isArray(message.content)) {
+						// Apply cache control to the last content item in the array
+						return {
+							...message,
+							content: message.content.map((item, contentIndex) =>
+								contentIndex === (message.content?.length || 0) - 1
+									? {
+											...item,
+											...cacheControl,
+										}
+									: item,
+							) as any,
+						}
+					}
+
+					return {
+						...message,
+						...cacheControl,
+					}
 				}
-			}
-			return message
-		})
+				return message
+			},
+		)
 
 		const stream = await client.chat.completions.create({
 			model: this.options.liteLlmModelId || liteLlmDefaultModelId,
-			messages: [enhancedSystemMessage, ...enhancedMessages],
+			messages: [systemMessage, ...enhancedMessages],
 			temperature,
 			stream: true,
 			stream_options: { include_usage: true },
@@ -244,16 +290,16 @@ export class LiteLlmHandler implements ApiHandler {
 				}
 			}
 
-			// Handle reasoning events (thinking)
-			// Thinking is not in the standard types but may be in the response
+			// Handle reasoning events
+			// This is not in the standard types but may be in the response
 			interface ThinkingDelta {
-				thinking?: string
+				reasoning_content?: string
 			}
 
-			if ((delta as ThinkingDelta)?.thinking) {
+			if ((delta as ThinkingDelta)?.reasoning_content) {
 				yield {
 					type: "reasoning",
-					reasoning: (delta as ThinkingDelta).thinking || "",
+					reasoning: (delta as ThinkingDelta).reasoning_content || "",
 				}
 			}
 

src/core/api/providers/moonshot.ts

@@ -49,6 +49,7 @@ export class MoonshotHandler implements ApiHandler {
 			model: model.id,
 			messages: openAiMessages,
 			temperature: 0,
+			max_tokens: model.info.maxTokens,
 			stream: true,
 			stream_options: { include_usage: true },
 		})

src/core/api/providers/oca.ts

@@ -0,0 +1,261 @@
+import { Anthropic } from "@anthropic-ai/sdk"
+import { LiteLLMModelInfo, liteLlmDefaultModelId, liteLlmModelInfoSaneDefaults } from "@shared/api"
+import OpenAI, { APIError, OpenAIError } from "openai"
+import type { FinalRequestOptions, Headers as OpenAIHeaders } from "openai/core"
+import { OcaAuthService } from "@/services/auth/oca/OcaAuthService"
+import { DEFAULT_OCA_BASE_URL, OCI_HEADER_OPC_REQUEST_ID } from "@/services/auth/oca/utils/constants"
+import { createOcaHeaders } from "@/services/auth/oca/utils/utils"
+import { Logger } from "@/services/logging/Logger"
+import { ApiHandler, type CommonApiHandlerOptions } from ".."
+import { withRetry } from "../retry"
+import { convertToOpenAiMessages } from "../transform/openai-format"
+import { ApiStream } from "../transform/stream"
+
+export interface OcaHandlerOptions extends CommonApiHandlerOptions {
+	ocaBaseUrl?: string
+	ocaModelId?: string
+	ocaModelInfo?: LiteLLMModelInfo
+	thinkingBudgetTokens?: number
+	ocaUsePromptCache?: boolean
+	taskId?: string
+}
+
+export class OcaHandler implements ApiHandler {
+	protected options: OcaHandlerOptions
+	protected client: OpenAI | undefined
+
+	constructor(options: OcaHandlerOptions) {
+		this.options = options
+	}
+
+	protected initializeClient(options: OcaHandlerOptions) {
+		return new (class OCIOpenAI extends OpenAI {
+			protected override async prepareOptions(opts: FinalRequestOptions<unknown>): Promise<void> {
+				const token = await OcaAuthService.getInstance().getAuthToken()
+				if (!token) {
+					throw new OpenAIError("Unable to handle auth, Oracle Code Assist (OCA) access token is not available")
+				}
+				opts.headers ??= {}
+				// OCA Headers
+				const ociHeaders = await createOcaHeaders(token, options.taskId!)
+				opts.headers = { ...opts.headers, ...ociHeaders }
+				Logger.log(`Making request with customer opc-request-id: ${opts.headers?.["opc-request-id"]}`)
+				return super.prepareOptions(opts)
+			}
+
+			protected override makeStatusError(
+				status: number | undefined,
+				error: Object | undefined,
+				message: string | undefined,
+				headers: OpenAIHeaders | undefined,
+			): APIError {
+				interface OciError {
+					code?: string
+					message?: string
+				}
+				let ociErrorMessage = message
+				if (typeof error === "object" && error !== null) {
+					try {
+						ociErrorMessage = JSON.stringify(error)
+						const ociErr = error as OciError
+						if (ociErr.code !== undefined && ociErr.message !== undefined) {
+							ociErrorMessage = `${ociErr.code}: ${ociErr.message}`
+						}
+					} catch {}
+				}
+				const opcRequestId = headers?.[OCI_HEADER_OPC_REQUEST_ID]
+				if (opcRequestId) {
+					ociErrorMessage += `\n(${OCI_HEADER_OPC_REQUEST_ID}: ${opcRequestId})`
+				}
+				return super.makeStatusError(status, error, ociErrorMessage, headers)
+			}
+		})({
+			baseURL: options.ocaBaseUrl || DEFAULT_OCA_BASE_URL,
+			apiKey: "noop",
+		})
+	}
+
+	protected ensureClient(): OpenAI {
+		if (!this.client) {
+			if (!this.options.ocaModelId) {
+				throw new Error("Oracle Code Assist (OCA) model is not selected")
+			}
+			try {
+				this.client = this.initializeClient(this.options)
+			} catch (error) {
+				throw new Error(`Error creating Oracle Code Assist (OCA) client: ${error.message}`)
+			}
+		}
+		return this.client
+	}
+
+	async calculateCost(prompt_tokens: number, completion_tokens: number): Promise<number | undefined> {
+		// Reference: https://github.com/BerriAI/litellm/blob/122ee634f434014267af104814022af1d9a0882f/litellm/proxy/spend_tracking/spend_management_endpoints.py#L1473
+		const client = this.ensureClient()
+		const modelId = this.options.ocaModelId || liteLlmDefaultModelId
+		const token = await OcaAuthService.getInstance().getAuthToken()
+		if (!token) {
+			throw new OpenAIError("Unable to handle auth, Oracle Code Assist (OCA) access token is not available")
+		}
+		const ociHeaders = await createOcaHeaders(token, this.options.taskId!)
+		Logger.log(`Making calculate cost request with customer opc-request-id: ${ociHeaders["opc-request-id"]}`)
+		try {
+			const response = await fetch(`${client.baseURL}/spend/calculate`, {
+				method: "POST",
+				headers: ociHeaders,
+				body: JSON.stringify({
+					completion_response: {
+						model: modelId,
+						usage: {
+							prompt_tokens,
+							completion_tokens,
+						},
+					},
+				}),
+			})
+
+			if (response.ok) {
+				const data: { cost: number } = await response.json()
+				return data.cost
+			} else {
+				console.error("Error calculating spend:", response.statusText)
+				return undefined
+			}
+		} catch (error) {
+			console.error("Error calculating spend:", error)
+			return undefined
+		}
+	}
+
+	@withRetry()
+	async *createMessage(systemPrompt: string, messages: Anthropic.Messages.MessageParam[]): ApiStream {
+		const client = this.ensureClient()
+		const formattedMessages = convertToOpenAiMessages(messages)
+		const systemMessage: OpenAI.Chat.ChatCompletionSystemMessageParam = {
+			role: "system",
+			content: systemPrompt,
+		}
+		const modelId = this.options.ocaModelId || liteLlmDefaultModelId
+		const isOminiModel = modelId.includes("o1-mini") || modelId.includes("o3-mini") || modelId.includes("o4-mini")
+
+		// Configuration for extended thinking
+		const budgetTokens = this.options.thinkingBudgetTokens || 0
+		const reasoningOn = budgetTokens !== 0 ? true : false
+		const thinkingConfig = reasoningOn ? { type: "enabled", budget_tokens: budgetTokens } : undefined
+
+		let temperature: number | undefined = this.options.ocaModelInfo?.temperature ?? 0
+		const maxTokens: number | undefined = this.options.ocaModelInfo?.maxTokens
+
+		if (isOminiModel && reasoningOn) {
+			temperature = undefined // Thinking mode doesn't support temperature
+		}
+
+		// Define cache control object if prompt caching is enabled
+		const cacheControl = this.options.ocaUsePromptCache ? { cache_control: { type: "ephemeral" } } : undefined
+
+		// Add cache_control to system message if enabled
+		const enhancedSystemMessage = {
+			...systemMessage,
+			...(cacheControl && cacheControl),
+		}
+
+		// Find the last two user messages to apply caching
+		const userMsgIndices = formattedMessages.reduce((acc, msg, index) => {
+			if (msg.role === "user") {
+				acc.push(index)
+			}
+			return acc
+		}, [] as number[])
+		const lastUserMsgIndex = userMsgIndices[userMsgIndices.length - 1] ?? -1
+		const secondLastUserMsgIndex = userMsgIndices[userMsgIndices.length - 2] ?? -1
+
+		// Apply cache_control to the last two user messages if enabled
+		const enhancedMessages = formattedMessages.map((message, index) => {
+			if ((index === lastUserMsgIndex || index === secondLastUserMsgIndex) && cacheControl) {
+				return {
+					...message,
+					...cacheControl,
+				}
+			}
+			return message
+		})
+
+		const stream = await client.chat.completions.create({
+			model: this.options.ocaModelId || liteLlmDefaultModelId,
+			messages: [enhancedSystemMessage, ...enhancedMessages],
+			temperature,
+			stream: true,
+			max_completion_tokens: maxTokens,
+			max_tokens: maxTokens,
+			stream_options: { include_usage: true },
+			...(thinkingConfig && { thinking: thinkingConfig }), // Add thinking configuration when applicable
+			...(this.options.taskId && {
+				litellm_session_id: `cline-${this.options.taskId}`,
+			}), // Add session ID for LiteLLM tracking
+		})
+
+		const inputCost = (await this.calculateCost(1e6, 0)) || 0
+		const outputCost = (await this.calculateCost(0, 1e6)) || 0
+
+		for await (const chunk of stream) {
+			const delta = chunk.choices[0]?.delta
+
+			// Handle normal text content
+			if (delta?.content) {
+				yield {
+					type: "text",
+					text: delta.content,
+				}
+			}
+
+			// Handle reasoning events (thinking)
+			// Thinking is not in the standard types but may be in the response
+			interface ThinkingDelta {
+				thinking?: string
+			}
+
+			if ((delta as ThinkingDelta)?.thinking) {
+				yield {
+					type: "reasoning",
+					reasoning: (delta as ThinkingDelta).thinking || "",
+				}
+			}
+
+			// Handle token usage information
+			if (chunk.usage) {
+				const totalCost =
+					(inputCost * chunk.usage.prompt_tokens) / 1e6 + (outputCost * chunk.usage.completion_tokens) / 1e6
+
+				// Extract cache-related information if available
+				// Need to use type assertion since these properties are not in the standard OpenAI types
+				const usage = chunk.usage as {
+					prompt_tokens: number
+					completion_tokens: number
+					cache_creation_input_tokens?: number
+					prompt_cache_miss_tokens?: number
+					cache_read_input_tokens?: number
+					prompt_cache_hit_tokens?: number
+				}
+
+				const cacheWriteTokens = usage.cache_creation_input_tokens || usage.prompt_cache_miss_tokens || 0
+				const cacheReadTokens = usage.cache_read_input_tokens || usage.prompt_cache_hit_tokens || 0
+
+				yield {
+					type: "usage",
+					inputTokens: usage.prompt_tokens || 0,
+					outputTokens: usage.completion_tokens || 0,
+					cacheWriteTokens: cacheWriteTokens > 0 ? cacheWriteTokens : undefined,
+					cacheReadTokens: cacheReadTokens > 0 ? cacheReadTokens : undefined,
+					totalCost,
+				}
+			}
+		}
+	}
+
+	getModel() {
+		return {
+			id: this.options.ocaModelId || liteLlmDefaultModelId,
+			info: this.options.ocaModelInfo || liteLlmModelInfoSaneDefaults,
+		}
+	}
+}

src/core/api/providers/ollama.ts

@@ -29,7 +29,7 @@ export class OllamaHandler implements ApiHandler {
 		if (!this.client) {
 			try {
 				const clientOptions: Partial<Config> = {
-					host: this.options.ollamaBaseUrl || "http://localhost:11434",
+					host: this.options.ollamaBaseUrl,
 				}
 
 				// Add API key if provided (for Ollama cloud or authenticated instances)

src/core/api/providers/qwen-code.ts

@@ -4,7 +4,7 @@ import { ModelInfo, QwenCodeModelId, qwenCodeDefaultModelId, qwenCodeModels } fr
 import OpenAI from "openai"
 import * as os from "os"
 import * as path from "path"
-import { ApiHandler } from "../"
+import { ApiHandler, CommonApiHandlerOptions } from "../"
 import { withRetry } from "../retry"
 import { convertToOpenAiMessages } from "../transform/openai-format"
 import { ApiStream } from "../transform/stream"
@@ -24,7 +24,7 @@ interface QwenOAuthCredentials {
 	resource_url?: string
 }
 
-interface QwenCodeHandlerOptions {
+interface QwenCodeHandlerOptions extends CommonApiHandlerOptions {
 	qwenCodeOauthPath?: string
 	apiModelId?: string
 }

src/core/api/providers/requesty.ts

@@ -2,6 +2,7 @@ import { Anthropic } from "@anthropic-ai/sdk"
 import { ModelInfo, requestyDefaultModelId, requestyDefaultModelInfo } from "@shared/api"
 import { calculateApiCostOpenAI } from "@utils/cost"
 import OpenAI from "openai"
+import { toRequestyServiceStringUrl } from "@/shared/providers/requesty"
 import { ApiHandler, CommonApiHandlerOptions } from "../index"
 import { withRetry } from "../retry"
 import { convertToOpenAiMessages } from "../transform/openai-format"
@@ -41,7 +42,7 @@ export class RequestyHandler implements ApiHandler {
 			}
 			try {
 				this.client = new OpenAI({
-					baseURL: this.options.requestyBaseUrl || "https://router.requesty.ai/v1",
+					baseURL: toRequestyServiceStringUrl(this.options.requestyBaseUrl),
 					apiKey: this.options.requestyApiKey,
 					defaultHeaders: {
 						"HTTP-Referer": "https://cline.bot",
@@ -82,7 +83,6 @@ export class RequestyHandler implements ApiHandler {
 				? thinking
 				: {}
 
-		// @ts-ignore-next-line
 		const stream = await client.chat.completions.create({
 			model: model.id,
 			max_tokens: model.info.maxTokens || undefined,

src/core/api/providers/sapaicore.ts

@@ -4,6 +4,7 @@ import {
 	ConversationRole as BedrockConversationRole,
 	type Message as BedrockMessage,
 } from "@aws-sdk/client-bedrock-runtime"
+import { ChatMessages, LlmModuleConfig, OrchestrationClient, TemplatingModuleConfig } from "@sap-ai-sdk/orchestration"
 import { ModelInfo, SapAiCoreModelId, sapAiCoreDefaultModelId, sapAiCoreModels } from "@shared/api"
 import axios from "axios"
 import OpenAI from "openai"
@@ -18,7 +19,9 @@ interface SapAiCoreHandlerOptions extends CommonApiHandlerOptions {
 	sapAiResourceGroup?: string
 	sapAiCoreBaseUrl?: string
 	apiModelId?: string
+	sapAiCoreUseOrchestrationMode?: boolean
 	thinkingBudgetTokens?: number
+	deploymentId?: string
 	reasoningEffort?: string
 }
 
@@ -26,6 +29,7 @@ interface Deployment {
 	id: string
 	name: string
 }
+
 interface Token {
 	access_token: string
 	expires_in: number
@@ -350,19 +354,33 @@ export class SapAiCoreHandler implements ApiHandler {
 	private options: SapAiCoreHandlerOptions
 	private token?: Token
 	private deployments?: Deployment[]
+	private isAiCoreEnvSetup: boolean = false
 
 	constructor(options: SapAiCoreHandlerOptions) {
 		this.options = options
 	}
 
+	private validateCredentials(): void {
+		if (
+			!this.options.sapAiCoreClientId ||
+			!this.options.sapAiCoreClientSecret ||
+			!this.options.sapAiCoreTokenUrl ||
+			!this.options.sapAiCoreBaseUrl
+		) {
+			throw new Error("Missing required SAP AI Core credentials. Please check your configuration.")
+		}
+	}
+
 	private async authenticate(): Promise<Token> {
+		this.validateCredentials()
+
 		const payload = {
 			grant_type: "client_credentials",
-			client_id: this.options.sapAiCoreClientId || "",
-			client_secret: this.options.sapAiCoreClientSecret || "",
+			client_id: this.options.sapAiCoreClientId,
+			client_secret: this.options.sapAiCoreClientSecret,
 		}
 
-		const tokenUrl = (this.options.sapAiCoreTokenUrl || "").replace(/\/+$/, "") + "/oauth/token"
+		const tokenUrl = this.options.sapAiCoreTokenUrl!.replace(/\/+$/, "") + "/oauth/token"
 		const response = await axios.post(tokenUrl, payload, {
 			headers: { "Content-Type": "application/x-www-form-urlencoded" },
 		})
@@ -378,11 +396,8 @@ export class SapAiCoreHandler implements ApiHandler {
 		return this.token.access_token
 	}
 
+	// TODO: these fallback fetching deployment id methods can be removed in future version if decided that users migration to fetching deployment id in design-time (open SAP AI Core provider UI) considered as completed.
 	private async getAiCoreDeployments(): Promise<Deployment[]> {
-		if (this.options.sapAiCoreClientSecret === "") {
-			return [{ id: "notconfigured", name: "ai-core-not-configured" }]
-		}
-
 		const token = await this.getToken()
 		const headers = {
 			Authorization: `Bearer ${token}`,
@@ -440,6 +455,86 @@ export class SapAiCoreHandler implements ApiHandler {
 	}
 
 	async *createMessage(systemPrompt: string, messages: Anthropic.Messages.MessageParam[]): ApiStream {
+		if (this.options.sapAiCoreUseOrchestrationMode) {
+			yield* this.createMessageWithOrchestration(systemPrompt, messages)
+		} else {
+			yield* this.createMessageWithDeployments(systemPrompt, messages)
+		}
+	}
+
+	// TODO: support credentials changes after initial setup
+	private ensureAiCoreEnvSetup(): void {
+		// Only set up once to avoid redundant operations
+		if (this.isAiCoreEnvSetup) {
+			return
+		}
+
+		// Validate required credentials
+		this.validateCredentials()
+
+		const aiCoreServiceCredentials = {
+			clientid: this.options.sapAiCoreClientId!,
+			clientsecret: this.options.sapAiCoreClientSecret!,
+			url: this.options.sapAiCoreTokenUrl!,
+			serviceurls: {
+				AI_API_URL: this.options.sapAiCoreBaseUrl!,
+			},
+		}
+		process.env["AICORE_SERVICE_KEY"] = JSON.stringify(aiCoreServiceCredentials)
+
+		// Mark as set up to avoid redundant calls
+		this.isAiCoreEnvSetup = true
+	}
+
+	private async *createMessageWithOrchestration(systemPrompt: string, messages: Anthropic.Messages.MessageParam[]): ApiStream {
+		try {
+			// Ensure AI Core environment variable is set up (only runs once)
+			this.ensureAiCoreEnvSetup()
+			const model = this.getModel()
+
+			// Define the LLM to be used by the Orchestration pipeline
+			const llm: LlmModuleConfig = {
+				model_name: model.id,
+			}
+
+			const templating: TemplatingModuleConfig = {
+				template: [
+					{
+						role: "system",
+						content: systemPrompt,
+					},
+				],
+			}
+			const orchestrationClient = new OrchestrationClient(
+				{ llm, templating },
+				{ resourceGroup: this.options.sapAiResourceGroup || "default" },
+			)
+
+			const sapMessages = this.convertMessageParamToSAPMessages(messages)
+
+			const response = await orchestrationClient.stream({
+				messages: sapMessages,
+			})
+
+			for await (const chunk of response.stream.toContentStream()) {
+				yield { type: "text", text: chunk }
+			}
+
+			const tokenUsage = response.getTokenUsage()
+			if (tokenUsage) {
+				yield {
+					type: "usage",
+					inputTokens: tokenUsage.prompt_tokens || 0,
+					outputTokens: tokenUsage.completion_tokens || 0,
+				}
+			}
+		} catch (error) {
+			console.error("Error in SAP orchestration mode:", error)
+			throw error
+		}
+	}
+
+	private async *createMessageWithDeployments(systemPrompt: string, messages: Anthropic.Messages.MessageParam[]): ApiStream {
 		const token = await this.getToken()
 		const headers = {
 			Authorization: `Bearer ${token}`,
@@ -449,7 +544,13 @@ export class SapAiCoreHandler implements ApiHandler {
 		}
 
 		const model = this.getModel()
-		const deploymentId = await this.getDeploymentForModel(model.id)
+		let deploymentId = this.options.deploymentId
+
+		if (!deploymentId) {
+			// Fallback to runtime deployment id fetching for users who haven't opened the SAP provider UI
+			console.log(`No pre-configured deployment ID found for model ${model.id}, falling back to runtime fetching`)
+			deploymentId = await this.getDeploymentForModel(model.id)
+		}
 
 		const anthropicModels = [
 			"anthropic--claude-4-sonnet",
@@ -722,16 +823,13 @@ export class SapAiCoreHandler implements ApiHandler {
 
 							// Handle metadata (token usage)
 							if (data.metadata?.usage) {
+								// inputTokens does not include cached write/read tokens
 								let inputTokens = data.metadata.usage.inputTokens || 0
 								const outputTokens = data.metadata.usage.outputTokens || 0
 
-								// calibrate input token
-								const totalTokens = data.metadata.usage.totalTokens || 0
 								const cacheReadInputTokens = data.metadata.usage.cacheReadInputTokens || 0
-								const cacheWriteOutputTokens = data.metadata.usage.cacheWriteOutputTokens || 0
-								if (inputTokens + outputTokens + cacheReadInputTokens + cacheWriteOutputTokens !== totalTokens) {
-									inputTokens = totalTokens - outputTokens - cacheReadInputTokens - cacheWriteOutputTokens
-								}
+								const cacheWriteInputTokens = data.metadata.usage.cacheWriteInputTokens || 0
+								inputTokens = inputTokens + cacheReadInputTokens + cacheWriteInputTokens
 
 								yield {
 									type: "usage",
@@ -934,4 +1032,8 @@ export class SapAiCoreHandler implements ApiHandler {
 		}
 		return { id: sapAiCoreDefaultModelId, info: sapAiCoreModels[sapAiCoreDefaultModelId] }
 	}
+	private convertMessageParamToSAPMessages(messages: Anthropic.Messages.MessageParam[]): ChatMessages {
+		// Use the existing OpenAI converter since the logic is identical
+		return convertToOpenAiMessages(messages) as ChatMessages
+	}
 }

src/core/api/providers/vercel-ai-gateway.ts

@@ -1,12 +1,12 @@
 import { Anthropic } from "@anthropic-ai/sdk"
 import { ModelInfo, vercelAiGatewayDefaultModelId, vercelAiGatewayDefaultModelInfo } from "@shared/api"
 import OpenAI from "openai"
-import { ApiHandler } from "../index"
+import { ApiHandler, CommonApiHandlerOptions } from "../index"
 import { withRetry } from "../retry"
 import { ApiStream } from "../transform/stream"
 import { createVercelAIGatewayStream } from "../transform/vercel-ai-gateway-stream"
 
-interface VercelAIGatewayHandlerOptions {
+interface VercelAIGatewayHandlerOptions extends CommonApiHandlerOptions {
 	vercelAiGatewayApiKey?: string
 	vercelAiGatewayModelId?: string
 	vercelAiGatewayModelInfo?: ModelInfo

src/core/api/providers/vscode-lm.ts

@@ -3,7 +3,7 @@ import { ModelInfo, openAiModelInfoSaneDefaults } from "@shared/api"
 import { SELECTOR_SEPARATOR, stringifyVsCodeLmModelSelector } from "@shared/vsCodeSelectorUtils"
 import { calculateApiCostAnthropic } from "@utils/cost"
 import * as vscode from "vscode"
-import { ApiHandler, SingleCompletionHandler, CommonApiHandlerOptions } from "../"
+import { ApiHandler, CommonApiHandlerOptions, SingleCompletionHandler } from "../"
 import { withRetry } from "../retry"
 import { ApiStream } from "../transform/stream"
 import { convertToVsCodeLmMessages } from "../transform/vscode-lm-format"

src/core/api/providers/zai.ts

@@ -10,12 +10,12 @@ import {
 } from "@shared/api"
 import OpenAI from "openai"
 import { version as extensionVersion } from "../../../../package.json"
-import { ApiHandler } from ".."
+import { ApiHandler, CommonApiHandlerOptions } from ".."
 import { withRetry } from "../retry"
 import { convertToOpenAiMessages } from "../transform/openai-format"
 import { ApiStream } from "../transform/stream"
 
-interface ZAiHandlerOptions {
+interface ZAiHandlerOptions extends CommonApiHandlerOptions {
 	zaiApiLine?: string
 	zaiApiKey?: string
 	apiModelId?: string
@@ -85,7 +85,6 @@ export class ZAiHandler implements ApiHandler {
 			messages: openAiMessages,
 			stream: true,
 			stream_options: { include_usage: true },
-			temperature: 0,
 		})
 
 		for await (const chunk of stream) {

src/core/api/retry.ts

@@ -12,6 +12,18 @@ const DEFAULT_OPTIONS: Required<RetryOptions> = {
 	retryAllErrors: false,
 }
 
+export class RetriableError extends Error {
+	status: number = 429
+	retryAfter?: number
+
+	constructor(message: string, retryAfter?: number, options?: ErrorOptions) {
+		super(message, options)
+		this.name = "RetriableError"
+
+		this.retryAfter = retryAfter
+	}
+}
+
 export function withRetry(options: RetryOptions = {}) {
 	const { maxRetries, baseDelay, maxDelay, retryAllErrors } = { ...DEFAULT_OPTIONS, ...options }
 
@@ -24,7 +36,7 @@ export function withRetry(options: RetryOptions = {}) {
 					yield* originalMethod.apply(this, args)
 					return
 				} catch (error: any) {
-					const isRateLimit = error?.status === 429
+					const isRateLimit = error?.status === 429 || error instanceof RetriableError
 					const isLastAttempt = attempt === maxRetries - 1
 
 					if ((!isRateLimit && !retryAllErrors) || isLastAttempt) {
@@ -36,7 +48,8 @@ export function withRetry(options: RetryOptions = {}) {
 					const retryAfter =
 						error.headers?.["retry-after"] ||
 						error.headers?.["x-ratelimit-reset"] ||
-						error.headers?.["ratelimit-reset"]
+						error.headers?.["ratelimit-reset"] ||
+						error.retryAfter
 
 					let delay: number
 					if (retryAfter) {

src/core/api/transform/openrouter-stream.ts

@@ -1,7 +1,6 @@
 import { Anthropic } from "@anthropic-ai/sdk"
 import { CLAUDE_SONNET_4_1M_SUFFIX, ModelInfo, openRouterClaudeSonnet41mModelId } from "@shared/api"
 import OpenAI from "openai"
-import { isGPT5ModelFamily } from "../../prompts/system-prompt/utils"
 import { convertToOpenAiMessages } from "./openai-format"
 import { convertToR1Format } from "./r1-format"
 
@@ -140,10 +139,6 @@ export async function createOpenRouterStream(
 				reasoning = { max_tokens: budget_tokens }
 			}
 			break
-		case "cline/sonic":
-			temperature = 0.7
-			topP = 0.95
-			break
 		default:
 			if (thinkingBudgetTokens && model.info?.thinkingConfig && thinkingBudgetTokens > 0) {
 				temperature = undefined // extended thinking does not support non-1 temperature
@@ -152,17 +147,6 @@ export async function createOpenRouterStream(
 			}
 	}
 
-	// Removes messages in the middle when close to context window limit. Should not be applied to models that support prompt caching since it would continuously break the cache.
-	let shouldApplyMiddleOutTransform = !model.info.supportsPromptCache
-	// except for deepseek (which we set supportsPromptCache to true for), where because the context window is so small our truncation algo might miss and we should use openrouter's middle-out transform as a fallback to ensure we don't exceed the context window (FIXME: once we have a more robust token estimator we should not rely on this)
-	if (model.id === "deepseek/deepseek-chat") {
-		shouldApplyMiddleOutTransform = true
-	}
-
-	if (isGPT5ModelFamily(model.id)) {
-		shouldApplyMiddleOutTransform = false
-	}
-
 	// hardcoded provider sorting for kimi-k2
 	const isKimiK2 = model.id === "moonshotai/kimi-k2"
 	openRouterProviderSorting = isKimiK2 ? undefined : openRouterProviderSorting
@@ -176,7 +160,6 @@ export async function createOpenRouterStream(
 		messages: openAiMessages,
 		stream: true,
 		stream_options: { include_usage: true },
-		transforms: shouldApplyMiddleOutTransform ? ["middle-out"] : undefined,
 		include_reasoning: true,
 		...(model.id.startsWith("openai/o") ? { reasoning_effort: reasoningEffort || "medium" } : {}),
 		...(reasoning ? { reasoning } : {}),

src/core/assistant-message/index.ts

@@ -1,3 +1,4 @@
+import { ClineDefaultTool } from "@shared/tools"
 export type AssistantMessageContent = TextContent | ToolUse
 
 export { parseAssistantMessageV2 } from "./parse-assistant-message"
@@ -8,32 +9,6 @@ export interface TextContent {
 	partial: boolean
 }
 
-export const toolUseNames = [
-	"execute_command",
-	"read_file",
-	"write_to_file",
-	"replace_in_file",
-	"search_files",
-	"list_files",
-	"list_code_definition_names",
-	"browser_action",
-	"use_mcp_tool",
-	"access_mcp_resource",
-	"ask_followup_question",
-	"plan_mode_respond",
-	"load_mcp_documentation",
-	"attempt_completion",
-	"new_task",
-	"condense",
-	"summarize_task",
-	"report_bug",
-	"new_rule",
-	"web_fetch",
-] as const
-
-// Converts array of tool call names into a union type ("execute_command" | "read_file" | ...)
-export type ToolUseName = (typeof toolUseNames)[number]
-
 export const toolParamNames = [
 	"command",
 	"requires_approval",
@@ -63,13 +38,14 @@ export const toolParamNames = [
 	"additional_context",
 	"needs_more_exploration",
 	"task_progress",
+	"timeout",
 ] as const
 
 export type ToolParamName = (typeof toolParamNames)[number]
 
 export interface ToolUse {
 	type: "tool_use"
-	name: ToolUseName
+	name: ClineDefaultTool // id of the tool being used
 	// params is a partial record, allowing only some or none of the possible parameters to be used
 	params: Partial<Record<ToolParamName, string>>
 	partial: boolean

src/core/assistant-message/parse-assistant-message.ts

@@ -1,4 +1,5 @@
-import { AssistantMessageContent, TextContent, ToolParamName, ToolUse, ToolUseName, toolParamNames, toolUseNames } from "." // Assuming types are defined in index.ts or a similar file
+import { ClineDefaultTool, toolUseNames } from "@shared/tools"
+import { AssistantMessageContent, TextContent, ToolParamName, ToolUse, toolParamNames } from "." // Assuming types are defined in index.ts or a similar file
 
 // parseAssistantmessageV1 removed in https://github.com/cline/cline/pull/5425
 
@@ -33,7 +34,7 @@ export function parseAssistantMessageV2(assistantMessage: string): AssistantMess
 	let currentParamName: ToolParamName | undefined
 
 	// Precompute tags for faster lookups
-	const toolUseOpenTags = new Map<string, ToolUseName>()
+	const toolUseOpenTags = new Map<string, ClineDefaultTool>()
 	const toolParamOpenTags = new Map<string, ToolParamName>()
 	for (const name of toolUseNames) {
 		toolUseOpenTags.set(`<${name}>`, name)

src/core/commands/reconstructTaskHistory.ts

@@ -0,0 +1,286 @@
+import {
+	ensureTaskDirectoryExists,
+	getSavedClineMessages,
+	getTaskMetadata,
+	readTaskHistoryFromState,
+	writeTaskHistoryToState,
+} from "@core/storage/disk"
+import { HostProvider } from "@hosts/host-provider"
+import { ClineMessage } from "@shared/ExtensionMessage"
+import { HistoryItem } from "@shared/HistoryItem"
+import { ShowMessageType } from "@shared/proto/host/window"
+import { fileExistsAtPath } from "@utils/fs"
+import * as path from "path"
+import { ulid } from "ulid"
+import * as vscode from "vscode"
+
+interface TaskReconstructionResult {
+	totalTasks: number
+	reconstructedTasks: number
+	skippedTasks: number
+	errors: string[]
+}
+
+/**
+ * Reconstructs task history from existing task folders
+ */
+export async function reconstructTaskHistory(context: vscode.ExtensionContext): Promise<void> {
+	try {
+		// Show confirmation dialog using HostProvider
+		const proceed = await HostProvider.window.showMessage({
+			type: ShowMessageType.WARNING,
+			message:
+				"This will rebuild your task history from existing task data. This operation will backup your current task history and attempt to reconstruct it from task folders. Continue?",
+			options: {
+				items: ["Yes, Reconstruct", "Cancel"],
+			},
+		})
+
+		if (proceed?.selectedOption !== "Yes, Reconstruct") {
+			return
+		}
+
+		// Show initial progress message
+		HostProvider.window.showMessage({
+			type: ShowMessageType.INFORMATION,
+			message: "Reconstructing task history...",
+		})
+
+		const result = await performTaskHistoryReconstruction(context)
+
+		// Show results
+		if (result.errors.length > 0) {
+			const errorMessage = `Reconstruction completed with warnings:\n- Reconstructed: ${result.reconstructedTasks} tasks\n- Skipped: ${result.skippedTasks} tasks\n- Errors: ${result.errors.length}\n\nFirst few errors:\n${result.errors.slice(0, 3).join("\n")}`
+
+			HostProvider.window.showMessage({
+				type: ShowMessageType.WARNING,
+				message: errorMessage,
+			})
+		} else {
+			HostProvider.window.showMessage({
+				type: ShowMessageType.INFORMATION,
+				message: `Task history successfully reconstructed! Found and restored ${result.reconstructedTasks} tasks.`,
+			})
+		}
+	} catch (error) {
+		const errorMessage = error instanceof Error ? error.message : String(error)
+		HostProvider.window.showMessage({
+			type: ShowMessageType.ERROR,
+			message: `Failed to reconstruct task history: ${errorMessage}`,
+		})
+	}
+}
+
+async function performTaskHistoryReconstruction(context: vscode.ExtensionContext): Promise<TaskReconstructionResult> {
+	const result: TaskReconstructionResult = {
+		totalTasks: 0,
+		reconstructedTasks: 0,
+		skippedTasks: 0,
+		errors: [],
+	}
+
+	// Backup existing task history
+	await backupExistingTaskHistory(context)
+
+	// Get tasks directory
+	const globalStoragePath = context.globalStorageUri.fsPath
+	const tasksDir = path.join(globalStoragePath, "tasks")
+
+	// Check if tasks directory exists
+	if (!(await fileExistsAtPath(tasksDir))) {
+		throw new Error("No tasks directory found. Nothing to reconstruct.")
+	}
+
+	// Scan for task directories
+	const taskIds = await scanTaskDirectories(tasksDir)
+	result.totalTasks = taskIds.length
+
+	if (taskIds.length === 0) {
+		throw new Error("No task directories found. Nothing to reconstruct.")
+	}
+
+	// Process each task
+	const reconstructedItems: HistoryItem[] = []
+
+	for (const taskId of taskIds) {
+		try {
+			const historyItem = await reconstructTaskHistoryItem(context, taskId)
+			if (historyItem) {
+				reconstructedItems.push(historyItem)
+				result.reconstructedTasks++
+			} else {
+				result.skippedTasks++
+			}
+		} catch (error) {
+			result.skippedTasks++
+			const errorMsg = error instanceof Error ? error.message : String(error)
+			result.errors.push(`Task ${taskId}: ${errorMsg}`)
+		}
+	}
+
+	// Sort by timestamp (newest first)
+	reconstructedItems.sort((a, b) => b.ts - a.ts)
+
+	// Write reconstructed history
+	await writeTaskHistoryToState(reconstructedItems)
+
+	return result
+}
+
+async function backupExistingTaskHistory(context: vscode.ExtensionContext): Promise<void> {
+	try {
+		const existingHistory = await readTaskHistoryFromState()
+		if (existingHistory.length > 0) {
+			const backupPath = path.join(context.globalStorageUri.fsPath, "state", `taskHistory.backup.${Date.now()}.json`)
+
+			// Ensure state directory exists
+			const fs = await import("fs/promises")
+			await fs.mkdir(path.dirname(backupPath), { recursive: true })
+			await fs.writeFile(backupPath, JSON.stringify(existingHistory, null, 2))
+		}
+	} catch (error) {
+		// Non-fatal error, just log it
+		console.warn("Failed to backup existing task history:", error)
+	}
+}
+
+async function scanTaskDirectories(tasksDir: string): Promise<string[]> {
+	const fs = await import("fs/promises")
+
+	try {
+		const entries = await fs.readdir(tasksDir, { withFileTypes: true })
+		return entries
+			.filter((entry) => entry.isDirectory())
+			.map((entry) => entry.name)
+			.filter((name) => /^\d+$/.test(name)) // Only numeric task IDs
+	} catch (error) {
+		throw new Error(`Failed to scan tasks directory: ${error}`)
+	}
+}
+
+async function reconstructTaskHistoryItem(context: vscode.ExtensionContext, taskId: string): Promise<HistoryItem | null> {
+	try {
+		// Get task directory
+		const taskDir = await ensureTaskDirectoryExists(context, taskId)
+
+		// Load UI messages to extract task info
+		const clineMessages = await getSavedClineMessages(context, taskId)
+		if (clineMessages.length === 0) {
+			return null // Skip empty tasks
+		}
+
+		// Load task metadata for token usage
+		const metadata = await getTaskMetadata(context, taskId)
+
+		// Extract task information
+		const taskInfo = extractTaskInformation(clineMessages, metadata)
+
+		// Create HistoryItem
+		const historyItem: HistoryItem = {
+			id: taskId,
+			ulid: taskInfo.ulid || ulid(), // Generate new ULID if missing
+			ts: taskInfo.timestamp,
+			task: taskInfo.taskDescription,
+			tokensIn: taskInfo.tokensIn,
+			tokensOut: taskInfo.tokensOut,
+			cacheWrites: taskInfo.cacheWrites,
+			cacheReads: taskInfo.cacheReads,
+			totalCost: taskInfo.totalCost,
+			size: taskInfo.size,
+			isFavorited: taskInfo.isFavorited,
+			conversationHistoryDeletedRange: taskInfo.conversationHistoryDeletedRange,
+		}
+
+		return historyItem
+	} catch (error) {
+		throw new Error(`Failed to reconstruct task ${taskId}: ${error}`)
+	}
+}
+
+interface TaskInfo {
+	ulid?: string
+	timestamp: number
+	taskDescription: string
+	tokensIn: number
+	tokensOut: number
+	cacheWrites?: number
+	cacheReads?: number
+	totalCost: number
+	size?: number
+	isFavorited?: boolean
+	conversationHistoryDeletedRange?: [number, number]
+}
+
+function extractTaskInformation(clineMessages: ClineMessage[], metadata: any): TaskInfo {
+	// Find the first user message (task description)
+	const firstUserMessage = clineMessages.find((msg) => msg.type === "say" && msg.say === "text" && msg.text)
+
+	// Extract timestamp from first message or use task ID as fallback
+	const timestamp = clineMessages.length > 0 ? clineMessages[0].ts : Date.now()
+
+	// Extract task description
+	let taskDescription = "Untitled Task"
+	if (firstUserMessage?.text) {
+		// Clean up the task description
+		const cleanText = firstUserMessage.text
+			.replace(/<task>\s*/g, "")
+			.replace(/\s*<\/task>/g, "")
+			.trim()
+
+		const firstLine = cleanText.split("\n")[0]
+		if (firstLine) {
+			taskDescription = firstLine.substring(0, 100) // Limit length
+		}
+	}
+
+	// Calculate token usage from API request messages
+	let tokensIn = 0
+	let tokensOut = 0
+	let cacheWrites = 0
+	let cacheReads = 0
+	let totalCost = 0
+
+	// Look for api_req_started messages with token info
+	const apiReqMessages = clineMessages.filter((msg) => msg.type === "say" && msg.say === "api_req_started" && msg.text)
+
+	for (const msg of apiReqMessages) {
+		try {
+			if (msg.text) {
+				const apiInfo = JSON.parse(msg.text)
+				if (apiInfo.tokensIn) tokensIn += apiInfo.tokensIn
+				if (apiInfo.tokensOut) tokensOut += apiInfo.tokensOut
+				if (apiInfo.cacheWrites) cacheWrites += apiInfo.cacheWrites
+				if (apiInfo.cacheReads) cacheReads += apiInfo.cacheReads
+				if (apiInfo.cost) totalCost += apiInfo.cost
+			}
+		} catch {
+			// Ignore parsing errors
+		}
+	}
+
+	// Use metadata if available and no tokens found in messages
+	if (tokensIn === 0 && tokensOut === 0 && metadata.model_usage) {
+		for (const usage of metadata.model_usage) {
+			tokensIn += usage.tokensIn || 0
+			tokensOut += usage.tokensOut || 0
+			cacheWrites += usage.cacheWrites || 0
+			cacheReads += usage.cacheReads || 0
+			totalCost += usage.totalCost || 0
+		}
+	}
+
+	// Calculate approximate size (rough estimate)
+	const messageSize = JSON.stringify(clineMessages).length
+	const size = Math.floor(messageSize / 1024) // KB
+
+	return {
+		timestamp,
+		taskDescription,
+		tokensIn,
+		tokensOut,
+		cacheWrites: cacheWrites > 0 ? cacheWrites : undefined,
+		cacheReads: cacheReads > 0 ? cacheReads : undefined,
+		totalCost,
+		size,
+	}
+}

src/core/context/context-management/ContextManager.ts

@@ -108,15 +108,22 @@ export class ContextManager {
 	/**
 	 * Determine whether we should compact context window, based on token counts
 	 */
-	shouldCompactContextWindow(clineMessages: ClineMessage[], api: ApiHandler, previousApiReqIndex: number): boolean {
+	shouldCompactContextWindow(
+		clineMessages: ClineMessage[],
+		api: ApiHandler,
+		previousApiReqIndex: number,
+		thresholdPercentage?: number,
+	): boolean {
 		if (previousApiReqIndex >= 0) {
 			const previousRequest = clineMessages[previousApiReqIndex]
 			if (previousRequest && previousRequest.text) {
 				const { tokensIn, tokensOut, cacheWrites, cacheReads }: ClineApiReqInfo = JSON.parse(previousRequest.text)
 				const totalTokens = (tokensIn || 0) + (tokensOut || 0) + (cacheWrites || 0) + (cacheReads || 0)
 
-				const { maxAllowedSize } = getContextWindowInfo(api)
-				return totalTokens >= maxAllowedSize
+				const { contextWindow, maxAllowedSize } = getContextWindowInfo(api)
+				const roundedThreshold = thresholdPercentage ? Math.floor(contextWindow * thresholdPercentage) : maxAllowedSize
+				const thresholdTokens = Math.min(roundedThreshold, maxAllowedSize)
+				return totalTokens >= thresholdTokens
 			}
 		}
 		return false
@@ -291,7 +298,7 @@ export class ContextManager {
 
 		// Make sure that the last message being removed is a assistant message, so the next message after the initial user-assistant pair is an assistant message. This preserves the user-assistant-user-assistant structure.
 		// NOTE: anthropic format messages are always user-assistant-user-assistant, while openai format messages can have multiple user messages in a row (we use anthropic format throughout cline)
-		if (apiMessages[rangeEndIndex].role !== "assistant") {
+		if (apiMessages[rangeEndIndex] && apiMessages[rangeEndIndex].role !== "assistant") {
 			rangeEndIndex -= 1
 		}
 

src/core/context/context-tracking/FileContextTracker.test.ts

@@ -48,7 +48,7 @@ describe("FileContextTracker", () => {
 
 		// Mock controller and context
 		mockController = {
-			context: { globalStorageUri: { fsPath: "/mock/storage" } } as vscode.ExtensionContext,
+			context: {} as vscode.ExtensionContext,
 		} as unknown as Controller
 
 		// Mock disk module functions

src/core/context/context-tracking/FileContextTracker.ts

@@ -1,10 +1,9 @@
-import { getTaskMetadata, saveTaskMetadata } from "@core/storage/disk"
+import { getTaskMetadata, readTaskHistoryFromState, saveTaskMetadata } from "@core/storage/disk"
 import type { ClineMessage } from "@shared/ExtensionMessage"
 import chokidar, { FSWatcher } from "chokidar"
 import * as path from "path"
 import * as vscode from "vscode"
 import { Controller } from "@/core/controller"
-import { HistoryItem } from "@/shared/HistoryItem"
 import { getCwd } from "@/utils/path"
 import type { FileMetadataEntry } from "./ContextTrackerTypes"
 
@@ -243,7 +242,7 @@ export class FileContextTracker {
 			const key = `pendingFileContextWarning_${this.taskId}`
 			// NOTE: Using 'as any' because dynamic keys like pendingFileContextWarning_${taskId}
 			// are legitimate workspace state keys but don't fit the strict LocalStateKey type system
-			this.controller.cacheService.setWorkspaceState(key as any, files)
+			this.controller.stateManager.setWorkspaceState(key as any, files)
 		} catch (error) {
 			console.error("Error storing pending file context warning:", error)
 		}
@@ -255,7 +254,7 @@ export class FileContextTracker {
 	async retrievePendingFileContextWarning(): Promise<string[] | undefined> {
 		try {
 			const key = `pendingFileContextWarning_${this.taskId}`
-			const files = this.controller.cacheService.getWorkspaceStateKey(key as any) as string[]
+			const files = this.controller.stateManager.getWorkspaceStateKey(key as any) as string[]
 			return files
 		} catch (error) {
 			console.error("Error retrieving pending file context warning:", error)
@@ -270,7 +269,7 @@ export class FileContextTracker {
 		try {
 			const files = await this.retrievePendingFileContextWarning()
 			if (files) {
-				this.controller.cacheService.setWorkspaceState(`pendingFileContextWarning_${this.taskId}` as any, undefined)
+				this.controller.stateManager.setWorkspaceState(`pendingFileContextWarning_${this.taskId}` as any, undefined)
 				return files
 			}
 		} catch (error) {
@@ -286,8 +285,7 @@ export class FileContextTracker {
 	static async cleanupOrphanedWarnings(context: vscode.ExtensionContext): Promise<void> {
 		const startTime = Date.now()
 		try {
-			// eslint-disable-next-line eslint-rules/no-direct-vscode-state-api
-			const taskHistory = (context.globalState.get("taskHistory") as HistoryItem[]) || []
+			const taskHistory = await readTaskHistoryFromState()
 			const existingTaskIds = new Set(taskHistory.map((task) => task.id))
 			const allStateKeys = context.workspaceState.keys()
 			const pendingWarningKeys = allStateKeys.filter((key) => key.startsWith("pendingFileContextWarning_"))
@@ -312,7 +310,7 @@ export class FileContextTracker {
 				`FileContextTracker: Processed ${existingTaskIds.size} tasks, found ${pendingWarningKeys.length} pending warnings, ${orphanedPendingContextTasks.length} orphaned, deleted ${orphanedPendingContextTasks.length}, took ${duration}ms`,
 			)
 		} catch (error) {
-			console.error("Error cleaning up orphaned file context warnings:", error)
+			console.error("[FileContextTracker] Error cleaning up orphaned file context warnings:", error)
 		}
 	}
 }

src/core/context/instructions/user-instructions/cline-rules.ts

@@ -74,18 +74,18 @@ export async function refreshClineRulesToggles(
 	localToggles: ClineRulesToggles
 }> {
 	// Global toggles
-	const globalClineRulesToggles = controller.cacheService.getGlobalStateKey("globalClineRulesToggles")
+	const globalClineRulesToggles = controller.stateManager.getGlobalSettingsKey("globalClineRulesToggles")
 	const globalClineRulesFilePath = await ensureRulesDirectoryExists()
 	const updatedGlobalToggles = await synchronizeRuleToggles(globalClineRulesFilePath, globalClineRulesToggles)
-	controller.cacheService.setGlobalState("globalClineRulesToggles", updatedGlobalToggles)
+	controller.stateManager.setGlobalState("globalClineRulesToggles", updatedGlobalToggles)
 
 	// Local toggles
-	const localClineRulesToggles = controller.cacheService.getWorkspaceStateKey("localClineRulesToggles")
+	const localClineRulesToggles = controller.stateManager.getWorkspaceStateKey("localClineRulesToggles")
 	const localClineRulesFilePath = path.resolve(workingDirectory, GlobalFileNames.clineRules)
 	const updatedLocalToggles = await synchronizeRuleToggles(localClineRulesFilePath, localClineRulesToggles, "", [
 		[".clinerules", "workflows"],
 	])
-	controller.cacheService.setWorkspaceState("localClineRulesToggles", updatedLocalToggles)
+	controller.stateManager.setWorkspaceState("localClineRulesToggles", updatedLocalToggles)
 
 	return {
 		globalToggles: updatedGlobalToggles,

src/core/context/instructions/user-instructions/external-rules.ts

@@ -23,13 +23,13 @@ export async function refreshExternalRulesToggles(
 	cursorLocalToggles: ClineRulesToggles
 }> {
 	// local windsurf toggles
-	const localWindsurfRulesToggles = controller.cacheService.getWorkspaceStateKey("localWindsurfRulesToggles")
+	const localWindsurfRulesToggles = controller.stateManager.getWorkspaceStateKey("localWindsurfRulesToggles")
 	const localWindsurfRulesFilePath = path.resolve(workingDirectory, GlobalFileNames.windsurfRules)
 	const updatedLocalWindsurfToggles = await synchronizeRuleToggles(localWindsurfRulesFilePath, localWindsurfRulesToggles)
-	controller.cacheService.setWorkspaceState("localWindsurfRulesToggles", updatedLocalWindsurfToggles)
+	controller.stateManager.setWorkspaceState("localWindsurfRulesToggles", updatedLocalWindsurfToggles)
 
 	// local cursor toggles
-	const localCursorRulesToggles = controller.cacheService.getWorkspaceStateKey("localCursorRulesToggles")
+	const localCursorRulesToggles = controller.stateManager.getWorkspaceStateKey("localCursorRulesToggles")
 
 	// cursor has two valid locations for rules files, so we need to check both and combine
 	// synchronizeRuleToggles will drop whichever rules files are not in each given path, but combining the results will result in no data loss
@@ -40,7 +40,7 @@ export async function refreshExternalRulesToggles(
 	const updatedLocalCursorToggles2 = await synchronizeRuleToggles(localCursorRulesFilePath, localCursorRulesToggles)
 
 	const updatedLocalCursorToggles = combineRuleToggles(updatedLocalCursorToggles1, updatedLocalCursorToggles2)
-	controller.cacheService.setWorkspaceState("localCursorRulesToggles", updatedLocalCursorToggles)
+	controller.stateManager.setWorkspaceState("localCursorRulesToggles", updatedLocalCursorToggles)
 
 	return {
 		windsurfLocalToggles: updatedLocalWindsurfToggles,

src/core/context/instructions/user-instructions/rule-helpers.ts

@@ -247,31 +247,31 @@ export async function deleteRuleFile(
 		// Update the appropriate toggles
 		if (isGlobal) {
 			if (type === "workflow") {
-				const toggles = controller.cacheService.getGlobalStateKey("globalWorkflowToggles")
+				const toggles = controller.stateManager.getGlobalSettingsKey("globalWorkflowToggles")
 				delete toggles[rulePath]
-				controller.cacheService.setGlobalState("globalWorkflowToggles", toggles)
+				controller.stateManager.setGlobalState("globalWorkflowToggles", toggles)
 			} else {
-				const toggles = controller.cacheService.getGlobalStateKey("globalClineRulesToggles")
+				const toggles = controller.stateManager.getGlobalSettingsKey("globalClineRulesToggles")
 				delete toggles[rulePath]
-				controller.cacheService.setGlobalState("globalClineRulesToggles", toggles)
+				controller.stateManager.setGlobalState("globalClineRulesToggles", toggles)
 			}
 		} else {
 			if (type === "workflow") {
-				const toggles = controller.cacheService.getWorkspaceStateKey("workflowToggles")
+				const toggles = controller.stateManager.getWorkspaceStateKey("workflowToggles")
 				delete toggles[rulePath]
-				controller.cacheService.setWorkspaceState("workflowToggles", toggles)
+				controller.stateManager.setWorkspaceState("workflowToggles", toggles)
 			} else if (type === "cursor") {
-				const toggles = controller.cacheService.getWorkspaceStateKey("localCursorRulesToggles")
+				const toggles = controller.stateManager.getWorkspaceStateKey("localCursorRulesToggles")
 				delete toggles[rulePath]
-				controller.cacheService.setWorkspaceState("localCursorRulesToggles", toggles)
+				controller.stateManager.setWorkspaceState("localCursorRulesToggles", toggles)
 			} else if (type === "windsurf") {
-				const toggles = controller.cacheService.getWorkspaceStateKey("localWindsurfRulesToggles")
+				const toggles = controller.stateManager.getWorkspaceStateKey("localWindsurfRulesToggles")
 				delete toggles[rulePath]
-				controller.cacheService.setWorkspaceState("localWindsurfRulesToggles", toggles)
+				controller.stateManager.setWorkspaceState("localWindsurfRulesToggles", toggles)
 			} else {
-				const toggles = controller.cacheService.getWorkspaceStateKey("localClineRulesToggles")
+				const toggles = controller.stateManager.getWorkspaceStateKey("localClineRulesToggles")
 				delete toggles[rulePath]
-				controller.cacheService.setWorkspaceState("localClineRulesToggles", toggles)
+				controller.stateManager.setWorkspaceState("localClineRulesToggles", toggles)
 			}
 		}
 

src/core/context/instructions/user-instructions/workflows.ts

@@ -15,15 +15,15 @@ export async function refreshWorkflowToggles(
 	localWorkflowToggles: ClineRulesToggles
 }> {
 	// Global workflows
-	const globalWorkflowToggles = controller.cacheService.getGlobalStateKey("globalWorkflowToggles")
+	const globalWorkflowToggles = controller.stateManager.getGlobalSettingsKey("globalWorkflowToggles")
 	const globalClineWorkflowsFilePath = await ensureWorkflowsDirectoryExists()
 	const updatedGlobalWorkflowToggles = await synchronizeRuleToggles(globalClineWorkflowsFilePath, globalWorkflowToggles)
-	controller.cacheService.setGlobalState("globalWorkflowToggles", updatedGlobalWorkflowToggles)
+	controller.stateManager.setGlobalState("globalWorkflowToggles", updatedGlobalWorkflowToggles)
 
-	const workflowRulesToggles = controller.cacheService.getWorkspaceStateKey("workflowToggles")
+	const workflowRulesToggles = controller.stateManager.getWorkspaceStateKey("workflowToggles")
 	const workflowsDirPath = path.resolve(workingDirectory, GlobalFileNames.workflows)
 	const updatedWorkflowToggles = await synchronizeRuleToggles(workflowsDirPath, workflowRulesToggles)
-	controller.cacheService.setWorkspaceState("workflowToggles", updatedWorkflowToggles)
+	controller.stateManager.setWorkspaceState("workflowToggles", updatedWorkflowToggles)
 
 	return {
 		globalWorkflowToggles: updatedGlobalWorkflowToggles,

src/core/controller/account/authStateChanged.ts

@@ -11,7 +11,7 @@ import type { Controller } from "../index"
 export async function authStateChanged(controller: Controller, request: AuthStateChangedRequest): Promise<AuthState> {
 	try {
 		// Store the user info directly in global state
-		controller.cacheService.setGlobalState("userInfo", request.user)
+		controller.stateManager.setGlobalState("userInfo", request.user)
 
 		// Return the same user info
 		return AuthState.create({ user: request.user })