Support prompt caching and thinking for Opus 4.1

* feat: Added Claude Opus 4.1 to Bedrock

* Create ninety-owls-develop.md

.changeset/famous-baboons-vanish.md

@@ -1,5 +0,0 @@
----
-"claude-dev": patch
----
-
-Fixed issue with sap ai core client credentials storage

.changeset/lovely-experts-cry.md

@@ -0,0 +1,5 @@
+---
+"claude-dev": patch
+---
+
+Fix prompt caching and support extended thinking for Opus 4.1

.changeset/ninety-beers-unite.md

@@ -1,5 +0,0 @@
----
-"claude-dev": patch
----
-
-Fix Qwen Api option inconsistency between UI and API layer

.changeset/ninety-owls-develop.md

@@ -0,0 +1,5 @@
+---
+"claude-dev": patch
+---
+
+feat: Added Claude Opus 4.1 to Bedrock

.changeset/olive-eyes-decide.md

@@ -1,5 +0,0 @@
----
-"claude-dev": patch
----
-
-Fix credit balance out of sync issue on account switching

.changeset/silly-hornets-raise.md

@@ -1,5 +0,0 @@
----
-"claude-dev": patch
----
-
-Fix Claude Code CLAUDE_CODE_MAX_OUTPUT_TOKENS

.clinerules/cline-overview.md

@@ -716,7 +716,7 @@ The Controller class manages MCP servers through the McpHub service:
 class Controller {
   mcpHub?: McpHub
 
-  constructor(context: vscode.ExtensionContext, outputChannel: vscode.OutputChannel, webviewProvider: WebviewProvider) {
+  constructor(context: vscode.ExtensionContext, webviewProvider: WebviewProvider) {
     this.mcpHub = new McpHub(this)
   }
 

.github/ISSUE_TEMPLATE/bug_report.yml

@@ -5,7 +5,7 @@ body:
     - type: markdown
       attributes:
           value: |
-              **Important:** All bug reports must be reproducible using Claude 3.5 Sonnet. Cline uses complex prompts so less capable models may not work as expected.
+              **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:
@@ -24,7 +24,7 @@ body:
               2.
               3.
       validations:
-          required: true
+          required: false
     - type: textarea
       id: logs
       attributes:
@@ -39,20 +39,19 @@ body:
           placeholder: "e.g., cline:anthropic/claude-3.7-sonnet, gemini:gemini-2.5-pro-exp-03-25"
       validations:
           required: true
-    - type: input
-      id: operating-system
-      attributes:
-          label: Operating System
-          description: What operating system are you using?
-          placeholder: "e.g., Windows 11, macOS Sonoma, Ubuntu 22.04"
-      validations:
-          required: true
     - type: textarea
       id: system-info
       attributes:
-          label: System Info
-          description: What system information is relevant to the issue?
-          placeholder: "e.g., CPU: Intel Core i7-11700K, GPU: NVIDIA GeForce RTX 3070, RAM: 32GB DDR4"
+          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
@@ -63,8 +62,3 @@ body:
           placeholder: "e.g., 1.2.3"
       validations:
           required: true
-    - type: textarea
-      id: additional-context
-      attributes:
-          label: Additional context
-          description: Add any other context about the problem here, such as screenshots or related issues.

.github/ISSUE_TEMPLATE/feature_contribution.yml

@@ -1,116 +0,0 @@
-name: 💡 Feature Proposal & Contribution
-description: Propose a new feature or improvement, and optionally offer to implement feature as a contributor
-labels: ["proposal"]
-body:
-    - type: markdown
-      attributes:
-          value: |
-              **Feature Proposal & Contribution for Cline**
-
-              Thank you for proposing a feature or improvement for Cline! This template helps us understand the problem, evaluate the solution, and coordinate implementation.
-
-              **For detailed proposals:** Please provide comprehensive information to enable fast prioritization and discussion.
-              **For contribution offers:** You can indicate your willingness to implement the feature yourself.
-
-              Before submitting:
-              - Search existing [Issues](https://github.com/cline/cline/issues) and [Discussions](https://github.com/cline/cline/discussions) to avoid duplicates
-              - Read the [Contributing Guide](https://github.com/cline/cline/blob/main/CONTRIBUTING.md) if you plan to contribute
-              - Don't start implementation until the proposal is reviewed and approved
-
-    - type: textarea
-      id: problem-description
-      attributes:
-          label: What problem does this solve?
-          description: |
-              Describe the problem clearly from a user's point of view. Focus on why this matters, who it affects, and when it occurs.
-
-              ✅ Good examples:
-              - "LLM provider returns 400 error when nearing the context window instead of truncating"
-              - "Submit button is invisible in dark mode"
-              - "Users can't easily share their Cline configurations with team members"
-
-              ❌ Avoid vague descriptions:
-              - "Performance is bad"
-              - "UI needs work"
-
-              Your description should include:
-              - Who is affected?
-              - When does it happen?
-              - What's the current vs expected behavior?
-              - What is the impact?
-          placeholder: Be specific about the problem, who it affects, and the impact.
-      validations:
-          required: true
-
-    - type: textarea
-      id: proposed-solution
-      attributes:
-          label: What's the proposed solution?
-          description: |
-              Describe how the problem should be solved. Be specific about UX, system behavior, and any flows that would change.
-
-              ✅ Good examples:
-              - "Add error handling immediately after attempting to create the llm stream and retry after manually truncating"
-              - "Update button styling to ensure contrast in all themes"
-              - "Add export/import functionality in settings with JSON format"
-
-              ❌ Avoid vague solutions:
-              - "Improve performance"
-              - "Fix the bug"
-
-              Your solution should include:
-              - What exactly will change?
-              - How will users interact with it?
-              - What's the expected outcome?
-          placeholder: Describe the proposed changes and how they solve the problem.
-      validations:
-          required: false
-
-    - type: dropdown
-      id: contribution-intent
-      attributes:
-          label: Are you interested in implementing this?
-          description: Let us know if you'd like to contribute to this feature
-          options:
-              - "No, just proposing the idea"
-              - "Yes, I'd like to implement this myself"
-              - "Yes, I'd like to collaborate with others"
-              - "Maybe, depending on complexity and guidance"
-      validations:
-          required: false
-
-    - type: textarea
-      id: implementation-approach
-      attributes:
-          label: Implementation approach (if contributing)
-          description: |
-              **Only fill this out if you selected "Yes" above.**
-
-              How do you plan to implement this? Include:
-              - High-level technical approach
-              - Files/components that would be affected
-              - Any new dependencies required
-              - Potential challenges or considerations you've identified
-
-              This helps us provide better guidance and ensures alignment before you start coding.
-          placeholder: "My implementation approach would be..."
-
-    - type: checkboxes
-      id: checklist
-      attributes:
-          label: Proposal checklist
-          options:
-              - label: I've checked for existing issues or related proposals
-                required: true
-              - label: I understand this needs review before implementation can start
-                required: true
-
-    - type: checkboxes
-      id: contribution-checklist
-      attributes:
-          label: Contribution checklist (if contributing)
-          description: Only check these if you plan to contribute
-          options:
-              - label: I've read the [Contributing Guide](https://github.com/cline/cline/blob/main/CONTRIBUTING.md)
-              - label: I'm willing to make changes based on feedback
-              - label: I understand the code review process and requirements

.github/pull_request_template.md

@@ -2,15 +2,14 @@
 Thank you for contributing to Cline!
 
 ⚠️ Important: Before submitting this PR, please ensure you have:
-- Opened an issue and discussed your proposed changes with the community / contributors
-- Received approval from a core Cline contributor prior to proceeding with the implementation
-- Link the associated issue in the "Related Issue" section
+- For feature requests: Created a discussion in our Feature Requests discussions board https://github.com/cline/cline/discussions/categories/feature-requests and received approval from core maintainers before implementation
+- For all changes: Link the associated issue/discussion in the "Related Issue" section below
 
 Limited exceptions:
-Small bug fixes, typo corrections, minor wording improvements, or simple type fixes that don't change functionality may be submitted directly.
+Small bug fixes, typo corrections, minor wording improvements, or simple type fixes that don't change functionality may be submitted directly without prior discussion.
 
 Why this requirement?
-We deeply appreciate all community contributions - they are the core reason we're able to operate successfully and keep innovating! We welcome community input and want to make it as easy as possible for people to submit quality work. This process helps our core maintainers review new ideas faster and saves contributor time by ensuring you have the go-ahead before spending time on implementation.
+We deeply appreciate all community contributions - they are essential to Cline's success! To ensure the best use of everyone's time and maintain project direction, we use our Feature Requests discussions board to gauge community interest and validate feature ideas before implementation begins. This helps us focus development efforts on features that will benefit the most users.
 -->
 
 ### Related Issue

.github/workflows/test.yml

@@ -96,16 +96,15 @@ jobs:
             - name: Build Tests and Extension
               run: npm run pretest
 
-            # Unit Tests disabled due to module system conflicts between backend and webview-ui
-            # - name: Unit Tests
-            #   run: npm run test:unit
+            - name: Unit Tests
+              run: npm run test:unit
 
             # Run extension tests with coverage
-            - name: Extension Tests with Coverage
+            - name: Extension Integration Tests with Coverage
               id: extension_coverage
               continue-on-error: true
               run: |
-                  node ./scripts/test-ci.js > extension_coverage.txt 2>&1
+                  node ./scripts/test-ci.js 2>&1 | tee extension_coverage.txt
                   # Default the encoding to UTF-8 - It's not the default on Windows
                   PYTHONUTF8=1 PYTHONPATH=.github/scripts python -m coverage_check extract-coverage extension_coverage.txt --type=extension --github-output --verbose
 
@@ -117,7 +116,7 @@ jobs:
                   cd webview-ui
                   # Ensure coverage dependency is installed
                   npm install --no-save @vitest/coverage-v8
-                  npm run test:coverage > webview_coverage.txt 2>&1
+                  npm run test:coverage 2>&1 | tee webview_coverage.txt
                   cd ..
                   # Default the encoding to UTF-8 - It's not the default on Windows
                   PYTHONUTF8=1 PYTHONPATH=.github/scripts python -m coverage_check extract-coverage webview-ui/webview_coverage.txt --type=webview --github-output --verbose
@@ -132,21 +131,19 @@ jobs:
                   path: |
                       extension_coverage.txt
                       webview-ui/webview_coverage.txt
-                  retention-period: workflow # Artifacts are automatically deleted when the workflow completes
 
             # Set the check as failed if any of the tests failed
-            - name: Print test results and check for failures
+            - name: Check for test failures
               run: |
-                  echo "Extension Tests Result: ${{ steps.extension_coverage.outcome }}"
-                  cat extension_coverage.txt
-
-                  echo "Webview Tests Result: ${{ steps.webview_coverage.outcome }}"
-                  cat webview-ui/webview_coverage.txt
-
                   # Check if any of the test steps failed
                   # https://docs.github.com/en/actions/writing-workflows/choosing-what-your-workflow-does/accessing-contextual-information-about-workflow-runs#steps-context
+                  if [ "${{ steps.extension_coverage.outcome }}" != "success" ]; then
+                      echo "Extension Integration Tests failed, see previous step for test output."
+                  fi
+                  if [ "${{ steps.webview_coverage.outcome }}" != "success" ]; then
+                      echo "Webview Tests failed, see previous step for test output."
+                  fi
                   if [ "${{ steps.extension_coverage.outcome }}" != "success" ] || [ "${{ steps.webview_coverage.outcome }}" != "success" ]; then
-                      echo "Tests failed."
                       exit 1
                   fi
 

.vscode/launch.json

@@ -71,7 +71,7 @@
 		{
 			"type": "node",
 			"request": "launch",
-			"name": "Run Standalone Service",
+			"name": "Run cline-core service",
 			"skipFiles": ["<node_internals>/**"],
 			"sourceMaps": true,
 			"resolveSourceMapLocations": ["${workspaceFolder}/**", "!**/node_modules/**"],
@@ -82,11 +82,9 @@
 				// Turns on grpc debug log.
 				//"GRPC_TRACE": "all",
 				//"GRPC_VERBOSITY": "DEBUG",
-				"NODE_PATH": "${workspaceFolder}/dist-standalone/node_modules",
-
-				"HOST_BRIDGE_ADDRESS": "localhost:50052"
+				"NODE_PATH": "${workspaceFolder}/dist-standalone/node_modules"
 			},
-			"program": "standalone.js"
+			"program": "cline-core.js"
 		}
 	]
 }

.vscode/settings.json

@@ -9,5 +9,9 @@
 		"dist": true // set this to false to include "dist" folder in search results
 	},
 	// Turn off tsc task auto detection since we have the necessary tasks as npm scripts
-	"typescript.tsc.autoDetect": "off"
+	"typescript.tsc.autoDetect": "off",
+	// Protobuf settings
+	"protoc": {
+		"options": ["--proto_path=proto"]
+	}
 }

CHANGELOG.md

@@ -1,5 +1,63 @@
 # Changelog
 
+## [3.20.11]
+
+Add gpt-oss-120b as a Cerebras model
+Add Opus 4.1 through Claude Code
+
+## [3.20.10]
+
+- Add OpenAI's new open-source models (GPT-OSS-120B and GPT-OSS-20B) to Hugging Face and Groq providers
+
+## [3.20.9]
+
+- Add support for Claude Opus 4.1 model in Anthropic provider
+- Add Baseten as a new API provider with support for DeepSeek, Llama, and Kimi K2 models (Thanks @AlexKer!)
+- Fix error messages not clearing from UI when retrying failed tasks
+- Fix chat input box positioning issues
+
+## [3.20.8]
+
+- Add navbar tooltips on hover
+
+## [3.20.7]
+
+- Fix circular dependency that affect the github workflow Tests / test (pull_request)
+
+## [3.20.6]
+
+- Fix login check on extension restart
+
+## [3.20.5]
+
+- Fix authentication persistence issues that could cause users to be logged out unexpectedly
+
+## [3.20.4]
+
+- Add new Cerebras models
+- Update rate limits for existing Cerebras models
+- Fix for delete task dialog
+
+## [3.20.3]
+
+- Add Huawei Cloud MaaS Provider (Thanks @ddling!)
+- Add Cerebras Qwen 3 235B instruct model (Thanks @kevint-cerebras!)
+- Add DeepSeek R1 0528 support under Hugging Face (Thanks @0ne0rZer0!)
+- Fix Global Rules directory documentation for Linux/WSL systems
+- Fix token counting when using VSCode LM API provider
+- Fix input field stealing focus issue by only focusing on visible and active editor panels
+- Fix duplicate tool registration for claude4-experimental
+- Trim input value for URL fields
+
+## [3.20.2]
+
+- Fixed issue with sap ai core client credentials storage
+- Fix Qwen Api option inconsistency between UI and API layer
+- Fix credit balance out of sync issue on account switching
+- Fix Claude Code CLAUDE_CODE_MAX_OUTPUT_TOKENS
+- Fix cursor state after restoring files to be disabled after checked out
+- Fix issue where checkpointing blocked UI
+
 ## [3.20.1]
 
 - Fix for files being deleted when switching modes or closing tasks

CONTRIBUTING.md

@@ -14,14 +14,11 @@ Bug reports help make Cline better for everyone! Before creating a new issue, pl
 ## Before Contributing
 
 All contributions must begin with a GitHub Issue, unless the change is for small bug fixes, typo corrections, minor wording improvements, or simple type fixes that don't change functionality.
-
-- **Check existing issues**: Search [GitHub Issues](https://github.com/cline/cline/issues).
-- **Create an issue**: Use appropriate templates:
-    - **Contributions:** Use the "Contribution Request" template to propose what you'd like to work on.
-    - **Bugs:** "Bug Report" template for reporting issues.
-    - **Features:** "Detailed Feature Proposal" template for suggesting new features.
-- **Wait for approval**: A core Cline contributor must approve your contribution request before you start implementation.
-- **Claim issues**: Once approved, the issue will be assigned to you.
+**For features and contributions**:
+- First check the [Feature Requests discussions board](https://github.com/cline/cline/discussions/categories/feature-requests) for similar ideas
+- If your idea is new, create a new feature request  
+- Wait for approval from core maintainers before starting implementation
+- Once approved, feel free to begin working on a PR with the help of our community!
 
 **PRs without approved issues may be closed.**
 

buf.yaml

@@ -9,7 +9,6 @@ lint:
 
     except: # Add exceptions for current patterns that contradict STANDARD settings
         - RPC_PASCAL_CASE # rpcs are camel case (start with lowercase)
-        - PACKAGE_DIRECTORY_MATCH # the protos in the cline package are not in a dir named cline.
         - RPC_REQUEST_RESPONSE_UNIQUE # request messages are not unique.
         - RPC_REQUEST_STANDARD_NAME # request messages dont all end with Request
         - RPC_RESPONSE_STANDARD_NAME # response messages dont all end with Response

docs/features/cline-rules.mdx

@@ -11,7 +11,19 @@ You can create a rule by clicking the `+` button in the Rules tab. This will ope
 Once you save the file:
 
 -   Your rule will be stored in the `.clinerules/` directory in your project (if it's a Workspace Rule)
--   Or in the `Documents/Cline/Rules` directory (if it's a Global Rule).
+-   Or in the Global Rules directory (if it's a Global Rule):
+
+### Global Rules Directory Location
+
+The location of your Global Rules directory depends on your operating system:
+
+| Operating System | Default Location | Notes |
+|------------------|------------------|-------|
+| **Windows** | `Documents\Cline\Rules` | Uses system Documents folder |
+| **macOS** | `~/Documents/Cline/Rules` | Uses user Documents folder |
+| **Linux/WSL** | `~/Documents/Cline/Rules` | May fall back to `~/Cline/Rules` on some systems |
+
+> **Note for Linux/WSL users**: If you don't find your global rules in `~/Documents/Cline/Rules`, check `~/Cline/Rules` as the location may vary depending on your system configuration and whether the Documents directory exists.
 
 You can also have Cline create a rule for you by using the [`/newrule` slash command](/features/slash-commands/new-rule) in the chat.
 

docs/mcp/adding-mcp-servers-from-github.mdx

@@ -17,6 +17,7 @@ There are multiple places online to find MCP servers:
     -   [mcpservers.org](https://mcpservers.org/)
     -   [mcp.so](https://mcp.so/)
     -   [glama.ai/mcp/servers](https://glama.ai/mcp/servers)
+    -   [mcp.composio.dev](https://mcp.composio.dev/)
 
     These directories allow users to sort the servers by various criteria such as downloads, date, stars, and use case. Each entry provides information such as features, tools, and configuration instructions.
 

docs/more-info/telemetry.mdx

@@ -4,17 +4,17 @@ title: "Telemetry"
 
 ### Overview
 
-To help make Cline better for everyone, we collect anonymous usage data that helps us understand how developers are using our open-source AI coding agent. This feedback loop is crucial for improving Cline's capabilities and user experience.
+To help make Cline better for everyone, we collect usage data that helps us understand how developers are using our open-source AI coding agent. This feedback loop is crucial for improving Cline's capabilities and user experience.
 
 We use PostHog, an open-source analytics platform, for data collection and analysis. Our telemetry implementation is fully transparent - you can review the [source code](https://github.com/cline/cline/blob/main/src/services/posthog/telemetry/TelemetryService.ts) to see exactly what we track.
 
 ### Tracking Policy
 
-Privacy is our priority. All collected data is anonymized before being sent to PostHog, with no personally identifiable information (PII) included. Your code, prompts, and conversation content always remain private and are never collected.
+Privacy is our priority. By default, all collected data is anonymized. If you log in with a Cline account, your telemetry data will be associated with your account to help us improve the product and provide better support when you encounter issues. Your code, prompts, and conversation content always remain private and are never collected.
 
 ### What We Track
 
-We collect basic anonymous usage data including:
+We collect basic usage data including:
 
 **Task Interactions:** When tasks start and finish, conversation flow (without content)\
 **Mode and Tool Usage:** Switches between plan/act modes, which tools are being used\
@@ -28,7 +28,7 @@ For complete transparency, you can inspect our [telemetry implementation](https:
 
 Telemetry in Cline is entirely optional:
 
--   When you update or install our VS Code extension, you'll see a message about our anonymous telemetry
+-   When you update or install our VS Code extension, you'll see a message about our telemetry
 -   You can change your preference anytime in settings
 
 Cline also respects VS Code's global telemetry settings. If you've disabled telemetry at the VS Code level, Cline's telemetry will automatically be disabled as well.

esbuild.mjs

@@ -1,6 +1,10 @@
-const esbuild = require("esbuild")
-const fs = require("fs")
-const path = require("path")
+import fs from "node:fs"
+import * as esbuild from "esbuild"
+import path from "node:path"
+import { fileURLToPath } from "node:url"
+
+const __filename = fileURLToPath(import.meta.url)
+const __dirname = path.dirname(__filename)
 
 const production = process.argv.includes("--production")
 const watch = process.argv.includes("--watch")
@@ -127,10 +131,8 @@ const baseConfig = {
 	sourcemap: !production,
 	logLevel: "silent",
 	define: production
-		? {
-				"process.env.IS_DEV": JSON.stringify(!production),
-			}
-		: undefined,
+		? { "import.meta.url": "_importMetaUrl", "process.env.IS_DEV": JSON.stringify(!production) }
+		: { "import.meta.url": "_importMetaUrl" },
 	tsconfig: path.resolve(__dirname, "tsconfig.json"),
 	plugins: [
 		copyWasmFiles,
@@ -141,6 +143,9 @@ const baseConfig = {
 	format: "cjs",
 	sourcesContent: false,
 	platform: "node",
+	banner: {
+		js: "const _importMetaUrl=require('url').pathToFileURL(__filename)",
+	},
 }
 
 // Extension-specific configuration

eslint-rules/__tests__/no-direct-vscode-api.test.ts

@@ -34,25 +34,30 @@ directApiRuleTester.run("no-direct-vscode-api", noDirectVscodeApiRule, {
 		},
 		// Should allow other vscode API calls
 		{
-			code: `vscode.window.showInformationMessage("Hello")`,
-			filename: "test.ts",
+			code: `vscode.commands.registerCommand("Hello")`,
+			filename: "/foo/bar.ts",
 		},
 		// Should allow postMessage calls on other objects
 		{
 			code: `window.postMessage({ type: "test" }, "*")`,
-			filename: "test.ts",
+			filename: "/foo/bar.ts",
 		},
 		// Should allow variables named vscode but not calling postMessage
 		{
 			code: `const vscode = { other: "method" }; vscode.other()`,
-			filename: "test.ts",
+			filename: "/foo/bar.ts",
+		},
+		// Should allow vscode.postMessage in test files
+		{
+			code: `vscode.postMessage({ type: "newTask", text: message.text })`,
+			filename: "/foo/bar.test.ts",
 		},
 	],
 	invalid: [
 		// Should disallow vscode.postMessage in regular files
 		{
 			code: `vscode.postMessage({ type: "test", data: {} })`,
-			filename: "test.ts",
+			filename: "/foo/bar.ts",
 			errors: [
 				{
 					messageId: "useGrpcClient",
@@ -69,23 +74,13 @@ directApiRuleTester.run("no-direct-vscode-api", noDirectVscodeApiRule, {
 				},
 			],
 		},
-		// Should disallow vscode.postMessage in test files
-		{
-			code: `vscode.postMessage({ type: "newTask", text: message.text })`,
-			filename: "test.test.ts",
-			errors: [
-				{
-					messageId: "useGrpcClient",
-				},
-			],
-		},
 		// Should disallow property access for disallowed APIs
 		{
 			code: `const folders = vscode.workspace.workspaceFolders;`,
 			filename: "workspace.ts",
 			errors: [
 				{
-					messageId: "useHostBridge",
+					messageId: "useHostBridgeWorkspace",
 				},
 			],
 		},

eslint-rules/no-direct-vscode-api.js

@@ -35,18 +35,24 @@ const disallowedApis = {
 	"vscode.env.openExternal": {
 		messageId: "useUtils",
 	},
-	// "vscode.window.showWarningMessage": {
-	// 	messageId: "useHostBridgeShowMessage",
-	// },
+	"vscode.window.showWarningMessage": {
+		messageId: "useHostBridgeShowMessage",
+	},
 	"vscode.window.showOpenDialog": {
 		messageId: "useHostBridgeShowMessage",
 	},
 	"vscode.window.showErrorMessage": {
 		messageId: "useHostBridgeShowMessage",
 	},
-	// "vscode.window.showInformationMessage": {
-	// 	messageId: "useHostBridgeShowMessage",
-	// },
+	"vscode.window.showInformationMessage": {
+		messageId: "useHostBridgeShowMessage",
+	},
+	"vscode.window.showInputBox": {
+		messageId: "useHostBridge",
+	},
+	"vscode.workspace.findFiles": {
+		messageId: "useNative",
+	},
 }
 
 module.exports = createRule({
@@ -87,6 +93,10 @@ module.exports = createRule({
 				"Use utilities in @/utils instead of calling vscode APIs directly.\n" +
 				"This provides a consistent abstraction across VSCode and standalone environments.\n" +
 				"Found: {{code}}",
+			useNative:
+				"Use a native Javascript API instead of calling the vscode API.\n" +
+				"This provides a consistent abstraction across VSCode and standalone environments.\n" +
+				"Found: {{code}}",
 		},
 		schema: [],
 	},
@@ -187,7 +197,7 @@ module.exports = createRule({
 			if (filename.includes("/standalone/runtime-files/")) {
 				return true
 			}
-			// Skip unit tests
+			// Skip checking test files
 			if (filename.endsWith(".test.ts")) {
 				return true
 			}

evals/cli/src/commands/runDiffEval.ts

@@ -10,6 +10,7 @@ interface RunDiffEvalOptions {
 	parsingFunction: string
 	diffEditFunction: string
 	thinkingBudget: number
+	provider: string
 	parallel: boolean
 	verbose: boolean
 	testPath: string
@@ -39,6 +40,8 @@ export async function runDiffEvalHandler(options: RunDiffEvalOptions) {
 		options.parsingFunction,
 		"--diff-edit-function",
 		options.diffEditFunction,
+		"--provider",
+		options.provider,
 	]
 
 	// Conditionally add the optional arguments

evals/cli/src/index.ts

@@ -92,6 +92,7 @@ program
 	.option("--parsing-function <name>", "The parsing function to use", "parseAssistantMessageV2")
 	.option("--diff-edit-function <name>", "The diff editing function to use", "constructNewFileContentV2")
 	.option("--thinking-budget <tokens>", "Set the thinking tokens budget", "0")
+	.option("--provider <provider>", "API provider to use (openrouter, openai)", "openrouter")
 	.option("--parallel", "Run tests in parallel", false)
 	.option("--replay", "Run evaluation from a pre-recorded LLM output, skipping the API call", false)
 	.option("--replay-run-id <run_id>", "The ID of the run to replay from the database")

evals/diff-edits/ClineWrapper.ts

@@ -1,4 +1,5 @@
 import { OpenRouterHandler } from "../../src/api/providers/openrouter"
+import { OpenAiNativeHandler } from "../../src/api/providers/openai-native"
 import { ApiHandlerOptions } from "../../src/shared/api"
 import { Anthropic } from "@anthropic-ai/sdk"
 
@@ -54,7 +55,7 @@ interface StreamResult {
  * Process the stream and return full response with timing data
  */
 async function processStream(
-	handler: OpenRouterHandler,
+	handler: OpenRouterHandler | OpenAiNativeHandler,
 	systemPrompt: string,
 	messages: Anthropic.Messages.MessageParam[],
 ): Promise<StreamResult> {
@@ -190,19 +191,7 @@ export async function runSingleEvaluation(input: TestInput): Promise<TestResult>
 			}
 		}
 
-		const options: ApiHandlerOptions = {
-			openRouterApiKey: apiKey,
-			openRouterModelId: modelId,
-			thinkingBudgetTokens: thinkingBudgetTokens,
-			openRouterModelInfo: {
-				maxTokens: 10_000,
-				contextWindow: 1_000_000,
-				supportsImages: true,
-				supportsPromptCache: true, // may need to turn this on
-				inputPrice: 0,
-				outputPrice: 0,
-			},
-		}
+		const provider = input.provider || "openrouter"
 
 		// Get the output of streaming output of this llm call
 		let streamResult: StreamResult
@@ -214,10 +203,34 @@ export async function runSingleEvaluation(input: TestInput): Promise<TestResult>
 				usage: { inputTokens: 0, outputTokens: 0, cacheWriteTokens: 0, cacheReadTokens: 0, totalCost: 0 },
 			}
 		} else {
-			// Live mode: existing API call logic
+			// Live mode: provider-specific API call logic
 			try {
-				const openRouterHandler = new OpenRouterHandler(options)
-				streamResult = await processStream(openRouterHandler, systemPrompt, messages)
+				let handler: OpenRouterHandler | OpenAiNativeHandler
+				
+				if (provider === "openai") {
+					const openAiOptions = {
+						openAiNativeApiKey: apiKey,
+						apiModelId: modelId,
+					}
+					handler = new OpenAiNativeHandler(openAiOptions)
+				} else {
+					const openRouterOptions = {
+						openRouterApiKey: apiKey,
+						openRouterModelId: modelId,
+						thinkingBudgetTokens: thinkingBudgetTokens,
+						openRouterModelInfo: {
+							maxTokens: 10_000,
+							contextWindow: 1_000_000,
+							supportsImages: true,
+							supportsPromptCache: true,
+							inputPrice: 0,
+							outputPrice: 0,
+						},
+					}
+					handler = new OpenRouterHandler(openRouterOptions)
+				}
+				
+				streamResult = await processStream(handler, systemPrompt, messages)
 			} catch (error: any) {
 				return {
 					success: false,

evals/diff-edits/TestRunner.ts

@@ -49,16 +49,25 @@ type TestResultSet = { [test_id: string]: (TestResult & { test_id?: string })[]
 
 class NodeTestRunner {
 	private apiKey: string | undefined
+	private provider: string
 	private currentRunId: string | null = null
 	private systemPromptHash: string | null = null
 	private processingFunctionsHash: string | null = null
 	private caseIdMap: Map<string, string> = new Map() // test_id -> case_id mapping
 
-	constructor(isReplay: boolean) {
+	constructor(isReplay: boolean, provider: string = "openrouter") {
+		this.provider = provider
 		if (!isReplay) {
-			this.apiKey = process.env.OPENROUTER_API_KEY
-			if (!this.apiKey) {
-				throw new Error("OPENROUTER_API_KEY environment variable not set for a non-replay run.")
+			if (provider === "openai") {
+				this.apiKey = process.env.OPENAI_API_KEY
+				if (!this.apiKey) {
+					throw new Error("OPENAI_API_KEY environment variable not set for a non-replay run with OpenAI provider.")
+				}
+			} else {
+				this.apiKey = process.env.OPENROUTER_API_KEY
+				if (!this.apiKey) {
+					throw new Error("OPENROUTER_API_KEY environment variable not set for a non-replay run with OpenRouter provider.")
+				}
 			}
 		}
 	}
@@ -635,6 +644,7 @@ class NodeTestRunner {
 			thinkingBudgetTokens: testConfig.thinking_tokens_budget,
 			originalDiffEditToolCallMessage: testConfig.replay ? testCase.original_diff_edit_tool_call_message : undefined,
 			diffApplyFile: testConfig.diff_apply_file,
+			provider: this.provider,
 			isVerbose: isVerbose,
 		}
 
@@ -927,6 +937,7 @@ async function main() {
 		.option("--parsing-function <name>", "The parsing function to use", "parseAssistantMessageV2")
 		.option("--diff-edit-function <name>", "The diff editing function to use", "diff-06-26-25")
 		.option("--thinking-budget <tokens>", "Set the thinking tokens budget", "0")
+		.option("--provider <provider>", "API provider to use (openrouter, openai)", "openrouter")
 		.option("--parallel", "Run tests in parallel", false)
 		.option("--replay", "Run evaluation from a pre-recorded LLM output, skipping the API call", false)
 		.option("--replay-run-id <run_id>", "The ID of the run to replay from the database")
@@ -959,7 +970,7 @@ async function main() {
 		? parseInt(options.maxAttemptsPerCase, 10)
 		: validAttemptsPerCase * 10;
 
-	const runner = new NodeTestRunner(options.replay || !!options.replayRunId)
+	const runner = new NodeTestRunner(options.replay || !!options.replayRunId, options.provider)
 
 	if (options.replayRunId) {
 		if (!options.diffApplyFile) {
@@ -979,7 +990,7 @@ async function main() {
 			log(isVerbose, "Warning: Could not load OpenRouter model data. Context window filtering might be affected for OpenRouter models.");
 		}
 
-		const runner = new NodeTestRunner(options.replay)
+		const runner = new NodeTestRunner(options.replay, options.provider)
 		let allLoadedTestCases = runner.loadTestCases(testPath, isVerbose) // Pass isVerbose
 		
 		const allProcessedTestCasesGlobal: ProcessedTestCase[] = allLoadedTestCases.map((tc) => ({

evals/diff-edits/dashboard/app.py

@@ -331,6 +331,42 @@ def get_performance_grade(success_rate):
     else:
         return "C", "poor"
 
+def get_error_description(error_enum, error_string=None):
+    """Map error enum values to user-friendly descriptions"""
+    error_map = {
+        1: "No tool calls - Model didn't use the replace_in_file tool",
+        2: "Multiple tool calls - Model called multiple tools instead of one", 
+        3: "Wrong tool call - Model used wrong tool (not replace_in_file)",
+        4: "Missing parameters - Tool call missing required path or diff",
+        5: "Wrong file edited - Model edited different file than expected",
+        6: "Wrong tool call - Model used wrong tool type",
+        7: "Wrong file edited - Model targeted incorrect file path",
+        8: "API/Stream error - Problem with model API connection",
+        9: "Configuration error - Invalid evaluation parameters",
+        10: "Function error - Invalid parsing/diff functions",
+        11: "Other error - Unexpected failure"
+    }
+    
+    base_description = error_map.get(error_enum, f"Unknown error (code: {error_enum})")
+    
+    if error_string:
+        return f"{base_description}: {error_string}"
+    return base_description
+
+def get_error_guidance(error_enum):
+    """Provide specific guidance based on error type"""
+    guidance_map = {
+        1: "💡 The model provided a response but didn't use the replace_in_file tool. Check the raw output to see what the model actually said.",
+        2: "💡 The model called multiple tools when it should only call replace_in_file once. Check the parsed tool call section.",
+        3: "💡 The model used a different tool instead of replace_in_file. This might indicate confusion about the task.",
+        4: "💡 The model called replace_in_file but didn't provide the required 'path' or 'diff' parameters.",
+        5: "💡 The model tried to edit a different file than expected. Check the parsed tool call to see which file it targeted.",
+        6: "💡 The model used the wrong tool type. Check the raw output to see what tool it attempted to use.",
+        7: "💡 The model tried to edit a different file path than expected. This could indicate path confusion or hallucination.",
+    }
+    
+    return guidance_map.get(error_enum, "")
+
 def render_hero_section(current_run, model_performance):
     """Render the hero section with key metrics"""
     run_title = current_run['description'] if current_run['description'] else f"Run {current_run['run_id'][:8]}..."
@@ -570,12 +606,16 @@ def render_result_detail(result):
     """Render detailed view of a single result"""
     st.markdown("### 🔬 Result Deep Dive")
     
-    # Check if this is a valid result
-    is_valid = (result['error_enum'] not in [1, 6, 7]) if not pd.isna(result['error_enum']) else True
+    # Check if this is a valid result (only invalid if no tool calls or wrong file)
+    is_valid = True
+    if not pd.isna(result['error_enum']):
+        # Only these specific errors make a result "invalid" for the benchmark:
+        # 1 = no_tool_calls, 5 = wrong_file_edited, 7 = wrong_file_edited
+        is_valid = result['error_enum'] not in [1, 5, 7]
     
     # Show validity warning if needed
     if not is_valid:
-        st.warning("⚠️ **This is an invalid result** - The model didn't properly call the diff edit tool or edited the wrong file. This result is excluded from success rate calculations.")
+        st.warning("⚠️ **This is an invalid result** - The model didn't call the replace_in_file tool or edited the wrong file. This result is excluded from success rate calculations.")
     
     # Result metadata
     col1, col2, col3, col4 = st.columns(4)
@@ -591,7 +631,10 @@ def render_result_detail(result):
         st.markdown(f"**Round Trip:** {result['time_round_trip_ms']:.0f}ms")
     
     with col4:
-        st.markdown(f"**Cost:** ${result['cost_usd']:.4f}")
+        if pd.notna(result['cost_usd']) and result['cost_usd'] is not None:
+            st.markdown(f"**Cost:** ${result['cost_usd']:.4f}")
+        else:
+            st.markdown(f"**Cost:** Free")
     
     # Tabbed interface for different views
     tab1, tab2, tab3, tab4 = st.tabs(["📄 File & Edits", "🤖 Raw Output", "🔧 Parsed Tool Call", "📊 Metrics"])
@@ -693,8 +736,46 @@ def render_file_and_edits_view(result):
             # Show error information
             st.error("❌ **Edit Failed**")
             
+            # Show detailed error reason
             if not pd.isna(result['error_enum']):
-                st.markdown(f"**Error Code:** {result['error_enum']}")
+                error_description = get_error_description(
+                    result['error_enum'], 
+                    result.get('error_string')
+                )
+                st.markdown(f"**Reason:** {error_description}")
+                
+                # Show specific guidance based on error type
+                guidance = get_error_guidance(result['error_enum'])
+                if guidance:
+                    st.info(guidance)
+            
+            # For valid results that failed, check for diff application failures
+            elif not result['succeeded']:
+                # This is a valid result that failed - likely due to diff application issues
+                raw_output = result.get('raw_model_output', '')
+                
+                # Check if we have specific error information in the raw output
+                if 'does not match anything in the file' in str(raw_output).lower():
+                    st.warning("⚠️ **Diff Application Failed**")
+                    st.info("💡 The SEARCH block in the diff didn't match any content in the original file. This usually means the model hallucinated code that doesn't exist.")
+                elif 'malformatted' in str(raw_output).lower() or 'malformed' in str(raw_output).lower():
+                    st.warning("⚠️ **Diff Format Error**")
+                    st.info("💡 The diff format was incorrect. Check the raw tool call to see the formatting issues.")
+                elif 'error:' in str(raw_output).lower():
+                    # Try to extract the specific error message
+                    lines = str(raw_output).split('\n')
+                    error_lines = [line for line in lines if 'error:' in line.lower()]
+                    if error_lines:
+                        error_msg = error_lines[0].strip()
+                        st.warning("⚠️ **Diff Application Failed**")
+                        st.info(f"💡 {error_msg}")
+                    else:
+                        st.warning("⚠️ **Diff Application Failed**")
+                        st.info("💡 The diff couldn't be applied to the original file. Check the raw output and parsed tool call for more details.")
+                else:
+                    # Generic diff application failure
+                    st.warning("⚠️ **Diff Application Failed**")
+                    st.info("💡 The model made a valid tool call but the diff couldn't be applied to the original file. This usually indicates a mismatch between the expected and actual file content.")
         else:
             # Show successful edit information
             st.success("✅ **Edit Successful**")
@@ -725,8 +806,25 @@ def render_file_and_edits_view(result):
                     if len(edited_lines) > 50:
                         st.text(f"... ({len(edited_lines) - 50} more lines)")
         
-        # Show parsed tool call if available
+        # Show raw and parsed tool calls if available
         if not pd.isna(result['parsed_tool_call_json']):
+            with st.expander("View Raw Tool Call"):
+                # Extract the raw tool call text from the model output
+                raw_output = result['raw_model_output'] if not pd.isna(result['raw_model_output']) else ""
+                
+                # Try to extract just the tool call portion
+                if raw_output and '<replace_in_file>' in raw_output:
+                    # Find the tool call block
+                    start_idx = raw_output.find('<replace_in_file>')
+                    end_idx = raw_output.find('</replace_in_file>') + len('</replace_in_file>')
+                    if start_idx != -1 and end_idx != -1:
+                        raw_tool_call = raw_output[start_idx:end_idx]
+                        st.code(raw_tool_call, language='xml')
+                    else:
+                        st.text("Tool call not found in raw output")
+                else:
+                    st.text("No raw tool call available")
+            
             with st.expander("View Parsed Tool Call"):
                 try:
                     parsed_call = json.loads(result['parsed_tool_call_json'])
@@ -795,8 +893,10 @@ def render_metrics_view(result):
         if not pd.isna(result['completion_tokens']):
             st.metric("Completion Tokens", int(result['completion_tokens']))
         
-        if not pd.isna(result['cost_usd']):
+        if pd.notna(result['cost_usd']) and result['cost_usd'] is not None:
             st.metric("Cost", f"${result['cost_usd']:.4f}")
+        else:
+            st.metric("Cost", "Free")
         
         if not pd.isna(result['tokens_in_context']):
             st.metric("Context Tokens", int(result['tokens_in_context']))

evals/diff-edits/types.ts

@@ -104,5 +104,6 @@ export interface TestInput {
 	thinkingBudgetTokens: number
 	originalDiffEditToolCallMessage?: string
 	diffApplyFile?: string
+	provider?: string
 	isVerbose: boolean
 }

package-lock.json

@@ -1,12 +1,12 @@
 {
 	"name": "claude-dev",
-	"version": "3.20.0",
+	"version": "3.20.11",
 	"lockfileVersion": 2,
 	"requires": true,
 	"packages": {
 		"": {
 			"name": "claude-dev",
-			"version": "3.20.0",
+			"version": "3.20.11",
 			"license": "Apache-2.0",
 			"dependencies": {
 				"@anthropic-ai/sdk": "^0.37.0",
@@ -7907,10 +7907,9 @@
 			"integrity": "sha512-AlcaJBi/pqqJBIQ8U9Mcpc9i8Aqxn88Skv5d+xBX006BY5u8N3mGLHa5Lgppa7L/HfwgwLgZ6NYs+Ag6uUmJRA=="
 		},
 		"node_modules/brace-expansion": {
-			"version": "2.0.1",
-			"resolved": "https://registry.npmjs.org/brace-expansion/-/brace-expansion-2.0.1.tgz",
-			"integrity": "sha512-XnAIvQ8eM+kC6aULx6wuQiwVsnzsi9d3WxzV3FpWTGA19F621kwdbsAcFKXgKUHZWsy+mY6iL1sHTxWEFCytDA==",
-			"license": "MIT",
+			"version": "2.0.2",
+			"resolved": "https://registry.npmjs.org/brace-expansion/-/brace-expansion-2.0.2.tgz",
+			"integrity": "sha512-Jt0vHyM+jmUBqojB7E1NIYadt0vI0Qxjxd2TErW94wDz+E2LAm5vKMXXwg6ZZBTHPuUlDgQHKXvjGBdfcF1ZDQ==",
 			"dependencies": {
 				"balanced-match": "^1.0.0"
 			}
@@ -17786,10 +17785,9 @@
 			}
 		},
 		"node_modules/tar-fs": {
-			"version": "3.0.8",
-			"resolved": "https://registry.npmjs.org/tar-fs/-/tar-fs-3.0.8.tgz",
-			"integrity": "sha512-ZoROL70jptorGAlgAYiLoBLItEKw/fUxg9BSYK/dF/GAGYFJOJJJMvjPAKDJraCXFwadD456FCuvLWgfhMsPwg==",
-			"license": "MIT",
+			"version": "3.1.0",
+			"resolved": "https://registry.npmjs.org/tar-fs/-/tar-fs-3.1.0.tgz",
+			"integrity": "sha512-5Mty5y/sOF1YWj1J6GiBodjlDc05CUR8PKXrsnFAiSG0xA+GHeWLovaZPYUDXkH/1iKRf2+M5+OrRgzC7O9b7w==",
 			"dependencies": {
 				"pump": "^3.0.0",
 				"tar-stream": "^3.1.5"
@@ -18418,10 +18416,9 @@
 			"license": "MIT"
 		},
 		"node_modules/undici": {
-			"version": "6.21.1",
-			"resolved": "https://registry.npmjs.org/undici/-/undici-6.21.1.tgz",
-			"integrity": "sha512-q/1rj5D0/zayJB2FraXdaWxbhWiNKDvu8naDT2dl1yTlvJp4BLtOcp2a5BvgGNQpYYJzau7tf1WgKv3b+7mqpQ==",
-			"license": "MIT",
+			"version": "6.21.3",
+			"resolved": "https://registry.npmjs.org/undici/-/undici-6.21.3.tgz",
+			"integrity": "sha512-gBLkYIlEnSp8pFbT64yFgGE6UIB9tAkhukC23PmMDCe5Nd+cRqKxSjw5y54MK2AZMgZfJWMaNE4nYUHgi1XEOw==",
 			"engines": {
 				"node": ">=18.17"
 			}
@@ -25163,9 +25160,9 @@
 			"integrity": "sha512-AlcaJBi/pqqJBIQ8U9Mcpc9i8Aqxn88Skv5d+xBX006BY5u8N3mGLHa5Lgppa7L/HfwgwLgZ6NYs+Ag6uUmJRA=="
 		},
 		"brace-expansion": {
-			"version": "2.0.1",
-			"resolved": "https://registry.npmjs.org/brace-expansion/-/brace-expansion-2.0.1.tgz",
-			"integrity": "sha512-XnAIvQ8eM+kC6aULx6wuQiwVsnzsi9d3WxzV3FpWTGA19F621kwdbsAcFKXgKUHZWsy+mY6iL1sHTxWEFCytDA==",
+			"version": "2.0.2",
+			"resolved": "https://registry.npmjs.org/brace-expansion/-/brace-expansion-2.0.2.tgz",
+			"integrity": "sha512-Jt0vHyM+jmUBqojB7E1NIYadt0vI0Qxjxd2TErW94wDz+E2LAm5vKMXXwg6ZZBTHPuUlDgQHKXvjGBdfcF1ZDQ==",
 			"requires": {
 				"balanced-match": "^1.0.0"
 			}
@@ -31801,9 +31798,9 @@
 			}
 		},
 		"tar-fs": {
-			"version": "3.0.8",
-			"resolved": "https://registry.npmjs.org/tar-fs/-/tar-fs-3.0.8.tgz",
-			"integrity": "sha512-ZoROL70jptorGAlgAYiLoBLItEKw/fUxg9BSYK/dF/GAGYFJOJJJMvjPAKDJraCXFwadD456FCuvLWgfhMsPwg==",
+			"version": "3.1.0",
+			"resolved": "https://registry.npmjs.org/tar-fs/-/tar-fs-3.1.0.tgz",
+			"integrity": "sha512-5Mty5y/sOF1YWj1J6GiBodjlDc05CUR8PKXrsnFAiSG0xA+GHeWLovaZPYUDXkH/1iKRf2+M5+OrRgzC7O9b7w==",
 			"requires": {
 				"bare-fs": "^4.0.1",
 				"bare-path": "^3.0.0",
@@ -32226,9 +32223,9 @@
 			"integrity": "sha512-GMXzWtsc57XAtguZgaQViUOzs0KTkk8ojr3/xAxXLITqf/3EMwxC0inyETfDFjH/Krbhuep0HNbbjI9i/q3F3g=="
 		},
 		"undici": {
-			"version": "6.21.1",
-			"resolved": "https://registry.npmjs.org/undici/-/undici-6.21.1.tgz",
-			"integrity": "sha512-q/1rj5D0/zayJB2FraXdaWxbhWiNKDvu8naDT2dl1yTlvJp4BLtOcp2a5BvgGNQpYYJzau7tf1WgKv3b+7mqpQ=="
+			"version": "6.21.3",
+			"resolved": "https://registry.npmjs.org/undici/-/undici-6.21.3.tgz",
+			"integrity": "sha512-gBLkYIlEnSp8pFbT64yFgGE6UIB9tAkhukC23PmMDCe5Nd+cRqKxSjw5y54MK2AZMgZfJWMaNE4nYUHgi1XEOw=="
 		},
 		"undici-types": {
 			"version": "5.26.5",

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": "3.20.1",
+	"version": "3.20.11",
 	"icon": "assets/icons/icon.png",
 	"engines": {
 		"vscode": "^1.84.0"
@@ -119,7 +119,8 @@
 				{
 					"type": "webview",
 					"id": "claude-dev.SidebarProvider",
-					"name": ""
+					"name": "",
+					"icon": "assets/icons/icon.svg"
 				}
 			]
 		},
@@ -338,14 +339,14 @@
 	},
 	"scripts": {
 		"vscode:prepublish": "npm run package",
-		"compile": "npm run check-types && npm run lint && node esbuild.js",
-		"compile-standalone": "npm run check-types && npm run lint && node esbuild.js --standalone",
+		"compile": "npm run check-types && npm run lint && node esbuild.mjs",
+		"compile-standalone": "npm run check-types && npm run lint && node esbuild.mjs --standalone",
 		"postcompile-standalone": "node scripts/package-standalone.mjs",
 		"watch": "npm-run-all -p watch:*",
-		"watch:esbuild": "node esbuild.js --watch",
+		"watch:esbuild": "node esbuild.mjs --watch",
 		"watch:tsc": "tsc --noEmit --watch --project tsconfig.json",
-		"package": "npm run check-types && npm run build:webview && npm run lint && node esbuild.js --production",
-		"protos": "node scripts/build-proto.mjs && node scripts/generate-protobus-setup.mjs && node scripts/generate-host-bridge-client.mjs",
+		"package": "npm run check-types && npm run build:webview && npm run lint && node esbuild.mjs --production",
+		"protos": "node scripts/build-proto.mjs",
 		"postprotos": "prettier src/shared/proto src/core/controller src/hosts/ webview-ui/src/services src/generated --write --log-level warn",
 		"clean": "rimraf dist dist-standalone webview-ui/build src/generated out/",
 		"compile-tests": "node ./scripts/build-tests.js",

playwright.config.ts

@@ -1,17 +1,31 @@
 import { defineConfig } from "@playwright/test"
 
-const isGitHubAction = !!process.env.CI
+const isCI = !!process?.env?.CI
+const isWindow = process?.platform?.startsWith("win")
 
 export default defineConfig({
 	workers: 1,
 	retries: 1,
 	testDir: "src/test/e2e",
-	timeout: 20000,
+	timeout: isCI || isWindow ? 40000 : 20000,
 	expect: {
-		timeout: 20000,
+		timeout: isCI || isWindow ? 5000 : 2000,
 	},
 	fullyParallel: true,
-	reporter: isGitHubAction ? [["github"], ["list"]] : [["list"]],
-	globalSetup: require.resolve("./src/test/e2e/utils/setup"),
-	globalTeardown: require.resolve("./src/test/e2e/utils/teardown"),
+	reporter: isCI ? [["github"], ["list"]] : [["list"]],
+	projects: [
+		{
+			name: "setup test environment",
+			testMatch: /global\.setup\.ts/,
+		},
+		{
+			name: "e2e tests",
+			testMatch: /.*\.test\.ts/,
+			dependencies: ["setup test environment"],
+		},
+		{
+			name: "cleanup test environment",
+			testMatch: /global\.teardown\.ts/,
+		},
+	],
 })

proto/cline/account.proto

@@ -1,7 +1,7 @@
 syntax = "proto3";
 
 package cline;
-import "common.proto";
+import "cline/common.proto";
 option java_package = "bot.cline.proto";
 option java_multiple_files = true;
 
@@ -127,4 +127,4 @@ message OrganizationUsageTransaction {
   int32 prompt_tokens = 10;
   int32 total_tokens = 11;
   string user_id = 12;
-}
+}

proto/cline/browser.proto

@@ -1,7 +1,7 @@
 syntax = "proto3";
 
 package cline;
-import "common.proto";
+import "cline/common.proto";
 option java_package = "bot.cline.proto";
 option java_multiple_files = true;
 

proto/cline/checkpoints.proto

@@ -1,7 +1,7 @@
 syntax = "proto3";
 
 package cline;
-import "common.proto";
+import "cline/common.proto";
 option java_package = "bot.cline.proto";
 option java_multiple_files = true;
 

proto/cline/file.proto

@@ -1,7 +1,7 @@
 syntax = "proto3";
 
 package cline;
-import "common.proto";
+import "cline/common.proto";
 option java_package = "bot.cline.proto";
 option java_multiple_files = true;
 

proto/cline/mcp.proto

@@ -1,7 +1,7 @@
 syntax = "proto3";
 
 package cline;
-import "common.proto";
+import "cline/common.proto";
 option java_package = "bot.cline.proto";
 option java_multiple_files = true;
 

proto/cline/models.proto

@@ -1,7 +1,7 @@
 syntax = "proto3";
 
 package cline;
-import "common.proto";
+import "cline/common.proto";
 option java_package = "bot.cline.proto";
 option java_multiple_files = true;
 
@@ -27,6 +27,8 @@ service ModelsService {
   rpc updateApiConfigurationProto(UpdateApiConfigurationRequest) returns (Empty);
    // Refreshes and returns Groq models
   rpc refreshGroqModels(EmptyRequest) returns (OpenRouterCompatibleModelInfo);
+  // Refreshes and returns Baseten models
+  rpc refreshBasetenModels(EmptyRequest) returns (OpenRouterCompatibleModelInfo);
 }
 
 // List of VS Code LM models
@@ -129,6 +131,8 @@ enum ApiProvider {
   CLAUDE_CODE = 26;
   MOONSHOT = 27;
   HUGGINGFACE = 28;
+  HUAWEI_CLOUD_MAAS = 29;
+  BASETEN = 30;
 }
 
 // Model info for OpenAI-compatible models
@@ -229,6 +233,8 @@ message ModelsApiConfiguration {
   optional string cline_account_id = 58;
   optional string groq_api_key = 59;
   optional string hugging_face_api_key = 60;
+  optional string huawei_cloud_maas_api_key = 61;
+  optional string baseten_api_key = 62;
 
   // Plan mode configurations
   optional ApiProvider plan_mode_api_provider = 100;
@@ -255,6 +261,10 @@ message ModelsApiConfiguration {
   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;
 
   // Act mode configurations
   optional ApiProvider act_mode_api_provider = 200;
@@ -281,6 +291,10 @@ message ModelsApiConfiguration {
   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;
 
   repeated string favorited_model_ids = 300;
 }

proto/cline/slash.proto

@@ -1,7 +1,7 @@
 syntax = "proto3";
 
 package cline;
-import "common.proto";
+import "cline/common.proto";
 option java_package = "bot.cline.proto";
 option java_multiple_files = true;
 

proto/cline/state.proto

@@ -1,19 +1,19 @@
 syntax = "proto3";
 package cline;
-import "common.proto";
+import "cline/common.proto";
 option java_package = "bot.cline.proto";
 option java_multiple_files = true;
 
 service StateService {
   rpc getLatestState(EmptyRequest) returns (State);
-  rpc updateTerminalConnectionTimeout(Int64Request) returns (Int64);
+  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);
   rpc resetState(ResetStateRequest) returns (Empty);
-  rpc togglePlanActMode(TogglePlanActModeRequest) returns (Boolean);
+  rpc togglePlanActModeProto(TogglePlanActModeRequest) returns (Boolean);
   rpc updateAutoApprovalSettings(AutoApprovalSettingsRequest) returns (Empty);
   rpc updateSettings(UpdateSettingsRequest) returns (Empty);
   rpc updateTelemetrySetting(TelemetrySettingRequest) returns (Empty);
@@ -43,7 +43,7 @@ message TerminalProfileUpdateResponse {
 
 message TogglePlanActModeRequest {
   Metadata metadata = 1;
-  ChatSettings chat_settings = 2;
+  PlanActMode mode = 2;
   optional ChatContent chat_content = 3;
 }
 
@@ -52,12 +52,6 @@ enum PlanActMode {
   ACT = 1;
 }
 
-message ChatSettings {
-  PlanActMode mode = 1;
-  optional string preferred_language = 2;
-  optional string open_ai_reasoning_effort = 3;
-}
-
 message ChatContent {
   optional string message = 1;
   repeated string images = 2;
@@ -108,12 +102,15 @@ message UpdateSettingsRequest {
   optional bool plan_act_separate_models_setting = 4;
   optional bool enable_checkpoints_setting = 5;
   optional bool mcp_marketplace_enabled = 6;
-  optional ChatSettings chat_settings = 7;
-  optional int64 shell_integration_timeout = 8;
+  optional int32 shell_integration_timeout = 8;
   optional bool terminal_reuse_enabled = 9;
   optional bool mcp_responses_collapsed = 10;
   optional string mcp_display_mode = 11;
-  optional int64 terminal_output_line_limit = 12;
+  optional int32 terminal_output_line_limit = 12;
+  optional PlanActMode mode = 13;
+  optional string preferred_language = 14;
+  optional string openai_reasoning_effort = 15;
+  optional bool strict_plan_mode_enabled = 16;
 }
 
 // Complete API Configuration message
@@ -153,8 +150,8 @@ message ApiConfiguration {
   optional string requesty_api_key = 32;
   optional string together_api_key = 33;
   optional string fireworks_api_key = 34;
-  optional int64 fireworks_model_max_completion_tokens = 35;
-  optional int64 fireworks_model_max_tokens = 36;
+  optional int32 fireworks_model_max_completion_tokens = 35;
+  optional int32 fireworks_model_max_tokens = 36;
   optional string qwen_api_key = 37;
   optional string doubao_api_key = 38;
   optional string mistral_api_key = 39;
@@ -166,7 +163,7 @@ message ApiConfiguration {
   optional string xai_api_key = 45;
   optional string sambanova_api_key = 46;
   optional string cerebras_api_key = 47;
-  optional int64 request_timeout_ms = 48;
+  optional int32 request_timeout_ms = 48;
   optional string sap_ai_core_client_id = 49;
   optional string sap_ai_core_client_secret = 50;
   optional string sap_ai_resource_group = 51;
@@ -174,11 +171,12 @@ message ApiConfiguration {
   optional string sap_ai_core_base_url = 53;
   optional string moonshot_api_key = 54;
   optional string moonshot_api_line = 55;
+  optional string huawei_cloud_maas_api_key = 56;
   
   // Plan mode configurations
   optional string plan_mode_api_provider = 100;
   optional string plan_mode_api_model_id = 101;
-  optional int64 plan_mode_thinking_budget_tokens = 102;
+  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 bool plan_mode_aws_bedrock_custom_selected = 105;
@@ -196,11 +194,13 @@ message ApiConfiguration {
   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;
   
   // Act mode configurations
   optional string act_mode_api_provider = 200;
   optional string act_mode_api_model_id = 201;
-  optional int64 act_mode_thinking_budget_tokens = 202;
+  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 bool act_mode_aws_bedrock_custom_selected = 205;
@@ -218,6 +218,8 @@ message ApiConfiguration {
   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;
   
   // Favorited model IDs
   repeated string favorited_model_ids = 300;
@@ -228,3 +230,11 @@ message ApiConfiguration {
 
   optional string cline_account_id = 303;
 }
+
+message UpdateTerminalConnectionTimeoutRequest {
+  optional int32 timeout_ms = 1;
+}
+
+message UpdateTerminalConnectionTimeoutResponse {
+  optional int32 timeout_ms = 1;
+}

proto/cline/task.proto

@@ -1,7 +1,7 @@
 syntax = "proto3";
 
 package cline;
-import "common.proto";
+import "cline/common.proto";
 option java_package = "bot.cline.proto";
 option java_multiple_files = true;
 

proto/cline/ui.proto

@@ -1,7 +1,7 @@
 syntax = "proto3";
 
 package cline;
-import "common.proto";
+import "cline/common.proto";
 option java_package = "bot.cline.proto";
 option java_multiple_files = true;
 

proto/cline/web.proto

@@ -1,7 +1,7 @@
 syntax = "proto3";
 
 package cline;
-import "common.proto";
+import "cline/common.proto";
 option java_package = "bot.cline.proto";
 option java_multiple_files = true;
 

proto/host/diff.proto

@@ -4,7 +4,7 @@ package host;
 option java_package = "bot.cline.host.proto";
 option java_multiple_files = true;
 
-import "common.proto";
+import "cline/common.proto";
 
 // Provides methods for diff views.
 service DiffService {

proto/host/env.proto

@@ -4,7 +4,7 @@ package host;
 option java_package = "bot.cline.host.proto";
 option java_multiple_files = true;
 
-import "common.proto";
+import "cline/common.proto";
 
 // Provides methods for working with the user's environment.
 service EnvService {
@@ -13,7 +13,4 @@ service EnvService {
 
   // Reads text from the system clipboard.
   rpc clipboardReadText(cline.EmptyRequest) returns (cline.String);
-
-  // Opens a URL in the user's default browser or application.
-  rpc openExternal(cline.StringRequest) returns (cline.Empty);
 }

proto/host/watch.proto

@@ -4,7 +4,7 @@ package host;
 option java_package = "bot.cline.host.proto";
 option java_multiple_files = true;
 
-import "common.proto";
+import "cline/common.proto";
 
 /** 
  * The watch service is only here as example of a streaming rpc in the host bridge. 

proto/host/window.proto

@@ -4,7 +4,7 @@ package host;
 option java_package = "bot.cline.host.proto";
 option java_multiple_files = true;
 
-import "common.proto";
+import "cline/common.proto";
 
 // Provides methods for working with IDE windows and editors.
 service WindowService {
@@ -14,6 +14,7 @@ service WindowService {
   rpc showMessage(ShowMessageRequest) returns (SelectedResponse);
   rpc showInputBox(ShowInputBoxRequest) returns (ShowInputBoxResponse);
   rpc showSaveDialog(ShowSaveDialogRequest) returns (ShowSaveDialogResponse);
+  rpc openFile(OpenFileRequest) returns (OpenFileResponse);
   rpc getOpenTabs(GetOpenTabsRequest) returns (GetOpenTabsResponse);
   rpc getVisibleTabs(GetVisibleTabsRequest) returns (GetVisibleTabsResponse);
 }
@@ -83,6 +84,8 @@ message ShowSaveDialogRequest {
 
 message ShowSaveDialogOptions {
   optional string default_path = 1;
+  // A map of file types to extensions, e.g
+  // "Text Files": { "extensions": ["txt", "md"] }
   map<string, FileExtensionList> filters = 2;
 }
 
@@ -91,6 +94,7 @@ message FileExtensionList {
 }
 
 message ShowSaveDialogResponse {
+  // If the user cancelled the dialog, this will be empty.
   optional string selected_path = 1;
 }
 
@@ -105,6 +109,15 @@ message ShowInputBoxResponse {
   optional string response = 1;
 }
 
+message OpenFileRequest {
+  cline.Metadata metadata = 1;
+  string file_path = 2;
+}
+
+message OpenFileResponse {
+  // empty
+}
+
 message GetOpenTabsRequest {
   // empty
 }
@@ -119,4 +132,4 @@ message GetVisibleTabsRequest {
 
 message GetVisibleTabsResponse {
   repeated string paths = 1;
-}
+}

proto/host/workspace.proto

@@ -4,14 +4,14 @@ package host;
 option java_package = "bot.cline.host.proto";
 option java_multiple_files = true;
 
-import "common.proto";
-
 // Provides methods for working with workspaces/projects.
 service WorkspaceService {
   // Returns a list of the top level directories of the workspace.
   rpc getWorkspacePaths(GetWorkspacePathsRequest) returns (GetWorkspacePathsResponse);
-  // Saves an open document if it's dirty
-  rpc saveOpenDocumentIfDirty(SaveOpenDocumentIfDirtyRequest) returns (cline.Empty);
+  // Saves an open document if it's open in the editor and has unsaved changes. 
+  // Returns true if the document was saved, returns false if the document was not found, or did not
+  // need to be saved.
+  rpc saveOpenDocumentIfDirty(SaveOpenDocumentIfDirtyRequest) returns (SaveOpenDocumentIfDirtyResponse);
 }
 
 message GetWorkspacePathsRequest {
@@ -28,6 +28,9 @@ message GetWorkspacePathsResponse {
 }
 
 message SaveOpenDocumentIfDirtyRequest {
-  cline.Metadata metadata = 1;
-  string file_path = 2;
+  optional string file_path = 2;
+}
+message SaveOpenDocumentIfDirtyResponse {
+  // Returns true if the document was saved. 
+  optional bool was_saved = 1;
 }

scripts/build-proto.mjs

@@ -7,8 +7,10 @@ import { globby } from "globby"
 import { createRequire } from "module"
 import os from "os"
 import * as path from "path"
-import { fileURLToPath } from "url"
 import { rmrf } from "./file-utils.mjs"
+import { main as generateProtoBusSetup } from "./generate-protobus-setup.mjs"
+import { main as generateHostBridgeClient } from "./generate-host-bridge-client.mjs"
+import { loadProtoDescriptorSet } from "./proto-utils.mjs"
 
 const require = createRequire(import.meta.url)
 const PROTOC = path.join(require.resolve("grpc-tools"), "../bin/protoc")
@@ -34,9 +36,14 @@ const TS_PROTO_OPTIONS = [
 ]
 
 async function main() {
-	console.log(chalk.bold.blue("Compiling Protocol Buffers..."))
-
 	await cleanup()
+	await compileProtos()
+	await checkProtos()
+	await generateProtoBusSetup()
+	await generateHostBridgeClient()
+}
+async function compileProtos() {
+	console.log(chalk.bold.blue("Compiling Protocol Buffers..."))
 
 	// Check for Apple Silicon compatibility before proceeding
 	checkAppleSiliconCompatibility()
@@ -180,6 +187,51 @@ 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/build-tests.js

@@ -1,3 +1,4 @@
+#!/usr/bin/env node
 const { execSync } = require("child_process")
 const esbuild = require("esbuild")
 

scripts/generate-host-bridge-client.mjs

@@ -15,7 +15,7 @@ const VSCODE_CLIENT_FILE = path.resolve("src/generated/hosts/vscode/hostbridge-g
 /**
  * Main function to generate the host bridge client
  */
-async function main() {
+export async function main() {
 	const { hostServices } = await loadServicesFromProtoDescriptor()
 
 	await generateTypesFile(hostServices)
@@ -234,8 +234,10 @@ const ${name}ServiceRegistry = createServiceRegistry("${name}")
 ${methods}`
 }
 
-// Run the main function
-main().catch((error) => {
-	console.error(chalk.red("Error:"), error)
-	process.exit(1)
-})
+// Only run main if this script is executed directly
+if (import.meta.url === `file://${process.argv[1]}`) {
+	main().catch((error) => {
+		console.error(chalk.red("Error:"), error)
+		process.exit(1)
+	})
+}

scripts/generate-protobus-setup.mjs

@@ -12,7 +12,7 @@ const STANDALONE_SERVER_SETUP_FILE = path.resolve("src/generated/hosts/standalon
 
 const SCRIPT_NAME = path.relative(process.cwd(), fileURLToPath(import.meta.url))
 
-async function main() {
+export async function main() {
 	const { protobusServices } = await loadServicesFromProtoDescriptor()
 	await generateWebviewProtobusClients(protobusServices)
 	await generateVscodeServiceTypes(protobusServices)
@@ -40,11 +40,11 @@ async function generateWebviewProtobusClients(protobusServices) {
 			}
 			if (!rpc.responseStream) {
 				rpcs.push(`    static async ${rpcName}(request: ${requestType}): Promise<${responseType}> {
-		return this.makeRequest("${rpcName}", request)
+		return this.makeUnaryRequest("${rpcName}", request, ${requestType}.toJSON, ${responseType}.fromJSON)
 	}`)
 			} else {
 				rpcs.push(`    static ${rpcName}(request: ${requestType}, callbacks: Callbacks<${responseType}>): ()=>void {
-		return this.makeStreamingRequest("${rpcName}", request, callbacks)
+		return this.makeStreamingRequest("${rpcName}", request, ${requestType}.toJSON, ${responseType}.fromJSON, callbacks)
 	}`)
 			}
 		}
@@ -205,4 +205,10 @@ function getDirName(serviceName) {
 	return domain.charAt(0).toLowerCase() + domain.slice(1)
 }
 
-main()
+// Only run main if this script is executed directly
+if (import.meta.url === `file://${process.argv[1]}`) {
+	main().catch((error) => {
+		console.error(chalk.red("Error:"), error)
+		process.exit(1)
+	})
+}

scripts/proto-utils.mjs

@@ -10,7 +10,7 @@ const DESCRIPTOR_SET = path.resolve("dist-standalone/proto/descriptor_set.pb")
 const typeNameToFQN = new Map()
 
 function addTypeNameToFqn(name, fqn) {
-	if (typeNameToFQN.has(name)) {
+	if (typeNameToFQN.has(name) && typeNameToFQN.get(name) !== fqn) {
 		throw new Error(`Proto type ${name} redefined (${fqn}).`)
 	}
 	typeNameToFQN.set(name, fqn)
@@ -23,11 +23,15 @@ export function getFqn(name) {
 	return typeNameToFQN.get(name)
 }
 
-export async function loadServicesFromProtoDescriptor() {
-	// Load service definitions from descriptor set
+export async function loadProtoDescriptorSet() {
 	const descriptorBuffer = await fs.readFile(DESCRIPTOR_SET)
 	const packageDefinition = protoLoader.loadFileDescriptorSetFromBuffer(descriptorBuffer)
-	const proto = grpc.loadPackageDefinition(packageDefinition)
+	return grpc.loadPackageDefinition(packageDefinition)
+}
+
+export async function loadServicesFromProtoDescriptor() {
+	// Load service definitions from descriptor set
+	const proto = await loadProtoDescriptorSet()
 
 	// Extract host services and proto messages from the proto definition
 	const hostServices = {}

src/api/index.ts

@@ -29,8 +29,10 @@ import { SapAiCoreHandler } from "./providers/sapaicore"
 import { ClaudeCodeHandler } from "./providers/claude-code"
 import { MoonshotHandler } from "./providers/moonshot"
 import { GroqHandler } from "./providers/groq"
-import { Mode } from "../shared/ChatSettings"
+import { Mode } from "@shared/storage/types"
 import { HuggingFaceHandler } from "./providers/huggingface"
+import { HuaweiCloudMaaSHandler } from "./providers/huawei-cloud-maas"
+import { BasetenHandler } from "./providers/baseten"
 
 export interface ApiHandler {
 	createMessage(systemPrompt: string, messages: Anthropic.Messages.MessageParam[]): ApiStream
@@ -256,6 +258,13 @@ function createHandlerForProvider(
 				groqModelInfo: mode === "plan" ? options.planModeGroqModelInfo : options.actModeGroqModelInfo,
 				apiModelId: mode === "plan" ? options.planModeApiModelId : options.actModeApiModelId,
 			})
+		case "baseten":
+			return new BasetenHandler({
+				basetenApiKey: options.basetenApiKey,
+				basetenModelId: mode === "plan" ? options.planModeBasetenModelId : options.actModeBasetenModelId,
+				basetenModelInfo: mode === "plan" ? options.planModeBasetenModelInfo : options.actModeBasetenModelInfo,
+				apiModelId: mode === "plan" ? options.planModeApiModelId : options.actModeApiModelId,
+			})
 		case "sapaicore":
 			return new SapAiCoreHandler({
 				sapAiCoreClientId: options.sapAiCoreClientId,
@@ -272,6 +281,14 @@ function createHandlerForProvider(
 				thinkingBudgetTokens:
 					mode === "plan" ? options.planModeThinkingBudgetTokens : options.actModeThinkingBudgetTokens,
 			})
+		case "huawei-cloud-maas":
+			return new HuaweiCloudMaaSHandler({
+				huaweiCloudMaasApiKey: options.huaweiCloudMaasApiKey,
+				huaweiCloudMaasModelId:
+					mode === "plan" ? options.planModeHuaweiCloudMaasModelId : options.actModeHuaweiCloudMaasModelId,
+				huaweiCloudMaasModelInfo:
+					mode === "plan" ? options.planModeHuaweiCloudMaasModelInfo : options.actModeHuaweiCloudMaasModelInfo,
+			})
 		default:
 			return new AnthropicHandler({
 				apiKey: options.apiKey,

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

@@ -612,101 +612,102 @@ describe("AwsBedrockHandler", () => {
 		})
 	})
 
-	describe("getModelId", () => {
-		it("should return raw model ID for custom models", async () => {
-			const customOptions: ApiHandlerOptions = {
-				...mockOptions,
-				actModeAwsBedrockCustomSelected: true,
-				actModeApiModelId:
-					"arn:aws:bedrock:us-west-2:123456789012:custom-model/anthropic.claude-3-5-sonnet-20241022-v2:0/Qk8MMyLmRd",
-			}
-			const customHandler = new AwsBedrockHandler(customOptions)
+	// TODO: Re-enable or remove these tests.
+	// describe("getModelId", () => {
+	// 	it("should return raw model ID for custom models", async () => {
+	// 		const customOptions: ApiHandlerOptions = {
+	// 			...mockOptions,
+	// 			actModeAwsBedrockCustomSelected: true,
+	// 			actModeApiModelId:
+	// 				"arn:aws:bedrock:us-west-2:123456789012:custom-model/anthropic.claude-3-5-sonnet-20241022-v2:0/Qk8MMyLmRd",
+	// 		}
+	// 		const customHandler = new AwsBedrockHandler(customOptions)
 
-			const modelId = await customHandler.getModelId()
-			modelId.should.equal(
-				"arn:aws:bedrock:us-west-2:123456789012:custom-model/anthropic.claude-3-5-sonnet-20241022-v2:0/Qk8MMyLmRd",
-			)
-		})
+	// 		const modelId = await customHandler.getModelId()
+	// 		modelId.should.equal(
+	// 			"arn:aws:bedrock:us-west-2:123456789012:custom-model/anthropic.claude-3-5-sonnet-20241022-v2:0/Qk8MMyLmRd",
+	// 		)
+	// 	})
 
-		it("should not encode custom model IDs with slashes", async () => {
-			const customOptions: ApiHandlerOptions = {
-				...mockOptions,
-				actModeAwsBedrockCustomSelected: true,
-				actModeApiModelId: "my-namespace/my-custom-model",
-			}
-			const customHandler = new AwsBedrockHandler(customOptions)
+	// 	it("should not encode custom model IDs with slashes", async () => {
+	// 		const customOptions: ApiHandlerOptions = {
+	// 			...mockOptions,
+	// 			actModeAwsBedrockCustomSelected: true,
+	// 			actModeApiModelId: "my-namespace/my-custom-model",
+	// 		}
+	// 		const customHandler = new AwsBedrockHandler(customOptions)
 
-			const modelId = await customHandler.getModelId()
-			modelId.should.equal("my-namespace/my-custom-model")
-			modelId.should.not.match(/%2F/)
-		})
+	// 		const modelId = await customHandler.getModelId()
+	// 		modelId.should.equal("my-namespace/my-custom-model")
+	// 		modelId.should.not.match(/%2F/)
+	// 	})
 
-		it("should apply cross-region prefix for non-custom models when enabled", async () => {
-			const crossRegionOptions: ApiHandlerOptions = {
-				...mockOptions,
-				awsUseCrossRegionInference: true,
-				awsRegion: "us-west-2",
-			}
-			const crossRegionHandler = new AwsBedrockHandler(crossRegionOptions)
+	// 	it("should apply cross-region prefix for non-custom models when enabled", async () => {
+	// 		const crossRegionOptions: ApiHandlerOptions = {
+	// 			...mockOptions,
+	// 			awsUseCrossRegionInference: true,
+	// 			awsRegion: "us-west-2",
+	// 		}
+	// 		const crossRegionHandler = new AwsBedrockHandler(crossRegionOptions)
 
-			const modelId = await crossRegionHandler.getModelId()
-			modelId.should.equal("us.anthropic.claude-3-7-sonnet-20250219-v1:0")
-		})
+	// 		const modelId = await crossRegionHandler.getModelId()
+	// 		modelId.should.equal("us.anthropic.claude-3-7-sonnet-20250219-v1:0")
+	// 	})
 
-		it("should apply EU cross-region prefix", async () => {
-			const euOptions: ApiHandlerOptions = {
-				...mockOptions,
-				awsUseCrossRegionInference: true,
-				awsRegion: "eu-central-1",
-			}
-			const euHandler = new AwsBedrockHandler(euOptions)
+	// 	it("should apply EU cross-region prefix", async () => {
+	// 		const euOptions: ApiHandlerOptions = {
+	// 			...mockOptions,
+	// 			awsUseCrossRegionInference: true,
+	// 			awsRegion: "eu-central-1",
+	// 		}
+	// 		const euHandler = new AwsBedrockHandler(euOptions)
 
-			const modelId = await euHandler.getModelId()
-			modelId.should.equal("eu.anthropic.claude-3-7-sonnet-20250219-v1:0")
-		})
+	// 		const modelId = await euHandler.getModelId()
+	// 		modelId.should.equal("eu.anthropic.claude-3-7-sonnet-20250219-v1:0")
+	// 	})
 
-		it("should apply APAC cross-region prefix", async () => {
-			const apacOptions: ApiHandlerOptions = {
-				...mockOptions,
-				awsUseCrossRegionInference: true,
-				awsRegion: "ap-northeast-1",
-			}
-			const apacHandler = new AwsBedrockHandler(apacOptions)
+	// 	it("should apply APAC cross-region prefix", async () => {
+	// 		const apacOptions: ApiHandlerOptions = {
+	// 			...mockOptions,
+	// 			awsUseCrossRegionInference: true,
+	// 			awsRegion: "ap-northeast-1",
+	// 		}
+	// 		const apacHandler = new AwsBedrockHandler(apacOptions)
 
-			const modelId = await apacHandler.getModelId()
-			modelId.should.equal("apac.anthropic.claude-3-7-sonnet-20250219-v1:0")
-		})
+	// 		const modelId = await apacHandler.getModelId()
+	// 		modelId.should.equal("apac.anthropic.claude-3-7-sonnet-20250219-v1:0")
+	// 	})
 
-		it("should not apply cross-region prefix for custom models even when enabled", async () => {
-			const customCrossRegionOptions: ApiHandlerOptions = {
-				...mockOptions,
-				actModeAwsBedrockCustomSelected: true,
-				actModeApiModelId: "arn:aws:bedrock:us-west-2:123456789012:custom-model/my-model",
-				awsUseCrossRegionInference: true,
-			}
-			const customCrossRegionHandler = new AwsBedrockHandler(customCrossRegionOptions)
+	// 	it("should not apply cross-region prefix for custom models even when enabled", async () => {
+	// 		const customCrossRegionOptions: ApiHandlerOptions = {
+	// 			...mockOptions,
+	// 			actModeAwsBedrockCustomSelected: true,
+	// 			actModeApiModelId: "arn:aws:bedrock:us-west-2:123456789012:custom-model/my-model",
+	// 			awsUseCrossRegionInference: true,
+	// 		}
+	// 		const customCrossRegionHandler = new AwsBedrockHandler(customCrossRegionOptions)
 
-			const modelId = await customCrossRegionHandler.getModelId()
-			modelId.should.equal("arn:aws:bedrock:us-west-2:123456789012:custom-model/my-model")
-		})
+	// 		const modelId = await customCrossRegionHandler.getModelId()
+	// 		modelId.should.equal("arn:aws:bedrock:us-west-2:123456789012:custom-model/my-model")
+	// 	})
 
-		it("should handle UltraThink model ARN correctly", async () => {
-			const ultraThinkOptions: ApiHandlerOptions = {
-				...mockOptions,
-				actModeAwsBedrockCustomSelected: true,
-				actModeApiModelId:
-					"arn:aws:bedrock:us-west-2:123456789012:custom-model/anthropic.claude-3-5-sonnet-20241022-v2:0/Qk8MMyLmRd",
-				actModeAwsBedrockCustomModelBaseId: "anthropic.claude-3-5-sonnet-20241022-v2:0",
-			}
-			const ultraThinkHandler = new AwsBedrockHandler(ultraThinkOptions)
+	// 	it("should handle UltraThink model ARN correctly", async () => {
+	// 		const ultraThinkOptions: ApiHandlerOptions = {
+	// 			...mockOptions,
+	// 			actModeAwsBedrockCustomSelected: true,
+	// 			actModeApiModelId:
+	// 				"arn:aws:bedrock:us-west-2:123456789012:custom-model/anthropic.claude-3-5-sonnet-20241022-v2:0/Qk8MMyLmRd",
+	// 			actModeAwsBedrockCustomModelBaseId: "anthropic.claude-3-5-sonnet-20241022-v2:0",
+	// 		}
+	// 		const ultraThinkHandler = new AwsBedrockHandler(ultraThinkOptions)
 
-			const modelId = await ultraThinkHandler.getModelId()
-			// Should return the raw ARN without any encoding
-			modelId.should.equal(
-				"arn:aws:bedrock:us-west-2:123456789012:custom-model/anthropic.claude-3-5-sonnet-20241022-v2:0/Qk8MMyLmRd",
-			)
-			modelId.should.not.match(/%2F/)
-			modelId.should.not.match(/%3A/)
-		})
-	})
+	// 		const modelId = await ultraThinkHandler.getModelId()
+	// 		// Should return the raw ARN without any encoding
+	// 		modelId.should.equal(
+	// 			"arn:aws:bedrock:us-west-2:123456789012:custom-model/anthropic.claude-3-5-sonnet-20241022-v2:0/Qk8MMyLmRd",
+	// 		)
+	// 		modelId.should.not.match(/%2F/)
+	// 		modelId.should.not.match(/%3A/)
+	// 	})
+	// })
 })

src/api/providers/anthropic.ts

@@ -55,6 +55,7 @@ export class AnthropicHandler implements ApiHandler {
 			case "claude-3-5-sonnet-20241022":
 			case "claude-3-5-haiku-20241022":
 			case "claude-opus-4-20250514":
+			case "claude-opus-4-1-20250805":
 			case "claude-3-opus-20240229":
 			case "claude-3-haiku-20240307": {
 				/*
@@ -122,6 +123,7 @@ export class AnthropicHandler implements ApiHandler {
 						switch (modelId) {
 							case "claude-sonnet-4-20250514":
 							case "claude-opus-4-20250514":
+							case "claude-opus-4-1-20250805":
 							case "claude-3-7-sonnet-20250219":
 							case "claude-3-5-sonnet-20241022":
 							case "claude-3-5-haiku-20241022":

src/api/providers/baseten.ts

@@ -0,0 +1,165 @@
+import { Anthropic } from "@anthropic-ai/sdk"
+import OpenAI from "openai"
+import { withRetry } from "../retry"
+import { ApiHandler } from "../"
+import { BasetenModelId, ModelInfo, basetenDefaultModelId, basetenModels } from "@shared/api"
+import { calculateApiCostOpenAI } from "../../utils/cost"
+import { convertToOpenAiMessages } from "../transform/openai-format"
+import { ApiStream } from "../transform/stream"
+
+interface BasetenHandlerOptions {
+	basetenApiKey?: string
+	basetenModelId?: string
+	basetenModelInfo?: ModelInfo
+	apiModelId?: string // For backward compatibility
+}
+
+export class BasetenHandler implements ApiHandler {
+	private options: BasetenHandlerOptions
+	private client: OpenAI | undefined
+
+	constructor(options: BasetenHandlerOptions) {
+		this.options = options
+	}
+
+	private ensureClient(): OpenAI {
+		if (!this.client) {
+			if (!this.options.basetenApiKey) {
+				throw new Error("Baseten API key is required")
+			}
+			try {
+				this.client = new OpenAI({
+					baseURL: "https://inference.baseten.co/v1",
+					apiKey: this.options.basetenApiKey,
+				})
+			} catch (error) {
+				throw new Error(`Error creating Baseten client: ${error.message}`)
+			}
+		}
+		return this.client
+	}
+
+	/**
+	 * Gets the optimal max_tokens based on model capabilities
+	 */
+	private getOptimalMaxTokens(model: { id: BasetenModelId; info: ModelInfo }): number {
+		// Use model-specific max tokens if available
+		if (model.info.maxTokens && model.info.maxTokens > 0) {
+			return model.info.maxTokens
+		}
+
+		// Default fallback
+		return 8192
+	}
+
+	getModel(): { id: BasetenModelId; info: ModelInfo } {
+		// First priority: basetenModelId and basetenModelInfo
+		const basetenModelId = this.options.basetenModelId
+		const basetenModelInfo = this.options.basetenModelInfo
+		if (basetenModelId && basetenModelInfo) {
+			return { id: basetenModelId as BasetenModelId, info: basetenModelInfo }
+		}
+
+		// Second priority: basetenModelId with static model info
+		if (basetenModelId && basetenModelId in basetenModels) {
+			const id = basetenModelId as BasetenModelId
+			return { id, info: basetenModels[id] }
+		}
+
+		// Third priority: apiModelId (for backward compatibility)
+		const apiModelId = this.options.apiModelId
+		if (apiModelId && apiModelId in basetenModels) {
+			const id = apiModelId as BasetenModelId
+			return { id, info: basetenModels[id] }
+		}
+
+		// Default fallback
+		return {
+			id: basetenDefaultModelId,
+			info: basetenModels[basetenDefaultModelId],
+		}
+	}
+
+	private async *yieldUsage(modelInfo: ModelInfo, usage: any): ApiStream {
+		if (usage.prompt_tokens || usage.completion_tokens) {
+			const cost = calculateApiCostOpenAI(modelInfo, usage.prompt_tokens || 0, usage.completion_tokens || 0)
+
+			yield {
+				type: "usage",
+				inputTokens: usage.prompt_tokens || 0,
+				outputTokens: usage.completion_tokens || 0,
+				cacheWriteTokens: 0,
+				cacheReadTokens: 0,
+				totalCost: cost,
+			}
+		}
+	}
+
+	@withRetry()
+	async *createMessage(systemPrompt: string, messages: Anthropic.Messages.MessageParam[]): ApiStream {
+		const client = this.ensureClient()
+		const model = this.getModel()
+		const maxTokens = this.getOptimalMaxTokens(model)
+
+		const openAiMessages: OpenAI.Chat.ChatCompletionMessageParam[] = [
+			{ role: "system", content: systemPrompt },
+			...convertToOpenAiMessages(messages),
+		]
+
+		const stream = await client.chat.completions.create({
+			model: model.id,
+			max_tokens: maxTokens,
+			messages: openAiMessages,
+			stream: true,
+			stream_options: { include_usage: true },
+			temperature: 0,
+		})
+
+		let didOutputUsage = false
+
+		for await (const chunk of stream) {
+			const delta = chunk.choices[0]?.delta
+
+			// Handle reasoning field if present (for reasoning models with parsed output)
+			if ((delta as any)?.reasoning) {
+				const reasoningContent = (delta as any).reasoning as string
+				yield {
+					type: "reasoning",
+					reasoning: reasoningContent,
+				}
+				continue
+			}
+
+			// Handle content field
+			if (delta?.content) {
+				yield {
+					type: "text",
+					text: delta.content,
+				}
+			}
+
+			// Handle usage information - only output once
+			if (!didOutputUsage && chunk.usage) {
+				yield* this.yieldUsage(model.info, chunk.usage)
+				didOutputUsage = true
+			}
+		}
+	}
+
+	/**
+	 * 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
+	}
+}

src/api/providers/cerebras.ts

@@ -102,6 +102,7 @@ export class CerebrasHandler implements ApiHandler {
 				messages: cerebrasMessages,
 				temperature: 0,
 				stream: true,
+				max_tokens: this.getModel().info.maxTokens,
 			})
 
 			// Handle streaming response
@@ -175,9 +176,15 @@ export class CerebrasHandler implements ApiHandler {
 	}
 
 	getModel(): { id: string; info: ModelInfo } {
-		const modelId = this.options.apiModelId
-		if (modelId && modelId in cerebrasModels) {
-			const id = modelId as CerebrasModelId
+		const originalModelId = this.options.apiModelId
+		let apiModelId = originalModelId
+		if (originalModelId === "qwen-3-coder-480b-free") {
+			apiModelId = "qwen-3-coder-480b"
+			return { id: apiModelId, info: cerebrasModels[originalModelId as CerebrasModelId] }
+		}
+
+		if (originalModelId && originalModelId in cerebrasModels) {
+			const id = originalModelId as CerebrasModelId
 			return { id, info: cerebrasModels[id] }
 		}
 		return {

src/api/providers/gemini-mock.test.ts

@@ -0,0 +1,53 @@
+// Mock for @google/genai module to avoid ESM compatibility issues in tests
+
+export class GoogleGenAI {
+	constructor(options: any) {
+		// Mock constructor
+	}
+
+	models = {
+		generateContentStream: async (params: any) => {
+			// Mock implementation that returns an async iterator
+			return {
+				async *[Symbol.asyncIterator]() {
+					yield {
+						text: "Mock response",
+						candidates: [],
+						usageMetadata: {
+							promptTokenCount: 100,
+							candidatesTokenCount: 50,
+							thoughtsTokenCount: 0,
+							cachedContentTokenCount: 0,
+						},
+					}
+				},
+			}
+		},
+		countTokens: async (params: any) => {
+			// Mock token counting
+			return {
+				totalTokens: 100,
+			}
+		},
+	}
+}
+
+// Export mock types
+export interface GenerateContentConfig {
+	httpOptions?: any
+	systemInstruction?: string
+	temperature?: number
+	thinkingConfig?: any
+}
+
+export interface GenerateContentResponseUsageMetadata {
+	promptTokenCount?: number
+	candidatesTokenCount?: number
+	thoughtsTokenCount?: number
+	cachedContentTokenCount?: number
+}
+
+export interface Part {
+	thought?: boolean
+	text?: string
+}

src/api/providers/gemini.ts

@@ -7,7 +7,7 @@ import { ApiHandler } from "../"
 import { ApiHandlerOptions, geminiDefaultModelId, GeminiModelId, geminiModels, ModelInfo } from "@shared/api"
 import { convertAnthropicMessageToGemini } from "../transform/gemini-format"
 import { ApiStream } from "../transform/stream"
-import { telemetryService } from "@services/posthog/telemetry/TelemetryService"
+import { telemetryService } from "@services/posthog/PostHogClientProvider"
 
 // Define a default TTL for the cache (e.g., 15 minutes in seconds)
 const DEFAULT_CACHE_TTL_SECONDS = 900

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

@@ -0,0 +1,132 @@
+import { ApiHandler } from ".."
+import { huaweiCloudMaasDefaultModelId, HuaweiCloudMaasModelId, huaweiCloudMaasModels, ModelInfo } from "@shared/api"
+import { Anthropic } from "@anthropic-ai/sdk"
+import OpenAI from "openai"
+import { convertToOpenAiMessages } from "../transform/openai-format"
+import { ApiStream } from "../transform/stream"
+import { withRetry } from "../retry"
+
+interface HuaweiCloudMaaSHandlerOptions {
+	huaweiCloudMaasApiKey?: string
+	huaweiCloudMaasModelId?: string
+	huaweiCloudMaasModelInfo?: ModelInfo
+}
+
+export class HuaweiCloudMaaSHandler implements ApiHandler {
+	private options: HuaweiCloudMaaSHandlerOptions
+	private client: OpenAI | undefined
+	constructor(options: HuaweiCloudMaaSHandlerOptions) {
+		this.options = options
+	}
+
+	private ensureClient(): OpenAI {
+		if (!this.client) {
+			if (!this.options.huaweiCloudMaasApiKey) {
+				throw new Error("Huawei Cloud MaaS API key is required")
+			}
+			try {
+				this.client = new OpenAI({
+					baseURL: "https://api.modelarts-maas.com/v1/",
+					apiKey: this.options.huaweiCloudMaasApiKey,
+				})
+			} catch (error) {
+				throw new Error(`Error creating Huawei Cloud MaaS client: ${error.message}`)
+			}
+		}
+		return this.client
+	}
+
+	getModel(): { id: HuaweiCloudMaasModelId; info: ModelInfo } {
+		// First priority: huaweiCloudMaasModelId and huaweiCloudMaasModelInfo (like Groq does)
+		const huaweiCloudMaasModelId = this.options.huaweiCloudMaasModelId
+		const huaweiCloudMaasModelInfo = this.options.huaweiCloudMaasModelInfo
+		if (huaweiCloudMaasModelId && huaweiCloudMaasModelInfo) {
+			return { id: huaweiCloudMaasModelId as HuaweiCloudMaasModelId, info: huaweiCloudMaasModelInfo }
+		}
+
+		// Second priority: huaweiCloudMaasModelId with static model info
+		if (huaweiCloudMaasModelId && huaweiCloudMaasModelId in huaweiCloudMaasModels) {
+			const id = huaweiCloudMaasModelId as HuaweiCloudMaasModelId
+			return { id, info: huaweiCloudMaasModels[id] }
+		}
+
+		// Default fallback
+		return {
+			id: huaweiCloudMaasDefaultModelId,
+			info: huaweiCloudMaasModels[huaweiCloudMaasDefaultModelId],
+		}
+	}
+
+	@withRetry()
+	async *createMessage(systemPrompt: string, messages: Anthropic.Messages.MessageParam[]): ApiStream {
+		const client = this.ensureClient()
+		const model = this.getModel()
+		let openAiMessages: OpenAI.Chat.ChatCompletionMessageParam[] = [
+			{ role: "system", content: systemPrompt },
+			...convertToOpenAiMessages(messages),
+		]
+		const stream = await client.chat.completions.create({
+			model: model.id,
+			max_completion_tokens: model.info.maxTokens,
+			messages: openAiMessages,
+			stream: true,
+			stream_options: { include_usage: true },
+			temperature: 0,
+		})
+
+		let reasoning: string | null = null
+		let didOutputUsage: boolean = false
+		let finalUsage: any = null
+
+		for await (const chunk of stream) {
+			const delta = chunk.choices[0]?.delta
+
+			// Handle reasoning content detection
+			if (delta?.content) {
+				if (reasoning || delta.content.includes("<think>")) {
+					reasoning = (reasoning || "") + delta.content
+				} else if (!reasoning) {
+					yield {
+						type: "text",
+						text: delta.content,
+					}
+				}
+			}
+
+			// Handle reasoning output
+			if (reasoning || (delta && "reasoning_content" in delta && delta.reasoning_content)) {
+				const reasoningContent = delta?.content || ((delta as any)?.reasoning_content as string | undefined) || ""
+				if (reasoningContent.trim()) {
+					yield {
+						type: "reasoning",
+						reasoning: reasoningContent,
+					}
+				}
+
+				// Check if reasoning is complete
+				if (reasoning?.includes("</think>")) {
+					reasoning = null
+				}
+			}
+
+			// Store usage information for later output
+			if (chunk.usage) {
+				finalUsage = chunk.usage
+			}
+
+			// Output usage when stream is finished
+			if (!didOutputUsage && chunk.choices?.[0]?.finish_reason) {
+				if (finalUsage) {
+					yield {
+						type: "usage",
+						inputTokens: finalUsage.prompt_tokens || 0,
+						outputTokens: finalUsage.completion_tokens || 0,
+						cacheWriteTokens: 0,
+						cacheReadTokens: 0,
+					}
+				}
+				didOutputUsage = true
+			}
+		}
+	}
+}

src/api/providers/vscode-lm.ts

@@ -1,13 +1,13 @@
 import { Anthropic } from "@anthropic-ai/sdk"
-import * as vscode from "vscode"
-import { ApiHandler, SingleCompletionHandler } from "../"
-import { calculateApiCostAnthropic } from "@utils/cost"
 import { ApiStream } from "@api/transform/stream"
 import { convertToVsCodeLmMessages } from "@api/transform/vscode-lm-format"
-import { SELECTOR_SEPARATOR, stringifyVsCodeLmModelSelector } from "@shared/vsCodeSelectorUtils"
 import { ModelInfo, openAiModelInfoSaneDefaults } from "@shared/api"
-import type { LanguageModelChatSelector as LanguageModelChatSelectorFromTypes } from "./types"
+import { SELECTOR_SEPARATOR, stringifyVsCodeLmModelSelector } from "@shared/vsCodeSelectorUtils"
+import { calculateApiCostAnthropic } from "@utils/cost"
+import * as vscode from "vscode"
+import { ApiHandler, SingleCompletionHandler } from "../"
 import { withRetry } from "../retry"
+import type { LanguageModelChatSelector as LanguageModelChatSelectorFromTypes } from "./types"
 
 interface VsCodeLmHandlerOptions {
 	vsCodeLmModelSelector?: any
@@ -237,82 +237,40 @@ export class VsCodeLmHandler implements ApiHandler, SingleCompletionHandler {
 		}
 	}
 
-	private async countTokens(text: string | vscode.LanguageModelChatMessage): Promise<number> {
-		// Check for required dependencies
-		if (!this.client) {
-			console.warn("Cline <Language Model API>: No client available for token counting")
-			return 0
-		}
-
-		if (!this.currentRequestCancellation) {
-			console.warn("Cline <Language Model API>: No cancellation token available for token counting")
-			return 0
-		}
-
-		// Validate input
-		if (!text) {
-			console.debug("Cline <Language Model API>: Empty text provided for token counting")
-			return 0
-		}
-
-		try {
-			// Handle different input types
-			let tokenCount: number
-
-			if (typeof text === "string") {
-				tokenCount = await this.client.countTokens(text, this.currentRequestCancellation.token)
-			} else if (text instanceof vscode.LanguageModelChatMessage) {
-				// For chat messages, ensure we have content
-				if (!text.content || (Array.isArray(text.content) && text.content.length === 0)) {
-					console.debug("Cline <Language Model API>: Empty chat message content")
-					return 0
-				}
-				tokenCount = await this.client.countTokens(text, this.currentRequestCancellation.token)
-			} else {
-				console.warn("Cline <Language Model API>: Invalid input type for token counting")
-				return 0
-			}
-
-			// Validate the result
-			if (typeof tokenCount !== "number") {
-				console.warn("Cline <Language Model API>: Non-numeric token count received:", tokenCount)
-				return 0
-			}
-
-			if (tokenCount < 0) {
-				console.warn("Cline <Language Model API>: Negative token count received:", tokenCount)
-				return 0
-			}
-
-			return tokenCount
-		} catch (error) {
-			// Handle specific error types
-			if (error instanceof vscode.CancellationError) {
-				console.debug("Cline <Language Model API>: Token counting cancelled by user")
-				return 0
-			}
-
-			const errorMessage = error instanceof Error ? error.message : "Unknown error"
-			console.warn("Cline <Language Model API>: Token counting failed:", errorMessage)
-
-			// Log additional error details if available
-			if (error instanceof Error && error.stack) {
-				console.debug("Token counting error stack:", error.stack)
-			}
-
-			return 0 // Fallback to prevent stream interruption
+	private extractTextFromMessage(message: vscode.LanguageModelChatMessage): string {
+		if (Array.isArray(message.content)) {
+			return message.content
+				.filter((part) => part instanceof vscode.LanguageModelTextPart)
+				.map((part) => (part as vscode.LanguageModelTextPart).value)
+				.join("")
 		}
+		return ""
 	}
 
-	private async calculateTotalInputTokens(
-		systemPrompt: string,
-		vsCodeLmMessages: vscode.LanguageModelChatMessage[],
-	): Promise<number> {
-		const systemTokens: number = await this.countTokens(systemPrompt)
+	private isClaudeModel(): boolean {
+		return this.client?.family?.startsWith("claude") || false
+	}
 
+	private async countTokens(text: string | vscode.LanguageModelChatMessage): Promise<number> {
+		/**
+		 * NOTE (intentional trade-off):
+		 * We use a coarse chars/4 heuristic here instead of a real tokenizer (e.g., js-tiktoken with o200k_base).
+		 * Rationale:
+		 *  - Avoid pulling multi‑MB rank files and increasing the extension install/download size.
+		 *  - Eliminate encoder lifecycle/memory concerns in long-running sessions.
+		 * Consequences:
+		 *  - This is not model-accurate and can under/over-estimate tokens, especially with tool/function calls.
+		 *  - It is “good enough” for budgeting/context checks, and we accept the inaccuracy by design.
+		 * If precise accounting becomes a requirement, reintroduce a tokenizer behind a feature flag or backend-only path.
+		 */
+		const textContent = typeof text === "string" ? text : this.extractTextFromMessage(text)
+		return Math.ceil((textContent || "").length / 4)
+	}
+
+	private async calculateTotalInputTokens(vsCodeLmMessages: vscode.LanguageModelChatMessage[]): Promise<number> {
 		const messageTokens: number[] = await Promise.all(vsCodeLmMessages.map((msg) => this.countTokens(msg)))
 
-		return systemTokens + messageTokens.reduce((sum: number, tokens: number): number => sum + tokens, 0)
+		return messageTokens.reduce((sum: number, tokens: number): number => sum + tokens, 0)
 	}
 
 	private ensureCleanState(): void {
@@ -434,7 +392,7 @@ export class VsCodeLmHandler implements ApiHandler, SingleCompletionHandler {
 		this.currentRequestCancellation = new vscode.CancellationTokenSource()
 
 		// Calculate input tokens before starting the stream
-		const totalInputTokens: number = await this.calculateTotalInputTokens(systemPrompt, vsCodeLmMessages)
+		const totalInputTokens: number = await this.calculateTotalInputTokens(vsCodeLmMessages)
 
 		// Accumulate the text and count at the end of the stream to reduce token counting overhead.
 		let accumulatedText: string = ""

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

@@ -1,14 +1,13 @@
-import { describe, it, beforeEach, afterEach } from "mocha"
+import { HostProvider } from "@/hosts/host-provider"
+import { setVscodeHostProviderMock } from "@/test/host-provider-test-utils"
+import * as diskModule from "@core/storage/disk"
 import { expect } from "chai"
+import { afterEach, beforeEach, describe, it } from "mocha"
+import * as path from "path"
 import * as sinon from "sinon"
 import * as vscode from "vscode"
-import * as path from "path"
+import type { FileMetadataEntry, TaskMetadata } from "./ContextTrackerTypes"
 import { FileContextTracker } from "./FileContextTracker"
-import * as diskModule from "@core/storage/disk"
-import type { TaskMetadata, FileMetadataEntry } from "./ContextTrackerTypes"
-import type { DiffViewProviderCreator, WebviewProviderCreator } from "@/hosts/host-provider"
-import { HostProvider } from "@/hosts/host-provider"
-import { vscodeHostBridgeClient } from "@/hosts/vscode/hostbridge/client/host-grpc-client"
 
 describe("FileContextTracker", () => {
 	let sandbox: sinon.SinonSandbox
@@ -54,13 +53,7 @@ describe("FileContextTracker", () => {
 		getTaskMetadataStub = sandbox.stub(diskModule, "getTaskMetadata").resolves(mockTaskMetadata)
 		saveTaskMetadataStub = sandbox.stub(diskModule, "saveTaskMetadata").resolves()
 
-		// Reset HostProvider before initializing to avoid "already initialized" errors
-		HostProvider.reset()
-		HostProvider.initialize(
-			((_) => {}) as WebviewProviderCreator,
-			(() => {}) as DiffViewProviderCreator,
-			vscodeHostBridgeClient,
-		)
+		setVscodeHostProviderMock()
 
 		// Create tracker instance
 		taskId = "test-task-id"

src/core/controller/account/accountLoginClicked.ts

@@ -1,6 +1,6 @@
 import { Controller } from "../index"
 import { AuthService } from "@/services/auth/AuthService"
-import { EmptyRequest, String } from "../../../shared/proto/common"
+import { EmptyRequest, String } from "@shared/proto/cline/common"
 
 const authService = AuthService.getInstance()
 

src/core/controller/account/accountLogoutClicked.ts

@@ -1,6 +1,6 @@
 import { AuthService } from "@/services/auth/AuthService"
-import { Empty } from "../../../shared/proto/common"
-import type { EmptyRequest } from "../../../shared/proto/common"
+import { Empty } from "@shared/proto/cline/common"
+import type { EmptyRequest } from "@shared/proto/cline/common"
 import type { Controller } from "../index"
 
 const authService = AuthService.getInstance()

src/core/controller/account/authStateChanged.ts

@@ -1,4 +1,4 @@
-import { AuthStateChangedRequest, AuthState } from "@shared/proto/account"
+import { AuthStateChangedRequest, AuthState } from "@shared/proto/cline/account"
 import type { Controller } from "../index"
 import { updateGlobalState } from "../../storage/state"
 

src/core/controller/account/getOrganizationCredits.ts

@@ -1,5 +1,5 @@
 import type { Controller } from "../index"
-import { GetOrganizationCreditsRequest, OrganizationCreditsData, OrganizationUsageTransaction } from "@shared/proto/account"
+import { GetOrganizationCreditsRequest, OrganizationCreditsData, OrganizationUsageTransaction } from "@shared/proto/cline/account"
 
 /**
  * Handles fetching all organization credits data (balance, usage, payments)

src/core/controller/account/getUserCredits.ts

@@ -1,6 +1,6 @@
 import type { Controller } from "../index"
-import type { EmptyRequest } from "@shared/proto/common"
-import { UserCreditsData } from "@shared/proto/account"
+import type { EmptyRequest } from "@shared/proto/cline/common"
+import { UserCreditsData } from "@shared/proto/cline/account"
 
 /**
  * Handles fetching all user credits data (balance, usage, payments)

src/core/controller/account/getUserOrganizations.ts

@@ -1,7 +1,6 @@
 import type { Controller } from "../index"
-import type { EmptyRequest } from "@shared/proto/common"
-import { UserOrganization, UserOrganizationsResponse } from "@shared/proto/account"
-import { getGlobalState } from "../../storage/state"
+import type { EmptyRequest } from "@shared/proto/cline/common"
+import { UserOrganization, UserOrganizationsResponse } from "@shared/proto/cline/account"
 
 /**
  * Handles fetching all user credits data (balance, usage, payments)
@@ -18,14 +17,11 @@ export async function getUserOrganizations(controller: Controller, request: Empt
 		// Fetch user organizations from the account service
 		const organizations = await controller.accountService.fetchUserOrganizationsRPC()
 
-		// Get the current active organization ID from global state
-		const currentActiveOrganizationId = await getGlobalState(controller.context, "currentActiveOrganizationId")
-
 		return UserOrganizationsResponse.create({
 			organizations:
 				organizations?.map((org) =>
 					UserOrganization.create({
-						active: org.organizationId === currentActiveOrganizationId,
+						active: org.active,
 						memberId: org.memberId,
 						name: org.name,
 						organizationId: org.organizationId,

src/core/controller/account/setUserOrganization.ts

@@ -1,10 +1,6 @@
 import type { Controller } from "../index"
-import { Empty } from "@shared/proto/common"
-import { UserOrganizationUpdateRequest } from "@shared/proto/account"
-import { updateGlobalState } from "../../storage/state"
-
-// Backend request deduplication tracking
-const ongoingOrganizationSwitches = new Map<string, Promise<Empty>>()
+import { Empty } from "@shared/proto/cline/common"
+import { UserOrganizationUpdateRequest } from "@shared/proto/cline/account"
 
 /**
  * Handles setting the user's active organization
@@ -13,39 +9,16 @@ const ongoingOrganizationSwitches = new Map<string, Promise<Empty>>()
  * @returns Empty response
  */
 export async function setUserOrganization(controller: Controller, request: UserOrganizationUpdateRequest): Promise<Empty> {
-	// Create a unique key for this request (user + organization)
-	const userId = controller.context.globalState.get("clineUserId") || "unknown"
-	const requestKey = `${userId}:${request.organizationId || "personal"}`
-
-	// Check if there's already an ongoing request for this user+organization combination
-	const existingRequest = ongoingOrganizationSwitches.get(requestKey)
-	if (existingRequest) {
-		return existingRequest
-	}
-
-	// Create the promise for this request
-	const requestPromise = (async (): Promise<Empty> => {
-		try {
-			if (!controller.accountService) {
-				throw new Error("Account service not available")
-			}
-
-			await controller.accountService.switchAccount(request.organizationId)
-
-			// Store the current active organization ID in global state
-			await updateGlobalState(controller.context, "currentActiveOrganizationId", request.organizationId)
-
-			return Empty.create({})
-		} catch (error) {
-			throw error
-		} finally {
-			// Clean up the ongoing request tracking
-			ongoingOrganizationSwitches.delete(requestKey)
+	try {
+		if (!controller.accountService) {
+			throw new Error("Account service not available")
 		}
-	})()
 
-	// Store the promise to prevent duplicate requests
-	ongoingOrganizationSwitches.set(requestKey, requestPromise)
+		// Switch to the specified organization using the account service
+		await controller.accountService.switchAccount(request.organizationId)
 
-	return requestPromise
+		return Empty.create({})
+	} catch (error) {
+		throw error
+	}
 }

src/core/controller/browser/discoverBrowser.ts

@@ -1,5 +1,5 @@
-import { BrowserConnection } from "@shared/proto/browser"
-import { EmptyRequest } from "@shared/proto/common"
+import { BrowserConnection } from "@shared/proto/cline/browser"
+import { EmptyRequest } from "@shared/proto/cline/common"
 import { Controller } from "../index"
 import { getAllExtensionState } from "@core/storage/state"
 import { BrowserSession } from "@services/browser/BrowserSession"

src/core/controller/browser/getBrowserConnectionInfo.ts

@@ -1,5 +1,5 @@
-import { BrowserConnectionInfo } from "@shared/proto/browser"
-import { EmptyRequest } from "@shared/proto/common"
+import { BrowserConnectionInfo } from "@shared/proto/cline/browser"
+import { EmptyRequest } from "@shared/proto/cline/common"
 import { Controller } from "../index"
 import { getAllExtensionState } from "@core/storage/state"
 

src/core/controller/browser/getDetectedChromePath.ts

@@ -1,5 +1,5 @@
-import { ChromePath } from "../../../shared/proto/browser"
-import { EmptyRequest } from "../../../shared/proto/common"
+import { ChromePath } from "@shared/proto/cline/browser"
+import { EmptyRequest } from "@shared/proto/cline/common"
 import { Controller } from "../index"
 import { getAllExtensionState } from "../../storage/state"
 import { BrowserSession } from "../../../services/browser/BrowserSession"

src/core/controller/browser/relaunchChromeDebugMode.ts

@@ -1,4 +1,4 @@
-import { EmptyRequest, String as StringMessage } from "../../../shared/proto/common"
+import { EmptyRequest, String as StringMessage } from "@shared/proto/cline/common"
 import { Controller } from "../index"
 import { BrowserSession } from "../../../services/browser/BrowserSession"
 

src/core/controller/browser/testBrowserConnection.ts

@@ -1,5 +1,5 @@
-import { BrowserConnection } from "@shared/proto/browser"
-import { StringRequest } from "@shared/proto/common"
+import { BrowserConnection } from "@shared/proto/cline/browser"
+import { StringRequest } from "@shared/proto/cline/common"
 import { Controller } from "../index"
 import { getAllExtensionState } from "@core/storage/state"
 import { BrowserSession } from "@services/browser/BrowserSession"

src/core/controller/browser/updateBrowserSettings.ts

@@ -1,5 +1,5 @@
-import { UpdateBrowserSettingsRequest } from "../../../shared/proto/browser"
-import { Boolean } from "../../../shared/proto/common"
+import { UpdateBrowserSettingsRequest } from "@shared/proto/cline/browser"
+import { Boolean } from "@shared/proto/cline/common"
 import { Controller } from "../index"
 import { updateGlobalState, getGlobalState } from "../../storage/state"
 import { BrowserSettings as SharedBrowserSettings, DEFAULT_BROWSER_SETTINGS } from "../../../shared/BrowserSettings"

src/core/controller/checkpoints/checkpointDiff.ts

@@ -1,5 +1,5 @@
 import { Controller } from ".."
-import { Empty, Int64Request } from "@shared/proto/common"
+import { Empty, Int64Request } from "@shared/proto/cline/common"
 
 export async function checkpointDiff(controller: Controller, request: Int64Request): Promise<Empty> {
 	if (request.value) {

src/core/controller/checkpoints/checkpointRestore.ts

@@ -1,8 +1,8 @@
 import { HostProvider } from "@/hosts/host-provider"
 import { Controller } from ".."
 import { ClineCheckpointRestore } from "../../../shared/WebviewMessage"
-import { CheckpointRestoreRequest } from "../../../shared/proto/checkpoints"
-import { Empty } from "../../../shared/proto/common"
+import { CheckpointRestoreRequest } from "@shared/proto/cline/checkpoints"
+import { Empty } from "@shared/proto/cline/common"
 import pWaitFor from "p-wait-for"
 import { ShowMessageType } from "@/shared/proto/index.host"
 

src/core/controller/file/copyToClipboard.ts

@@ -1,5 +1,5 @@
 import { Controller } from ".."
-import { Empty, StringRequest } from "@shared/proto/common"
+import { Empty, StringRequest } from "@shared/proto/cline/common"
 import { writeTextToClipboard } from "@/utils/env"
 
 /**

src/core/controller/file/createRuleFile.ts

@@ -1,5 +1,5 @@
 import { Controller } from ".."
-import { RuleFileRequest, RuleFile } from "@shared/proto/file"
+import { RuleFileRequest, RuleFile } from "@shared/proto/cline/file"
 import { refreshClineRulesToggles } from "@core/context/instructions/user-instructions/cline-rules"
 import { createRuleFile as createRuleFileImpl } from "@core/context/instructions/user-instructions/rule-helpers"
 import * as path from "path"

src/core/controller/file/deleteRuleFile.ts

@@ -1,5 +1,5 @@
 import { deleteRuleFile as deleteRuleFileImpl } from "@core/context/instructions/user-instructions/rule-helpers"
-import { RuleFile, RuleFileRequest } from "@shared/proto/file"
+import { RuleFile, RuleFileRequest } from "@shared/proto/cline/file"
 import * as path from "path"
 import { Controller } from ".."
 import { HostProvider } from "@/hosts/host-provider"

src/core/controller/file/getRelativePaths.ts

@@ -1,5 +1,5 @@
 import { asRelativePath } from "@/utils/path"
-import { RelativePaths, RelativePathsRequest } from "@shared/proto/file"
+import { RelativePaths, RelativePathsRequest } from "@shared/proto/cline/file"
 import * as path from "path"
 import { URI } from "vscode-uri"
 import { Controller } from ".."

src/core/controller/file/openFile.ts

@@ -1,5 +1,5 @@
 import { Controller } from ".."
-import { Empty, StringRequest } from "@shared/proto/common"
+import { Empty, StringRequest } from "@shared/proto/cline/common"
 import { openFile as openFileIntegration } from "@integrations/misc/open-file"
 
 /**

src/core/controller/file/openImage.ts

@@ -1,5 +1,5 @@
 import { Controller } from ".."
-import { Empty, StringRequest } from "@shared/proto/common"
+import { Empty, StringRequest } from "@shared/proto/cline/common"
 import { openImage as openImageIntegration } from "@integrations/misc/open-file"
 
 /**

src/core/controller/file/openMention.ts

@@ -1,5 +1,5 @@
 import { Controller } from ".."
-import { Empty, StringRequest } from "../../../shared/proto/common"
+import { Empty, StringRequest } from "@shared/proto/cline/common"
 import { openMention as coreOpenMention } from "../../mentions"
 
 /**

src/core/controller/file/openTaskHistory.ts

@@ -1,5 +1,5 @@
 import { Controller } from ".."
-import { Empty, StringRequest } from "@shared/proto/common"
+import { Empty, StringRequest } from "@shared/proto/cline/common"
 import { openFile as openFileIntegration } from "@integrations/misc/open-file"
 import path from "path"
 /**

src/core/controller/file/refreshRules.ts

@@ -1,5 +1,5 @@
-import { EmptyRequest } from "@shared/proto/common"
-import { RefreshedRules } from "@shared/proto/file"
+import { EmptyRequest } from "@shared/proto/cline/common"
+import { RefreshedRules } from "@shared/proto/cline/file"
 import type { Controller } from "../index"
 import { refreshClineRulesToggles } from "@core/context/instructions/user-instructions/cline-rules"
 import { refreshExternalRulesToggles } from "@core/context/instructions/user-instructions/external-rules"

src/core/controller/file/searchCommits.ts

@@ -1,6 +1,6 @@
 import { Controller } from ".."
-import { GitCommits } from "@shared/proto/file"
-import { StringRequest } from "@shared/proto/common"
+import { GitCommits } from "@shared/proto/cline/file"
+import { StringRequest } from "@shared/proto/cline/common"
 import { searchCommits as searchCommitsUtil } from "@utils/git"
 import { getWorkspacePath } from "@utils/path"
 import { convertGitCommitsToProtoGitCommits } from "@shared/proto-conversions/file/git-commit-conversion"

src/core/controller/file/searchFiles.ts

@@ -1,5 +1,5 @@
 import { Controller } from ".."
-import { FileSearchRequest, FileSearchResults } from "@shared/proto/file"
+import { FileSearchRequest, FileSearchResults } from "@shared/proto/cline/file"
 import { searchWorkspaceFiles } from "@services/search/file-search"
 import { getWorkspacePath } from "@utils/path"
 import { convertSearchResultsToProtoFileInfos } from "@shared/proto-conversions/file/search-result-conversion"

src/core/controller/file/selectFiles.ts

@@ -1,5 +1,5 @@
 import { Controller } from ".."
-import { BooleanRequest, StringArrays } from "@shared/proto/common"
+import { BooleanRequest, StringArrays } from "@shared/proto/cline/common"
 import { selectFiles as selectFilesIntegration } from "@integrations/misc/process-files"
 
 /**

src/core/controller/file/subscribeToWorkspaceUpdates.ts

@@ -1,5 +1,5 @@
 import { Controller } from "../index"
-import { EmptyRequest, StringArray } from "@shared/proto/common"
+import { EmptyRequest, StringArray } from "@shared/proto/cline/common"
 import { StreamingResponseHandler, getRequestRegistry } from "../grpc-handler"
 
 // Keep track of active subscriptions

src/core/controller/file/toggleClineRule.ts

@@ -1,5 +1,5 @@
-import { ToggleClineRules } from "../../../shared/proto/file"
-import type { ToggleClineRuleRequest } from "../../../shared/proto/file"
+import { ToggleClineRules } from "@shared/proto/cline/file"
+import type { ToggleClineRuleRequest } from "@shared/proto/cline/file"
 import type { Controller } from "../index"
 import { getGlobalState, getWorkspaceState, updateGlobalState, updateWorkspaceState } from "../../../core/storage/state"
 import { ClineRulesToggles as AppClineRulesToggles } from "@shared/cline-rules"

src/core/controller/file/toggleCursorRule.ts

@@ -1,5 +1,5 @@
-import type { ToggleCursorRuleRequest } from "../../../shared/proto/file"
-import { ClineRulesToggles } from "../../../shared/proto/file"
+import type { ToggleCursorRuleRequest } from "@shared/proto/cline/file"
+import { ClineRulesToggles } from "@shared/proto/cline/file"
 import type { Controller } from "../index"
 import { getWorkspaceState, updateWorkspaceState } from "../../../core/storage/state"
 import { ClineRulesToggles as AppClineRulesToggles } from "@shared/cline-rules"

src/core/controller/file/toggleWindsurfRule.ts

@@ -1,5 +1,5 @@
-import type { ToggleWindsurfRuleRequest } from "../../../shared/proto/file"
-import { ClineRulesToggles } from "../../../shared/proto/file"
+import type { ToggleWindsurfRuleRequest } from "@shared/proto/cline/file"
+import { ClineRulesToggles } from "@shared/proto/cline/file"
 import type { Controller } from "../index"
 import { getWorkspaceState, updateWorkspaceState } from "../../../core/storage/state"
 import { ClineRulesToggles as AppClineRulesToggles } from "@shared/cline-rules"

src/core/controller/file/toggleWorkflow.ts

@@ -1,6 +1,6 @@
 import { Controller } from ".."
-import { Metadata } from "../../../shared/proto/common"
-import { ToggleWorkflowRequest, ClineRulesToggles } from "../../../shared/proto/file"
+import { Metadata } from "@shared/proto/cline/common"
+import { ToggleWorkflowRequest, ClineRulesToggles } from "@shared/proto/cline/file"
 import { getWorkspaceState, updateWorkspaceState, getGlobalState, updateGlobalState } from "../../../core/storage/state"
 import { ClineRulesToggles as AppClineRulesToggles } from "../../../shared/cline-rules"
 

src/core/controller/grpc-handler.test.ts

@@ -0,0 +1,412 @@
+import { describe, it, beforeEach, afterEach } from "mocha"
+import { expect } from "chai"
+import * as sinon from "sinon"
+import { handleGrpcRequest, handleGrpcRequestCancel, getRequestRegistry } from "./grpc-handler"
+import { Controller } from "@core/controller"
+import { GrpcRequest, GrpcCancel } from "@shared/WebviewMessage"
+import { serviceHandlers } from "@generated/hosts/vscode/protobus-services"
+
+describe("grpc-handler", () => {
+	let sandbox: sinon.SinonSandbox
+	let mockController: sinon.SinonStubbedInstance<Controller>
+
+	let mockUnaryHandler: sinon.SinonStub
+	let mockUnaryFailingHandler: sinon.SinonStub
+	let mockStreamingHandler: sinon.SinonStub
+	let mockStreamingFailingHandler: sinon.SinonStub
+
+	const serviceName = "cline.TestService"
+	const mockResponse = { result: "result-1234" }
+
+	beforeEach(() => {
+		sandbox = sinon.createSandbox()
+
+		// Create a mock controller
+		mockController = {
+			postMessageToWebview: sandbox.stub().resolves(),
+		} as any
+
+		// Create mock service handlers
+		mockUnaryHandler = sandbox.stub().resolves(mockResponse)
+		mockStreamingHandler = sandbox.stub().resolves()
+		mockUnaryFailingHandler = sandbox.stub().rejects(new Error("Test error unary"))
+		mockStreamingFailingHandler = sandbox.stub().rejects(new Error("Stream error"))
+		serviceHandlers[serviceName] = {
+			testUnary: mockUnaryHandler,
+			testUnaryFailing: mockUnaryFailingHandler,
+			testStreaming: mockStreamingHandler,
+			testStreamingFailing: mockStreamingFailingHandler,
+		}
+	})
+
+	afterEach(() => {
+		sandbox.restore()
+	})
+
+	describe("handleGrpcRequest", () => {
+		describe("Unary requests", () => {
+			it("should handle successful unary requests", async () => {
+				const request: GrpcRequest = {
+					service: serviceName,
+					method: "testUnary",
+					message: { input: "test" },
+					request_id: "test-123",
+					is_streaming: false,
+				}
+
+				await handleGrpcRequest(mockController as any, request)
+
+				// Verify the handler was called
+				expect(mockUnaryHandler.calledOnce).to.be.true
+				expect(mockUnaryHandler.firstCall.args[0]).to.equal(mockController)
+				expect(mockUnaryHandler.firstCall.args[1]).to.deep.equal({ input: "test" })
+
+				// Verify the response was sent
+				expect(mockController.postMessageToWebview.calledOnce).to.be.true
+				const sentMessage = mockController.postMessageToWebview.firstCall.args[0]
+				expect(sentMessage).to.deep.equal({
+					type: "grpc_response",
+					grpc_response: {
+						message: mockResponse,
+						request_id: "test-123",
+					},
+				})
+			})
+
+			it("should handle errors in unary requests", async () => {
+				const request: GrpcRequest = {
+					service: serviceName,
+					method: "testUnaryFailing",
+					message: { input: "test" },
+					request_id: "test-456",
+					is_streaming: false,
+				}
+
+				await handleGrpcRequest(mockController as any, request)
+
+				// Verify the error response was sent
+				expect(mockController.postMessageToWebview.calledOnce).to.be.true
+				const sentMessage = mockController.postMessageToWebview.firstCall.args[0]
+				expect(sentMessage).to.deep.equal({
+					type: "grpc_response",
+					grpc_response: {
+						error: "Test error unary",
+						request_id: "test-456",
+						is_streaming: false,
+					},
+				})
+			})
+
+			it("should handle unknown service errors", async () => {
+				const request: GrpcRequest = {
+					service: "UnknownService",
+					method: "someMethod",
+					message: {},
+					request_id: "test-789",
+					is_streaming: false,
+				}
+
+				await handleGrpcRequest(mockController as any, request)
+
+				// Verify the error response was sent
+				expect(mockController.postMessageToWebview.calledOnce).to.be.true
+				const sentMessage = mockController.postMessageToWebview.firstCall.args[0]
+				expect(sentMessage.type).to.equal("grpc_response")
+				expect(sentMessage.grpc_response?.error).to.include("Unknown service: UnknownService")
+				expect(sentMessage.grpc_response?.request_id).to.equal("test-789")
+			})
+
+			it("should handle unknown method errors", async () => {
+				const request: GrpcRequest = {
+					service: serviceName,
+					method: "unknownMethod",
+					message: {},
+					request_id: "test-999",
+					is_streaming: false,
+				}
+
+				await handleGrpcRequest(mockController as any, request)
+
+				// Verify the error response was sent
+				expect(mockController.postMessageToWebview.calledOnce).to.be.true
+				const sentMessage = mockController.postMessageToWebview.firstCall.args[0]
+				expect(sentMessage.type).to.equal("grpc_response")
+				expect(sentMessage.grpc_response?.error).to.include("Unknown rpc: cline.TestService.unknownMethod")
+				expect(sentMessage.grpc_response?.request_id).to.equal("test-999")
+			})
+		})
+		describe("Streaming requests", () => {
+			it("should handle successful streaming requests", async () => {
+				// Set up a streaming handler that sends multiple responses
+				const request: GrpcRequest = {
+					service: serviceName,
+					method: "testStreaming",
+					message: { input: "stream" },
+					request_id: "stream-123",
+					is_streaming: true,
+				}
+
+				// Reset the mock and set up the handler using callsFake
+				mockStreamingHandler.reset()
+				mockStreamingHandler.callsFake(async (controller: any, message: any, responseStream: any, requestId: string) => {
+					// Simulate streaming multiple messages
+					await responseStream({ value: 1 }, false, 0)
+					await responseStream({ value: 2 }, false, 1)
+					await responseStream({ value: 3 }, true, 2) // Last message
+				})
+
+				await handleGrpcRequest(mockController as any, request)
+
+				// Verify the handler was called
+				expect(mockStreamingHandler.calledOnce).to.be.true
+				expect(mockStreamingHandler.firstCall.args[0]).to.equal(mockController)
+				expect(mockStreamingHandler.firstCall.args[1]).to.deep.equal({ input: "stream" })
+				expect(mockStreamingHandler.firstCall.args[3]).to.equal("stream-123")
+
+				// Verify all streaming responses were sent
+				expect(mockController.postMessageToWebview.callCount).to.equal(3)
+
+				// Check all responses
+				expect(mockController.postMessageToWebview.firstCall.args[0]).to.deep.equal({
+					type: "grpc_response",
+					grpc_response: {
+						message: { value: 1 },
+						request_id: "stream-123",
+						is_streaming: true,
+						sequence_number: 0,
+					},
+				})
+				expect(mockController.postMessageToWebview.secondCall.args[0]).to.deep.equal({
+					type: "grpc_response",
+					grpc_response: {
+						message: { value: 2 },
+						request_id: "stream-123",
+						is_streaming: true,
+						sequence_number: 1,
+					},
+				})
+				expect(mockController.postMessageToWebview.thirdCall.args[0]).to.deep.equal({
+					type: "grpc_response",
+					grpc_response: {
+						message: { value: 3 },
+						request_id: "stream-123",
+						is_streaming: false, // Last message has is_streaming: false
+						sequence_number: 2,
+					},
+				})
+			})
+
+			it("should handle errors in streaming requests", async () => {
+				const request: GrpcRequest = {
+					service: serviceName,
+					method: "testStreamingFailing",
+					message: { input: "stream" },
+					request_id: "stream-456",
+					is_streaming: true,
+				}
+
+				await handleGrpcRequest(mockController as any, request)
+
+				// Verify the error response was sent
+				expect(mockController.postMessageToWebview.calledOnce).to.be.true
+				const sentMessage = mockController.postMessageToWebview.firstCall.args[0]
+				expect(sentMessage).to.deep.equal({
+					type: "grpc_response",
+					grpc_response: {
+						error: "Stream error",
+						request_id: "stream-456",
+						is_streaming: false,
+					},
+				})
+			})
+
+			it("should handle streaming with message, error, then another message", async () => {
+				// This test simulates a scenario where:
+				// 1. First message is sent successfully
+				// 2. An error occurs
+				// 3. Another message is attempted (which should not be sent after error)
+
+				const request: GrpcRequest = {
+					service: serviceName,
+					method: "testStreaming",
+					message: { input: "stream-with-error" },
+					request_id: "stream-error-mid",
+					is_streaming: true,
+				}
+
+				// Reset the mock and set up the handler to throw an error after being called
+				mockStreamingHandler.reset()
+				mockStreamingHandler.callsFake(async (controller: any, message: any, responseStream: any, requestId: string) => {
+					// Send first message successfully
+					await responseStream({ value: "first" }, false, 0)
+					// Throw an error
+					throw new Error("Mid-stream error")
+				})
+
+				await handleGrpcRequest(mockController as any, request)
+
+				// Verify the handler was called
+				expect(mockStreamingHandler.calledOnce).to.be.true
+
+				// Verify that we got the first message and then the error
+				expect(mockController.postMessageToWebview.callCount).to.equal(2)
+
+				// Check first message was sent successfully
+				expect(mockController.postMessageToWebview.firstCall.args[0]).to.deep.equal({
+					type: "grpc_response",
+					grpc_response: {
+						message: { value: "first" },
+						request_id: "stream-error-mid",
+						is_streaming: true,
+						sequence_number: 0,
+					},
+				})
+
+				// Check error response was sent
+				expect(mockController.postMessageToWebview.secondCall.args[0]).to.deep.equal({
+					type: "grpc_response",
+					grpc_response: {
+						error: "Mid-stream error",
+						request_id: "stream-error-mid",
+						is_streaming: false,
+					},
+				})
+
+				// Try to send another message after the error (simulating what might happen
+				// if the handler tried to continue after an error)
+				const responseStream = mockStreamingHandler.firstCall.args[2]
+
+				// This should still work as the responseStream function is still valid
+				await responseStream({ value: "after-error" }, false, 1)
+
+				// Verify we now have 3 total calls (first message, error, after-error message)
+				expect(mockController.postMessageToWebview.callCount).to.equal(3)
+
+				// Verify the message after error was still sent
+				// (In a real scenario, the handler would have stopped due to the error,
+				// but this tests that the responseStream function itself still works)
+				expect(mockController.postMessageToWebview.thirdCall.args[0]).to.deep.equal({
+					type: "grpc_response",
+					grpc_response: {
+						message: { value: "after-error" },
+						request_id: "stream-error-mid",
+						is_streaming: true,
+						sequence_number: 1,
+					},
+				})
+			})
+		})
+
+		describe("handleGrpcRequestCancel", () => {
+			it("should cancel an active request", async () => {
+				// Register a request in the registry
+				const registry = getRequestRegistry()
+				const cleanupStub = sandbox.stub()
+				registry.registerRequest("cancel-123", cleanupStub)
+
+				const cancelRequest: GrpcCancel = {
+					request_id: "cancel-123",
+				}
+
+				await handleGrpcRequestCancel(mockController as any, cancelRequest)
+
+				// Verify the cleanup was called
+				expect(cleanupStub.calledOnce).to.be.true
+
+				// Verify the cancellation confirmation was sent
+				expect(mockController.postMessageToWebview.calledOnce).to.be.true
+				const sentMessage = mockController.postMessageToWebview.firstCall.args[0]
+				expect(sentMessage).to.deep.equal({
+					type: "grpc_response",
+					grpc_response: {
+						message: { cancelled: true },
+						request_id: "cancel-123",
+						is_streaming: false,
+					},
+				})
+
+				// Verify the request was removed from the registry
+				expect(registry.hasRequest("cancel-123")).to.be.false
+			})
+
+			it("should handle cancellation of non-existent request", async () => {
+				const cancelRequest: GrpcCancel = {
+					request_id: "non-existent",
+				}
+
+				await handleGrpcRequestCancel(mockController as any, cancelRequest)
+
+				// Verify no message was sent (request not found)
+				expect(mockController.postMessageToWebview.called).to.be.false
+			})
+
+			it("should handle cleanup errors gracefully", async () => {
+				// Register a request with a failing cleanup
+				const registry = getRequestRegistry()
+				const cleanupStub = sandbox.stub().throws(new Error("Cleanup failed"))
+				registry.registerRequest("cancel-error", cleanupStub)
+
+				const cancelRequest: GrpcCancel = {
+					request_id: "cancel-error",
+				}
+
+				// Should not throw
+				await handleGrpcRequestCancel(mockController as any, cancelRequest)
+
+				// Verify the cleanup was attempted
+				expect(cleanupStub.calledOnce).to.be.true
+
+				// Verify the cancellation confirmation was still sent
+				expect(mockController.postMessageToWebview.calledOnce).to.be.true
+
+				// Verify the request was removed despite the error
+				expect(registry.hasRequest("cancel-error")).to.be.false
+			})
+		})
+
+		describe("Concurrent requests", () => {
+			it("should handle concurrent requests", async () => {
+				// Set up handlers
+				mockUnaryHandler.resolves({ result: "unary" })
+				mockStreamingHandler.callsFake(async (controller: any, message: any, responseStream: any) => {
+					await responseStream({ value: "stream1" }, false, 0)
+					await responseStream({ value: "stream2" }, true, 1)
+				})
+
+				// Send multiple requests concurrently
+				const requests = [
+					handleGrpcRequest(mockController as any, {
+						service: serviceName,
+						method: "testUnary",
+						message: { id: 1 },
+						request_id: "concurrent-1",
+						is_streaming: false,
+					}),
+					handleGrpcRequest(mockController as any, {
+						service: serviceName,
+						method: "testStreaming",
+						message: { id: 2 },
+						request_id: "concurrent-2",
+						is_streaming: true,
+					}),
+					handleGrpcRequest(mockController as any, {
+						service: serviceName,
+						method: "testUnary",
+						message: { id: 3 },
+						request_id: "concurrent-3",
+						is_streaming: false,
+					}),
+				]
+
+				await Promise.all(requests)
+
+				// Verify all handlers were called
+				expect(mockUnaryHandler.callCount).to.equal(2)
+				expect(mockStreamingHandler.callCount).to.equal(1)
+
+				// Verify all responses were sent (2 unary + 2 streaming)
+				expect(mockController.postMessageToWebview.callCount).to.equal(4)
+			})
+		})
+	})
+})

src/core/controller/grpc-handler.ts

@@ -1,6 +1,7 @@
 import { Controller } from "./index"
 import { serviceHandlers } from "@generated/hosts/vscode/protobus-services"
 import { GrpcRequestRegistry } from "./grpc-request-registry"
+import { GrpcCancel, GrpcRequest } from "@/shared/WebviewMessage"
 
 /**
  * Type definition for a streaming response handler
@@ -12,152 +13,34 @@ export type StreamingResponseHandler<TResponse> = (
 ) => Promise<void>
 
 /**
- * Handles gRPC requests from the webview
+ * Handles a gRPC request from the webview.
  */
-export class GrpcHandler {
-	constructor(private controller: Controller) {}
-
-	/**
-	 * Handle a gRPC request from the webview
-	 * @param service The service name
-	 * @param method The method name
-	 * @param message The request message
-	 * @param requestId The request ID for response correlation
-	 * @param isStreaming Whether this is a streaming request
-	 * @returns The response message or error for unary requests, void for streaming requests
-	 */
-	async handleRequest(
-		service: string,
-		method: string,
-		message: any,
-		requestId: string,
-		isStreaming: boolean = false,
-	): Promise<{
-		message?: any
-		error?: string
-		request_id: string
-	} | void> {
-		try {
-			// If this is a streaming request, use the streaming handler
-			if (isStreaming) {
-				await this.handleStreamingRequest(service, method, message, requestId)
-				return
-			}
-
-			// Get the service handler from the config
-			const handler = getHandler(service, method)
-
-			// Handle unary request
-			return {
-				message: await handler(this.controller, message),
-				request_id: requestId,
-			}
-		} catch (error) {
-			return {
-				error: error instanceof Error ? error.message : String(error),
-				request_id: requestId,
-			}
-		}
-	}
-
-	/**
-	 * Handle a streaming gRPC request
-	 * @param service The service name
-	 * @param method The method name
-	 * @param message The request message
-	 * @param requestId The request ID for response correlation
-	 */
-	private async handleStreamingRequest(service: string, method: string, message: any, requestId: string): Promise<void> {
-		// Create a response stream function
-		const responseStream: StreamingResponseHandler<any> = async (
-			response: any,
-			isLast: boolean = false,
-			sequenceNumber?: number,
-		) => {
-			await this.controller.postMessageToWebview({
-				type: "grpc_response",
-				grpc_response: {
-					message: response,
-					request_id: requestId,
-					is_streaming: !isLast,
-					sequence_number: sequenceNumber,
-				},
-			})
-		}
-
-		try {
-			// Get the service handler from the config
-			const handler = getHandler(service, method)
-
-			// Handle streaming request and pass the requestId to all streaming handlers
-			await handler(this.controller, message, responseStream, requestId)
-
-			// Don't send a final message here - the stream should stay open for future updates
-			// The stream will be closed when the client disconnects or when the service explicitly ends it
-		} catch (error) {
-			// Send error response
-			console.log("Protobus error:", error)
-			await this.controller.postMessageToWebview({
-				type: "grpc_response",
-				grpc_response: {
-					error: error instanceof Error ? error.message : String(error),
-					request_id: requestId,
-					is_streaming: false,
-				},
-			})
-		}
+export async function handleGrpcRequest(controller: Controller, request: GrpcRequest): Promise<void> {
+	if (request.is_streaming) {
+		await handleStreamingRequest(controller, request)
+	} else {
+		await handleUnaryRequest(controller, request)
 	}
 }
 
-// Registry to track active gRPC requests and their cleanup functions
-const requestRegistry = new GrpcRequestRegistry()
-
 /**
- * Handle a gRPC request from the webview
- * @param controller The controller instance
- * @param request The gRPC request
+ * Handles a gRPC unary request from the webview.
+ *
+ * Calls the handler using the service and method name, and then posts the result back to the webview.
  */
-export async function handleGrpcRequest(
-	controller: Controller,
-	request: {
-		service: string
-		method: string
-		message: any
-		request_id: string
-		is_streaming?: boolean
-	},
-) {
+async function handleUnaryRequest(controller: Controller, request: GrpcRequest): Promise<void> {
 	try {
-		const grpcHandler = new GrpcHandler(controller)
-
-		// For streaming requests, handleRequest handles sending responses directly
-		if (request.is_streaming) {
-			try {
-				await grpcHandler.handleRequest(request.service, request.method, request.message, request.request_id, true)
-			} finally {
-				// Note: We don't automatically clean up here anymore
-				// The request will be cleaned up when it completes or is cancelled
-			}
-			return
-		}
-
-		// For unary requests, we get a response and send it back
-		const response = (await grpcHandler.handleRequest(
-			request.service,
-			request.method,
-			request.message,
-			request.request_id,
-			false,
-		)) as {
-			message?: any
-			error?: string
-			request_id: string
-		}
-
-		// Send the response back to the webview
+		// Get the service handler from the config
+		const handler = getHandler(request.service, request.method)
+		// Handle unary request
+		const response = await handler(controller, request.message)
+		// Send response to the webview
 		await controller.postMessageToWebview({
 			type: "grpc_response",
-			grpc_response: response,
+			grpc_response: {
+				message: response,
+				request_id: request.request_id,
+			},
 		})
 	} catch (error) {
 		// Send error response
@@ -167,22 +50,65 @@ export async function handleGrpcRequest(
 			grpc_response: {
 				error: error instanceof Error ? error.message : String(error),
 				request_id: request.request_id,
+				is_streaming: false,
 			},
 		})
 	}
 }
 
 /**
- * Handle a gRPC request cancellation from the webview
+ * Handle a streaming gRPC request from the webview.
+ *
+ * Calls the handler using the service and method name, and creates a streaming response handler
+ * which posts results back to the webview.
+ */
+async function handleStreamingRequest(controller: Controller, request: GrpcRequest): Promise<void> {
+	// Create a response stream function
+	const responseStream: StreamingResponseHandler<any> = async (
+		response: any,
+		isLast: boolean = false,
+		sequenceNumber?: number,
+	) => {
+		await controller.postMessageToWebview({
+			type: "grpc_response",
+			grpc_response: {
+				message: response,
+				request_id: request.request_id,
+				is_streaming: !isLast,
+				sequence_number: sequenceNumber,
+			},
+		})
+	}
+
+	try {
+		// Get the service handler from the config
+		const handler = getHandler(request.service, request.method)
+
+		// Handle streaming request and pass the requestId to all streaming handlers
+		await handler(controller, request.message, responseStream, request.request_id)
+
+		// Don't send a final message here - the stream should stay open for future updates
+		// The stream will be closed when the client disconnects or when the service explicitly ends it
+	} catch (error) {
+		// Send error response
+		console.log("Protobus error:", error)
+		await controller.postMessageToWebview({
+			type: "grpc_response",
+			grpc_response: {
+				error: error instanceof Error ? error.message : String(error),
+				request_id: request.request_id,
+				is_streaming: false,
+			},
+		})
+	}
+}
+
+/**
+ * Handles a gRPC request cancellation from the webview.
  * @param controller The controller instance
  * @param request The cancellation request
  */
-export async function handleGrpcRequestCancel(
-	controller: Controller,
-	request: {
-		request_id: string
-	},
-) {
+export async function handleGrpcRequestCancel(controller: Controller, request: GrpcCancel) {
 	const cancelled = requestRegistry.cancelRequest(request.request_id)
 
 	if (cancelled) {
@@ -200,6 +126,17 @@ export async function handleGrpcRequestCancel(
 	}
 }
 
+// Registry to track active gRPC requests and their cleanup functions
+const requestRegistry = new GrpcRequestRegistry()
+
+/**
+ * Get the request registry instance
+ * This allows other parts of the code to access the registry
+ */
+export function getRequestRegistry(): GrpcRequestRegistry {
+	return requestRegistry
+}
+
 function getHandler(serviceName: string, methodName: string): any {
 	// Get the service handler from the config
 	const serviceConfig = serviceHandlers[serviceName]
@@ -212,11 +149,3 @@ function getHandler(serviceName: string, methodName: string): any {
 	}
 	return handler
 }
-
-/**
- * Get the request registry instance
- * This allows other parts of the code to access the registry
- */
-export function getRequestRegistry(): GrpcRequestRegistry {
-	return requestRegistry
-}

src/core/controller/index.ts

@@ -1,7 +1,7 @@
 import { clineEnvConfig } from "@/config"
 import { HostProvider } from "@/hosts/host-provider"
 import { AuthService } from "@/services/auth/AuthService"
-import { telemetryService } from "@/services/posthog/telemetry/TelemetryService"
+import { PostHogClientProvider, telemetryService } from "@/services/posthog/PostHogClientProvider"
 import { ShowMessageType } from "@/shared/proto/host/window"
 import { getCwd, getDesktopDir } from "@/utils/path"
 import { Anthropic } from "@anthropic-ai/sdk"
@@ -13,7 +13,7 @@ import { ClineAccountService } from "@services/account/ClineAccountService"
 import { McpHub } from "@services/mcp/McpHub"
 import { ApiProvider, ModelInfo } from "@shared/api"
 import { ChatContent } from "@shared/ChatContent"
-import { ChatSettings, Mode, StoredChatSettings } from "@shared/ChatSettings"
+import { Mode } from "@shared/storage/types"
 import { ClineRulesToggles } from "@shared/cline-rules"
 import { ExtensionMessage, ExtensionState, Platform } from "@shared/ExtensionMessage"
 import { HistoryItem } from "@shared/HistoryItem"
@@ -30,12 +30,13 @@ import * as path from "path"
 import * as vscode from "vscode"
 import { ensureMcpServersDirectoryExists, ensureSettingsDirectoryExists, GlobalFileNames } from "../storage/disk"
 import { getAllExtensionState, getGlobalState, getWorkspaceState, storeSecret, updateGlobalState } from "../storage/state"
+import { CacheService, PersistenceErrorEvent } from "../storage/CacheService"
 import { Task } from "../task"
 import { handleGrpcRequest, handleGrpcRequestCancel } from "./grpc-handler"
 import { sendMcpMarketplaceCatalogEvent } from "./mcp/subscribeToMcpMarketplaceCatalog"
 import { sendStateUpdate } from "./state/subscribeToState"
 import { sendAddToInputEvent } from "./ui/subscribeToAddToInput"
-
+import { getLatestAnnouncementId } from "@/utils/announcements"
 /*
 https://github.com/microsoft/vscode-webview-ui-toolkit-samples/blob/main/default/weather-webview/src/providers/WeatherViewProvider.ts
 
@@ -52,34 +53,59 @@ export class Controller {
 	workspaceTracker: WorkspaceTracker
 	mcpHub: McpHub
 	accountService: ClineAccountService
-	authService: AuthService
-	get latestAnnouncementId(): string {
-		return this.context.extension?.packageJSON?.version?.split(".").slice(0, 2).join(".") ?? ""
-	}
+	readonly cacheService: CacheService
 
 	constructor(
 		readonly context: vscode.ExtensionContext,
-		private readonly outputChannel: vscode.OutputChannel,
 		postMessage: (message: ExtensionMessage) => Thenable<boolean> | undefined,
 		id: string,
 	) {
 		this.id = id
-		this.outputChannel.appendLine("ClineProvider instantiated")
+
+		HostProvider.get().logToChannel("ClineProvider instantiated")
 		this.postMessage = postMessage
+		this.accountService = ClineAccountService.getInstance()
+		this.cacheService = new CacheService(context)
+		const authService = AuthService.getInstance(this)
+
+		// Initialize cache service asynchronously - critical for extension functionality
+		this.cacheService
+			.initialize()
+			.then(() => {
+				authService.restoreRefreshTokenAndRetrieveAuthInfo()
+			})
+			.catch((error) => {
+				console.error("CRITICAL: Failed to initialize CacheService - extension may not function properly:", error)
+			})
+
+		// Set up persistence error recovery
+		this.cacheService.onPersistenceError = async ({ error }: PersistenceErrorEvent) => {
+			console.error("Cache persistence failed, recovering:", error)
+			try {
+				await this.cacheService.reInitialize()
+				await this.postStateToWebview()
+				HostProvider.window.showMessage({
+					type: ShowMessageType.WARNING,
+					message: "Saving settings to storage failed.",
+				})
+			} catch (recoveryError) {
+				console.error("Cache recovery failed:", recoveryError)
+				HostProvider.window.showMessage({
+					type: ShowMessageType.ERROR,
+					message: "Failed to save settings. Please restart the extension.",
+				})
+			}
+		}
 
 		this.workspaceTracker = new WorkspaceTracker()
 		this.mcpHub = new McpHub(
 			() => ensureMcpServersDirectoryExists(),
 			() => ensureSettingsDirectoryExists(this.context),
-			(msg) => this.postMessageToWebview(msg),
 			this.context.extension?.packageJSON?.version ?? "1.0.0",
 		)
-		this.accountService = ClineAccountService.getInstance()
-		this.authService = AuthService.getInstance(context)
-		this.authService.restoreRefreshTokenAndRetrieveAuthInfo()
 
 		// Clean up legacy checkpoints
-		cleanupLegacyCheckpoints(this.context.globalStorageUri.fsPath, this.outputChannel).catch((error) => {
+		cleanupLegacyCheckpoints(this.context.globalStorageUri.fsPath).catch((error) => {
 			console.error("Failed to cleanup legacy checkpoints:", error)
 		})
 	}
@@ -111,12 +137,18 @@ export class Controller {
 	async handleSignOut() {
 		try {
 			// TODO: update to clineAccountId and then move clineApiKey to a clear function.
-			await storeSecret(this.context, "clineAccountId", undefined)
+			this.cacheService.setSecret("clineAccountId", undefined)
 			await updateGlobalState(this.context, "userInfo", undefined)
-			await Promise.all([
-				updateGlobalState(this.context, "planModeApiProvider", "openrouter"),
-				updateGlobalState(this.context, "actModeApiProvider", "openrouter"),
-			])
+
+			// Update API providers through cache service
+			const apiConfiguration = this.cacheService.getApiConfiguration()
+			const updatedConfig = {
+				...apiConfiguration,
+				planModeApiProvider: "openrouter" as ApiProvider,
+				actModeApiProvider: "openrouter" as ApiProvider,
+			}
+			this.cacheService.setApiConfiguration(updatedConfig)
+
 			await this.postStateToWebview()
 			HostProvider.window.showMessage({
 				type: ShowMessageType.INFORMATION,
@@ -136,11 +168,16 @@ export class Controller {
 
 	async initTask(task?: string, images?: string[], files?: string[], historyItem?: HistoryItem) {
 		await this.clearTask() // ensures that an existing task doesn't exist before starting a new one, although this shouldn't be possible since user must clear task before starting a new one
+
+		// Get API configuration from cache for immediate access
+		const apiConfiguration = this.cacheService.getApiConfiguration()
+
 		const {
-			apiConfiguration,
 			autoApprovalSettings,
 			browserSettings,
-			chatSettings: storedChatSettings,
+			preferredLanguage,
+			openaiReasoningEffort,
+			mode,
 			shellIntegrationTimeout,
 			terminalReuseEnabled,
 			terminalOutputLineLimit,
@@ -148,17 +185,9 @@ export class Controller {
 			enableCheckpointsSetting,
 			isNewUser,
 			taskHistory,
+			strictPlanModeEnabled,
 		} = await getAllExtensionState(this.context)
 
-		// Get current mode using helper function
-		const currentMode = await this.getCurrentMode()
-
-		// Reconstruct ChatSettings with mode from global state and stored preferences
-		const chatSettings: ChatSettings = {
-			...storedChatSettings, // Spread stored preferences (preferredLanguage, openAIReasoningEffort)
-			mode: currentMode, // Use mode from global state
-		}
-
 		const NEW_USER_TASK_COUNT_THRESHOLD = 10
 
 		// Check if the user has completed enough tasks to no longer be considered a "new user"
@@ -185,13 +214,17 @@ export class Controller {
 			apiConfiguration,
 			autoApprovalSettings,
 			browserSettings,
-			chatSettings,
+			preferredLanguage,
+			openaiReasoningEffort,
+			mode,
+			strictPlanModeEnabled ?? false,
 			shellIntegrationTimeout,
 			terminalReuseEnabled ?? true,
 			terminalOutputLineLimit ?? 500,
 			defaultTerminalProfile ?? "default",
 			enableCheckpointsSetting ?? true,
 			await getCwd(getDesktopDir()),
+			this.cacheService,
 			task,
 			images,
 			files,
@@ -219,10 +252,6 @@ export class Controller {
 	 */
 	async handleWebviewMessage(message: WebviewMessage) {
 		switch (message.type) {
-			case "fetchMcpMarketplace": {
-				await this.fetchMcpMarketplace(message.bool)
-				break
-			}
 			case "grpc_request": {
 				if (message.grpc_request) {
 					await handleGrpcRequest(this, message.grpc_request)
@@ -235,9 +264,9 @@ export class Controller {
 				}
 				break
 			}
-
-			// Add more switch case statements here as more webview message commands
-			// are created within the webview context (i.e. inside media/main.js)
+			default: {
+				console.error("Received unhandled WebviewMessage type:", JSON.stringify(message))
+			}
 		}
 	}
 
@@ -248,28 +277,25 @@ export class Controller {
 		await this.postStateToWebview()
 	}
 
-	async togglePlanActModeWithChatSettings(chatSettings: ChatSettings, chatContent?: ChatContent): Promise<boolean> {
-		const didSwitchToActMode = chatSettings.mode === "act"
+	async togglePlanActMode(modeToSwitchTo: Mode, chatContent?: ChatContent): Promise<boolean> {
+		const didSwitchToActMode = modeToSwitchTo === "act"
 
 		// Store mode to global state
-		await updateGlobalState(this.context, "mode", chatSettings.mode)
+		await updateGlobalState(this.context, "mode", modeToSwitchTo)
 
 		// Capture mode switch telemetry | Capture regardless of if we know the taskId
-		telemetryService.captureModeSwitch(this.task?.taskId ?? "0", chatSettings.mode)
+		telemetryService.captureModeSwitch(this.task?.taskId ?? "0", modeToSwitchTo)
 
 		// Update API handler with new mode (buildApiHandler now selects provider based on mode)
 		if (this.task) {
-			const { apiConfiguration } = await getAllExtensionState(this.context)
-			this.task.api = buildApiHandler({ ...apiConfiguration, taskId: this.task.taskId }, chatSettings.mode)
+			const apiConfiguration = this.cacheService.getApiConfiguration()
+			this.task.api = buildApiHandler({ ...apiConfiguration, taskId: this.task.taskId }, modeToSwitchTo)
 		}
 
-		// Save only non-mode properties to global storage
-		const { mode, ...persistentChatSettings }: { mode: string } & StoredChatSettings = chatSettings
-		await updateGlobalState(this.context, "chatSettings", persistentChatSettings)
 		await this.postStateToWebview()
 
 		if (this.task) {
-			this.task.chatSettings = chatSettings
+			this.task.updateMode(modeToSwitchTo)
 			if (this.task.taskState.isAwaitingPlanResponse && didSwitchToActMode) {
 				this.task.taskState.didRespondToPlanAskBySwitchingMode = true
 				// Use chatContent if provided, otherwise use default message
@@ -321,7 +347,7 @@ export class Controller {
 
 	async handleAuthCallback(customToken: string, provider: string | null = null) {
 		try {
-			await this.authService.handleAuthCallback(customToken, provider ? provider : "google")
+			await AuthService.getInstance(this).handleAuthCallback(customToken, provider ? provider : "google")
 
 			const clineProvider: ApiProvider = "cline"
 
@@ -329,27 +355,26 @@ export class Controller {
 			const { planActSeparateModelsSetting } = await getAllExtensionState(this.context)
 			const currentMode = await this.getCurrentMode()
 
+			// Get current API configuration from cache
+			const currentApiConfiguration = this.cacheService.getApiConfiguration()
+
+			const updatedConfig = { ...currentApiConfiguration }
+
 			if (planActSeparateModelsSetting) {
 				// Only update the current mode's provider
 				if (currentMode === "plan") {
-					await updateGlobalState(this.context, "planModeApiProvider", clineProvider)
+					updatedConfig.planModeApiProvider = clineProvider
 				} else {
-					await updateGlobalState(this.context, "actModeApiProvider", clineProvider)
+					updatedConfig.actModeApiProvider = clineProvider
 				}
 			} else {
 				// Update both modes to keep them in sync
-				await Promise.all([
-					updateGlobalState(this.context, "planModeApiProvider", clineProvider),
-					updateGlobalState(this.context, "actModeApiProvider", clineProvider),
-				])
+				updatedConfig.planModeApiProvider = clineProvider
+				updatedConfig.actModeApiProvider = clineProvider
 			}
 
-			// Get the updated API configuration (now includes the updated providers)
-			const { apiConfiguration } = await getAllExtensionState(this.context)
-			const updatedConfig = {
-				...apiConfiguration,
-				apiProvider: clineProvider,
-			}
+			// Update the API configuration through cache service
+			this.cacheService.setApiConfiguration(updatedConfig)
 
 			// Mark welcome view as completed since user has successfully logged in
 			await updateGlobalState(this.context, "welcomeViewCompleted", true)
@@ -468,31 +493,6 @@ export class Controller {
 		}
 	}
 
-	private async fetchMcpMarketplace(forceRefresh: boolean = false) {
-		try {
-			// Check if we have cached data
-			const cachedCatalog = (await getGlobalState(this.context, "mcpMarketplaceCatalog")) as
-				| McpMarketplaceCatalog
-				| undefined
-			if (!forceRefresh && cachedCatalog?.items) {
-				await sendMcpMarketplaceCatalogEvent(cachedCatalog)
-				return
-			}
-
-			const catalog = await this.fetchMcpMarketplaceFromApi(false)
-			if (catalog) {
-				await sendMcpMarketplaceCatalogEvent(catalog)
-			}
-		} catch (error) {
-			console.error("Failed to handle cached MCP marketplace:", error)
-			const errorMessage = error instanceof Error ? error.message : "Failed to handle cached MCP marketplace"
-			HostProvider.window.showMessage({
-				type: ShowMessageType.ERROR,
-				message: errorMessage,
-			})
-		}
-	}
-
 	// OpenRouter
 
 	async handleOpenRouterCallback(code: string) {
@@ -511,21 +511,20 @@ export class Controller {
 
 		const openrouter: ApiProvider = "openrouter"
 		const currentMode = await this.getCurrentMode()
-		await Promise.all([
-			updateGlobalState(this.context, "planModeApiProvider", openrouter),
-			updateGlobalState(this.context, "actModeApiProvider", openrouter),
-		])
-		await storeSecret(this.context, "openRouterApiKey", apiKey)
+
+		// Update API configuration through cache service
+		const currentApiConfiguration = this.cacheService.getApiConfiguration()
+		const updatedConfig = {
+			...currentApiConfiguration,
+			planModeApiProvider: openrouter,
+			actModeApiProvider: openrouter,
+			openRouterApiKey: apiKey,
+		}
+		this.cacheService.setApiConfiguration(updatedConfig)
+
 		await this.postStateToWebview()
 		if (this.task) {
-			// Get the updated API configuration (now includes the updated providers)
-			const { apiConfiguration } = await getAllExtensionState(this.context)
-			const updatedConfig = {
-				...apiConfiguration,
-				openRouterApiKey: apiKey,
-				taskId: this.task.taskId,
-			}
-			this.task.api = buildApiHandler(updatedConfig, currentMode)
+			this.task.api = buildApiHandler({ ...updatedConfig, taskId: this.task.taskId }, currentMode)
 		}
 		// await this.postMessageToWebview({ type: "action", action: "settingsButtonClicked" }) // bad ux if user is on welcome
 	}
@@ -699,13 +698,18 @@ export class Controller {
 	}
 
 	async getStateToPostToWebview(): Promise<ExtensionState> {
+		// Get API configuration from cache for immediate access
+		const apiConfiguration = this.cacheService.getApiConfiguration()
+
 		const {
-			apiConfiguration,
 			lastShownAnnouncementId,
 			taskHistory,
 			autoApprovalSettings,
 			browserSettings,
-			chatSettings: storedChatSettings,
+			preferredLanguage,
+			openaiReasoningEffort,
+			mode,
+			strictPlanModeEnabled,
 			userInfo,
 			mcpMarketplaceEnabled,
 			mcpDisplayMode,
@@ -721,51 +725,51 @@ export class Controller {
 			welcomeViewCompleted,
 			mcpResponsesCollapsed,
 			terminalOutputLineLimit,
+			localClineRulesToggles,
+			localWindsurfRulesToggles,
+			localCursorRulesToggles,
+			localWorkflowToggles,
 		} = await getAllExtensionState(this.context)
 
-		// Get current mode using helper function
-		const currentMode = await this.getCurrentMode()
+		const currentTaskItem = this.task?.taskId ? (taskHistory || []).find((item) => item.id === this.task?.taskId) : undefined
+		const checkpointTrackerErrorMessage = this.task?.taskState.checkpointTrackerErrorMessage
+		const clineMessages = this.task?.messageStateHandler.getClineMessages() || []
 
-		// Reconstruct ChatSettings with mode from global state and stored preferences
-		const chatSettings: ChatSettings = {
-			...storedChatSettings, // Spread stored preferences (preferredLanguage, openAIReasoningEffort)
-			mode: currentMode, // Use mode from global state
-		}
+		const processedTaskHistory = (taskHistory || [])
+			.filter((item) => item.ts && item.task)
+			.sort((a, b) => b.ts - a.ts)
+			.slice(0, 100) // for now we're only getting the latest 100 tasks, but a better solution here is to only pass in 3 for recent task history, and then get the full task history on demand when going to the task history view (maybe with pagination?)
 
-		const localClineRulesToggles =
-			((await getWorkspaceState(this.context, "localClineRulesToggles")) as ClineRulesToggles) || {}
-
-		const localWindsurfRulesToggles =
-			((await getWorkspaceState(this.context, "localWindsurfRulesToggles")) as ClineRulesToggles) || {}
-
-		const localCursorRulesToggles =
-			((await getWorkspaceState(this.context, "localCursorRulesToggles")) as ClineRulesToggles) || {}
-
-		const localWorkflowToggles = ((await getWorkspaceState(this.context, "workflowToggles")) as ClineRulesToggles) || {}
+		const latestAnnouncementId = getLatestAnnouncementId(this.context)
+		const shouldShowAnnouncement = lastShownAnnouncementId !== latestAnnouncementId
+		const platform = process.platform as Platform
+		const distinctId = PostHogClientProvider.getInstance().distinctId
+		const version = this.context.extension?.packageJSON?.version ?? ""
+		const uriScheme = vscode.env.uriScheme
 
 		return {
-			version: this.context.extension?.packageJSON?.version ?? "",
+			version,
 			apiConfiguration,
-			uriScheme: vscode.env.uriScheme,
-			currentTaskItem: this.task?.taskId ? (taskHistory || []).find((item) => item.id === this.task?.taskId) : undefined,
-			checkpointTrackerErrorMessage: this.task?.taskState.checkpointTrackerErrorMessage,
-			clineMessages: this.task?.messageStateHandler.getClineMessages() || [],
-			taskHistory: (taskHistory || [])
-				.filter((item) => item.ts && item.task)
-				.sort((a, b) => b.ts - a.ts)
-				.slice(0, 100), // for now we're only getting the latest 100 tasks, but a better solution here is to only pass in 3 for recent task history, and then get the full task history on demand when going to the task history view (maybe with pagination?)
-			shouldShowAnnouncement: lastShownAnnouncementId !== this.latestAnnouncementId,
-			platform: process.platform as Platform,
+			uriScheme,
+			currentTaskItem,
+			checkpointTrackerErrorMessage,
+			clineMessages,
+			taskHistory: processedTaskHistory,
+			shouldShowAnnouncement,
+			platform,
 			autoApprovalSettings,
 			browserSettings,
-			chatSettings,
+			preferredLanguage,
+			openaiReasoningEffort,
+			mode,
+			strictPlanModeEnabled,
 			userInfo,
 			mcpMarketplaceEnabled,
 			mcpDisplayMode,
 			telemetrySetting,
 			planActSeparateModelsSetting,
 			enableCheckpointsSetting: enableCheckpointsSetting ?? true,
-			distinctId: telemetryService.distinctId,
+			distinctId,
 			globalClineRulesToggles: globalClineRulesToggles || {},
 			localClineRulesToggles: localClineRulesToggles || {},
 			localWindsurfRulesToggles: localWindsurfRulesToggles || {},
@@ -840,18 +844,4 @@ export class Controller {
 		await updateGlobalState(this.context, "taskHistory", history)
 		return history
 	}
-
-	// private async clearState() {
-	// 	this.context.workspaceState.keys().forEach((key) => {
-	// 		this.context.workspaceState.update(key, undefined)
-	// 	})
-	// 	this.context.globalState.keys().forEach((key) => {
-	// 		this.context.globalState.update(key, undefined)
-	// 	})
-	// 	this.context.secrets.delete("apiKey")
-	// }
-
-	// secrets
-
-	// dev
 }

src/core/controller/mcp/addRemoteMcpServer.ts

@@ -1,6 +1,6 @@
 import { convertMcpServersToProtoMcpServers } from "@/shared/proto-conversions/mcp/mcp-server-conversion"
-import type { AddRemoteMcpServerRequest } from "../../../shared/proto/mcp"
-import { McpServers } from "../../../shared/proto/mcp"
+import type { AddRemoteMcpServerRequest } from "@shared/proto/cline/mcp"
+import { McpServers } from "@shared/proto/cline/mcp"
 import type { Controller } from "../index"
 
 /**