fix proto request

* rename CacheService to StateManager

* fix types

.changeset/early-actors-marry.md

@@ -1,5 +0,0 @@
----
-"claude-dev": patch
----
-
-Fix Provider Options to properly handle retry attempts

.changeset/fluffy-toys-beam.md

@@ -0,0 +1,5 @@
+---
+"claude-dev": patch
+---
+
+Fix Ollama connection issue to default endpoint at port 11434

.changeset/funny-ants-travel.md

@@ -0,0 +1,5 @@
+---
+"claude-dev": patch
+---
+
+Optimized Cline for GPT-5 model family with an aligned system prompt

.changeset/gentle-jars-listen.md

@@ -1,5 +0,0 @@
----
-"claude-dev": patch
----
-
-fix issue with gpt-5 not using auto-compact even when toggled on

.changeset/hip-feet-approve.md

@@ -1,5 +0,0 @@
----
-"claude-dev": patch
----
-
-Make markdown and code block parsing styling reactive to theme change

.changeset/smart-flowers-deny.md

@@ -0,0 +1,5 @@
+---
+"claude-dev": patch
+---
+
+Dify.ai api integration

.changeset/tasty-rocks-sit.md

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

.changeset/yellow-cycles-join.md

@@ -0,0 +1,5 @@
+---
+"claude-dev": patch
+---
+
+Prompt changes for deep-planning in windows/powershell

.mocharc.json

@@ -3,8 +3,7 @@
 		"ts"
 	],
 	"spec": [
-		"src/**/__tests__/*.ts",
-		"eslint-rules/__tests__/**/*.test.ts"
+		"src/**/__tests__/*.ts"
 	],
 	"require": [
 		"ts-node/register",

CHANGELOG.md

@@ -1,5 +1,12 @@
 # Changelog
 
+## [3.26.6]
+
+- Add free Grok Coder model to Cline provider for users looking for a fast, free coding model option
+- Fix GPT-5 models not respecting auto-compact setting when enabled, improving context window management
+- Fix provider retry attempts not showing proper user feedback during rate limiting scenarios
+- Improve markdown and code block styling to automatically adapt when switching VS Code themes
+
 ## [3.26.5]
 
 - fix (provider/vercel-ai-gateway): reduce model list load frequency in settings view

docs/docs.json

@@ -90,6 +90,7 @@
 					"features/focus-chain",
 					"features/auto-compact",
 					"features/editing-messages",
+					"features/dictation",
 					{
 						"group": "@ Mentions",
 						"pages": [

docs/features/dictation.mdx

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

package-lock.json

@@ -1,12 +1,12 @@
 {
 	"name": "claude-dev",
-	"version": "3.26.5",
+	"version": "3.26.6",
 	"lockfileVersion": 3,
 	"requires": true,
 	"packages": {
 		"": {
 			"name": "claude-dev",
-			"version": "3.26.5",
+			"version": "3.26.6",
 			"license": "Apache-2.0",
 			"dependencies": {
 				"@anthropic-ai/sdk": "^0.37.0",

package.json

@@ -2,7 +2,7 @@
 	"name": "claude-dev",
 	"displayName": "Cline",
 	"description": "Autonomous coding agent right in your IDE, capable of creating/editing files, running commands, using the browser, and more with your permission every step of the way.",
-	"version": "3.26.5",
+	"version": "3.26.6",
 	"icon": "assets/icons/icon.png",
 	"engines": {
 		"vscode": "^1.84.0"

proto/cline/dictation.proto

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

proto/cline/models.proto

@@ -150,6 +150,7 @@ enum ApiProvider {
   ZAI = 31;
   VERCEL_AI_GATEWAY = 32;
   QWEN_CODE = 33;
+  DIFY = 34;
 }
 
 // Model info for OpenAI-compatible models
@@ -259,6 +260,8 @@ message ModelsApiConfiguration {
   optional string lm_studio_max_tokens = 67;
   optional string vercel_ai_gateway_api_key = 68;
   optional string qwen_code_oauth_path = 69;
+  optional string dify_api_key = 70;
+  optional string dify_base_url = 71;
 
   // Plan mode configurations
   optional ApiProvider plan_mode_api_provider = 100;

proto/cline/state.proto

@@ -123,9 +123,10 @@ message UpdateSettingsRequest {
   optional string preferred_language = 14;
   optional OpenaiReasoningEffort openai_reasoning_effort = 15;
   optional bool strict_plan_mode_enabled = 16;
-  optional FocusChainSettings focus_chain_settings = 17;
-  optional bool use_auto_condense = 18;
-  optional string custom_prompt = 19;
+  optional DictationSettings dictation_settings = 17;
+  optional FocusChainSettings focus_chain_settings = 18;
+  optional bool use_auto_condense = 19;
+  optional string custom_prompt = 20;
 }
 
 // Complete API Configuration message
@@ -194,6 +195,8 @@ message ApiConfiguration {
   optional string lm_studio_max_tokens = 61;
   optional string vercel_ai_gateway_api_key = 62;
   optional string qwen_code_oauth_path = 63;
+  optional string dify_api_key = 64;
+  optional string dify_base_url = 65;
   
   // Plan mode configurations
   optional string plan_mode_api_provider = 100;
@@ -261,6 +264,11 @@ message UpdateTerminalConnectionTimeoutRequest {
   optional int32 timeout_ms = 1;
 }
 
+message DictationSettings {
+  bool dictation_enabled = 1;
+  string dictation_language = 2;
+}
+
 message FocusChainSettings {
   bool enabled = 1;
   int32 remind_cline_interval = 2;

src/api/providers/dify.ts

@@ -0,0 +1,288 @@
+import { Anthropic } from "@anthropic-ai/sdk"
+import { ApiHandlerOptions, ModelInfo } from "@shared/api"
+import { ApiHandler } from "../../core/api/index"
+import { ApiStream } from "../../core/api/transform/stream"
+
+export class DifyHandler implements ApiHandler {
+	private options: ApiHandlerOptions
+	private baseUrl: string
+	private apiKey: string
+	private conversationId: string | null = null
+
+	constructor(options: ApiHandlerOptions) {
+		this.options = options
+		this.apiKey = options.difyApiKey || ""
+		this.baseUrl = options.difyBaseUrl || ""
+
+		console.log("[DIFY DEBUG] Constructor called with:", {
+			hasApiKey: !!this.apiKey,
+			baseUrl: this.baseUrl,
+		})
+
+		if (!this.apiKey) {
+			throw new Error("Dify API key is required")
+		}
+		if (!this.baseUrl) {
+			throw new Error("Dify base URL is required")
+		}
+	}
+
+	async *createMessage(systemPrompt: string, messages: Anthropic.Messages.MessageParam[]): ApiStream {
+		console.log("[DIFY DEBUG] createMessage called with:", {
+			systemPromptLength: systemPrompt?.length || 0,
+			messagesCount: messages?.length || 0,
+		})
+
+		// Convert messages to Dify format
+		const query = this.convertMessagesToQuery(systemPrompt, messages)
+		const requestBody = {
+			inputs: {},
+			query: query,
+			response_mode: "streaming",
+			conversation_id: this.conversationId || "",
+			user: "cline-user", // A unique user identifier
+			files: [],
+		}
+
+		const fullUrl = `${this.baseUrl}/chat-messages`
+		console.log("[DIFY DEBUG] Making request to:", fullUrl)
+		console.log("[DIFY DEBUG] Request body:", JSON.stringify(requestBody, null, 2))
+		console.log("[DIFY DEBUG] Current process environment variables (for proxy debugging):", process.env)
+
+		let response: Response
+		try {
+			response = await fetch(fullUrl, {
+				method: "POST",
+				headers: {
+					Authorization: `Bearer ${this.apiKey}`,
+					"Content-Type": "application/json",
+				},
+				body: JSON.stringify(requestBody),
+			})
+		} catch (error: any) {
+			console.error("[DIFY DEBUG] Network error during fetch:", error)
+			// Log more detailed error information if available (e.g., from undici)
+			const cause = error.cause ? ` | Cause: ${error.cause}` : ""
+			throw new Error(`Dify API network error: ${error.message}${cause}`)
+		}
+
+		console.log("[DIFY DEBUG] Response status:", response.status)
+		const headersObj: Record<string, string> = {}
+		response.headers.forEach((value, key) => {
+			headersObj[key] = value
+		})
+		console.log("[DIFY DEBUG] Response headers:", headersObj)
+
+		if (!response.ok) {
+			const errorText = await response.text()
+			console.error("[DIFY DEBUG] Error response:", errorText)
+			throw new Error(`Dify API error: ${response.status} ${response.statusText} - ${errorText}`)
+		}
+
+		if (!response.body) {
+			throw new Error("No response body from Dify API")
+		}
+
+		const reader = response.body.getReader()
+		const decoder = new TextDecoder()
+		let buffer = ""
+		let fullText = ""
+
+		console.log("[DIFY DEBUG] Starting to read streaming response...")
+
+		try {
+			while (true) {
+				const { done, value } = await reader.read()
+				if (done) {
+					console.log("[DIFY DEBUG] Stream ended naturally")
+					break
+				}
+
+				const chunk = decoder.decode(value, { stream: true })
+				console.log("[DIFY DEBUG] Raw chunk received:", JSON.stringify(chunk))
+
+				buffer += chunk
+				const lines = buffer.split("\n")
+
+				// Keep the last incomplete line in the buffer
+				buffer = lines.pop() || ""
+
+				for (const line of lines) {
+					console.log("[DIFY DEBUG] Processing line:", JSON.stringify(line))
+
+					if (line.startsWith("data: ")) {
+						const data = line.slice(6).trim()
+						console.log("[DIFY DEBUG] Extracted data:", JSON.stringify(data))
+
+						if (data === "[DONE]") {
+							console.log("[DIFY DEBUG] Received [DONE] signal")
+							return // Explicitly return on [DONE]
+						}
+
+						if (data === "") {
+							console.log("[DIFY DEBUG] Empty data line, skipping")
+							continue
+						}
+
+						try {
+							const parsed = JSON.parse(data)
+							console.log("[DIFY DEBUG] Parsed JSON:", parsed)
+
+							// Capture conversation_id as soon as it's available
+							if (parsed.conversation_id && !this.conversationId) {
+								this.conversationId = parsed.conversation_id
+								console.log("[DIFY DEBUG] Captured conversation_id:", this.conversationId)
+							}
+
+							// Handle different Dify event types based on actual Dify API
+							if (parsed.event === "message") {
+								console.log("[DIFY DEBUG] Message event, answer:", parsed.answer)
+								// Dify sends the full text in each "answer" chunk, so we replace.
+								if (typeof parsed.answer === "string") {
+									fullText = parsed.answer
+									console.log("[DIFY DEBUG] Updated fullText length:", fullText.length)
+									yield {
+										type: "text",
+										text: fullText,
+									}
+								}
+							} else if (parsed.event === "message_replace") {
+								console.log("[DIFY DEBUG] Replace message event:", parsed)
+								if (parsed.answer) {
+									fullText = parsed.answer // Replace instead of append
+									console.log("[DIFY DEBUG] Replaced fullText length:", fullText.length)
+									yield {
+										type: "text",
+										text: fullText,
+									}
+								}
+							} else if (parsed.event === "message_end") {
+								console.log("[DIFY DEBUG] Message end event", parsed)
+								// Message completed. Yield final text if we have any.
+								if (fullText) {
+									yield {
+										type: "text",
+										text: fullText,
+									}
+								}
+								// Yield usage data if available
+								if (parsed.usage) {
+									yield {
+										type: "usage",
+										inputTokens: parsed.usage.prompt_tokens || 0,
+										outputTokens: parsed.usage.completion_tokens || parsed.usage.total_tokens || 0,
+										totalCost: parsed.usage.total_price || 0,
+									}
+								}
+								return // End of stream
+							} else if (parsed.event === "error") {
+								console.error("[DIFY DEBUG] Error event:", parsed)
+								throw new Error(`Dify API error: ${parsed.message || "Unknown error"}`)
+							} else if (parsed.event === "workflow_started" || parsed.event === "workflow_finished") {
+								console.log("[DIFY DEBUG] Workflow event:", parsed.event)
+								// These are informational events, continue processing
+							} else if (parsed.event === "node_started" || parsed.event === "node_finished") {
+								console.log("[DIFY DEBUG] Node event:", parsed.event, parsed.data)
+								// These are informational events, continue processing
+							} else if (parsed.event === "ping") {
+								console.log("[DIFY DEBUG] Ping event received, keeping connection alive.")
+								// Ping event, do nothing
+							} else {
+								console.log("[DIFY DEBUG] Unknown event type:", parsed.event, "Full object:", parsed)
+								// Try to extract text from other possible fields
+								if (parsed.text) {
+									fullText += parsed.text
+									yield {
+										type: "text",
+										text: fullText,
+									}
+								} else if (parsed.content) {
+									fullText += parsed.content
+									yield {
+										type: "text",
+										text: fullText,
+									}
+								}
+							}
+						} catch (e) {
+							console.warn("[DIFY DEBUG] Failed to parse JSON:", data, "Error:", e)
+						}
+					} else if (line.trim() !== "") {
+						console.log(
+							"[DIFY DEBUG] Non-data line (not starting with 'data:'), trying to parse as direct JSON:",
+							JSON.stringify(line),
+						)
+						// Try to parse as direct JSON (fallback for non-SSE responses, though Dify uses SSE)
+						try {
+							const parsed = JSON.parse(line.trim())
+							console.log("[DIFY DEBUG] Parsed direct JSON:", parsed)
+
+							// Handle the same event types as above
+							if (parsed.event === "message" && parsed.answer) {
+								fullText += parsed.answer
+								yield {
+									type: "text",
+									text: fullText,
+								}
+							} else if (parsed.event === "message_end") {
+								if (fullText) {
+									yield {
+										type: "text",
+										text: fullText,
+									}
+								}
+								return
+							} else if (parsed.event === "error") {
+								console.error("[DIFY DEBUG] Direct JSON Error event:", parsed)
+								throw new Error(`Dify API error: ${parsed.message || "Unknown error"}`)
+							}
+						} catch (e) {
+							// Not JSON, continue
+							console.log("[DIFY DEBUG] Line is not direct JSON, continuing")
+						}
+					}
+				}
+			}
+		} finally {
+			reader.releaseLock()
+			console.log("[DIFY DEBUG] Stream reader released")
+		}
+	}
+
+	private convertMessagesToQuery(systemPrompt: string, messages: Anthropic.Messages.MessageParam[]): string {
+		// Dify's context is managed by `conversation_id`. The `query` should be the last user message.
+		// The system prompt is typically configured in the Dify App itself.
+		const lastUserMessage = messages.filter((m) => m.role === "user").pop()
+
+		if (!lastUserMessage) {
+			return "" // Should not happen in normal flow
+		}
+
+		const userQuery = Array.isArray(lastUserMessage.content)
+			? lastUserMessage.content.map((c) => ("text" in c ? c.text : "")).join("\n")
+			: (lastUserMessage.content as string)
+
+		// Only prepend the system prompt if it's the very first message of a new conversation.
+		if (!this.conversationId && systemPrompt) {
+			console.log("[DIFY DEBUG] Prepending system prompt for new conversation.")
+			return `${systemPrompt}\n\n---\n\n${userQuery}`
+		}
+
+		return userQuery
+	}
+
+	getModel(): { id: string; info: ModelInfo } {
+		return {
+			id: "dify-workflow",
+			info: {
+				maxTokens: 8192,
+				contextWindow: 128000,
+				supportsImages: true,
+				supportsPromptCache: false,
+				inputPrice: 0,
+				outputPrice: 0,
+				description: "Dify workflow - model selection is configured in your Dify application",
+			},
+		}
+	}
+}

src/core/api/index.ts

@@ -9,6 +9,7 @@ import { CerebrasHandler } from "./providers/cerebras"
 import { ClaudeCodeHandler } from "./providers/claude-code"
 import { ClineHandler } from "./providers/cline"
 import { DeepSeekHandler } from "./providers/deepseek"
+import { DifyHandler } from "./providers/dify"
 import { DoubaoHandler } from "./providers/doubao"
 import { FireworksHandler } from "./providers/fireworks"
 import { GeminiHandler } from "./providers/gemini"
@@ -345,6 +346,15 @@ function createHandlerForProvider(
 				huaweiCloudMaasModelInfo:
 					mode === "plan" ? options.planModeHuaweiCloudMaasModelInfo : options.actModeHuaweiCloudMaasModelInfo,
 			})
+		case "dify": // Add Dify.ai handler
+			console.log("[DIFY DEBUG] Instantiating DifyHandler with options:", {
+				difyApiKeyPresent: !!options.difyApiKey,
+				difyBaseUrl: options.difyBaseUrl,
+			})
+			return new DifyHandler({
+				difyApiKey: options.difyApiKey,
+				difyBaseUrl: options.difyBaseUrl,
+			})
 		case "vercel-ai-gateway":
 			return new VercelAIGatewayHandler({
 				vercelAiGatewayApiKey: options.vercelAiGatewayApiKey,

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

@@ -2,7 +2,7 @@ import type { Anthropic } from "@anthropic-ai/sdk"
 import { filterMessagesForClaudeCode } from "@/integrations/claude-code/message-filter"
 import { runClaudeCode } from "@/integrations/claude-code/run"
 import { ClaudeCodeModelId, claudeCodeDefaultModelId, claudeCodeModels } from "@/shared/api"
-import { CommonApiHandlerOptions, type ApiHandler } from ".."
+import { type ApiHandler, CommonApiHandlerOptions } from ".."
 import { withRetry } from "../retry"
 import { type ApiStream, ApiStreamUsageChunk } from "../transform/stream"
 

src/core/api/providers/dify.ts

@@ -0,0 +1,655 @@
+import { Anthropic } from "@anthropic-ai/sdk"
+import { ApiHandlerOptions, ModelInfo } from "../../../shared/api"
+import { ApiHandler } from "../index"
+import { ApiStream } from "../transform/stream"
+
+// Dify API Response Types
+export interface DifyFileResponse {
+	id: string
+	name: string
+	size: number
+	extension: string
+	mime_type: string
+	created_by: string
+	created_at: number
+}
+
+export interface DifyMessage {
+	id: string
+	conversation_id: string
+	inputs: Record<string, any>
+	query: string
+	message_files: Array<{
+		id: string
+		type: string
+		url: string
+		belongs_to: string
+	}>
+	answer: string
+	created_at: number
+	feedback?: {
+		rating: string
+	}
+	retriever_resources?: any[]
+}
+
+interface DifyHistoryResponse {
+	data: DifyMessage[]
+	has_more: boolean
+	limit: number
+}
+
+interface DifyConversation {
+	id: string
+	name: string
+	inputs: Record<string, any>
+	status: string
+	introduction: string
+	created_at: number
+	updated_at: number
+}
+
+interface DifyConversationsResponse {
+	data: DifyConversation[]
+	has_more: boolean
+	limit: number
+}
+
+interface DifyConversationResponse {
+	id: string
+	name: string
+	inputs: Record<string, any>
+	status: string
+	introduction: string
+	created_at: number
+	updated_at: number
+}
+
+export class DifyHandler implements ApiHandler {
+	private options: ApiHandlerOptions
+	private baseUrl: string
+	private apiKey: string
+	private conversationId: string | null = null
+	private currentTaskId: string | null = null
+	private abortController: AbortController | null = null
+
+	constructor(options: ApiHandlerOptions) {
+		this.options = options
+		this.apiKey = options.difyApiKey || ""
+		this.baseUrl = options.difyBaseUrl || ""
+
+		console.log("[DIFY DEBUG] Constructor called with:", {
+			hasApiKey: !!this.apiKey,
+			baseUrl: this.baseUrl,
+		})
+
+		if (!this.apiKey) {
+			throw new Error("Dify API key is required")
+		}
+		if (!this.baseUrl) {
+			throw new Error("Dify base URL is required")
+		}
+	}
+
+	async *createMessage(systemPrompt: string, messages: Anthropic.Messages.MessageParam[]): ApiStream {
+		console.log("[DIFY DEBUG] createMessage called with:", {
+			systemPromptLength: systemPrompt?.length || 0,
+			messagesCount: messages?.length || 0,
+		})
+
+		// Convert messages to Dify format
+		const query = this.convertMessagesToQuery(systemPrompt, messages)
+		const requestBody = {
+			inputs: {},
+			query: query,
+			response_mode: "streaming",
+			conversation_id: this.conversationId || "",
+			user: "cline-user", // A unique user identifier
+			files: [],
+		}
+
+		const fullUrl = `${this.baseUrl}/chat-messages`
+		console.log("[DIFY DEBUG] Making request to:", fullUrl)
+		console.log("[DIFY DEBUG] Request body:", JSON.stringify(requestBody, null, 2))
+
+		let response: Response
+		try {
+			response = await fetch(fullUrl, {
+				method: "POST",
+				headers: {
+					Authorization: `Bearer ${this.apiKey}`,
+					"Content-Type": "application/json",
+				},
+				body: JSON.stringify(requestBody),
+			})
+		} catch (error: any) {
+			console.error("[DIFY DEBUG] Network error during fetch:", error)
+			const cause = error.cause ? ` | Cause: ${error.cause}` : ""
+			throw new Error(`Dify API network error: ${error.message}${cause}`)
+		}
+
+		console.log("[DIFY DEBUG] Response status:", response.status)
+		const headersObj: Record<string, string> = {}
+		response.headers.forEach((value, key) => {
+			headersObj[key] = value
+		})
+		console.log("[DIFY DEBUG] Response headers:", headersObj)
+
+		if (!response.ok) {
+			const errorText = await response.text()
+			console.error("[DIFY DEBUG] Error response:", errorText)
+			throw new Error(`Dify API error: ${response.status} ${response.statusText} - ${errorText}`)
+		}
+
+		if (!response.body) {
+			throw new Error("No response body from Dify API")
+		}
+
+		const reader = response.body.getReader()
+		const decoder = new TextDecoder()
+		let buffer = ""
+		let fullText = ""
+		let hasYieldedContent = false
+		const processedEvents: string[] = []
+		let lastEventTime = Date.now()
+
+		console.log("[DIFY DEBUG] Starting to read streaming response...")
+
+		try {
+			while (true) {
+				const { done, value } = await reader.read()
+				if (done) {
+					console.log("[DIFY DEBUG] Stream ended naturally")
+					console.log(
+						"[DIFY DEBUG] Final state - hasYieldedContent:",
+						hasYieldedContent,
+						"fullText length:",
+						fullText.length,
+						"processedEvents:",
+						processedEvents,
+					)
+					break
+				}
+
+				const chunk = decoder.decode(value, { stream: true })
+				console.log("[DIFY DEBUG] Raw chunk received:", JSON.stringify(chunk))
+
+				buffer += chunk
+				const lines = buffer.split("\n")
+
+				// Keep the last incomplete line in the buffer
+				buffer = lines.pop() || ""
+
+				for (const line of lines) {
+					console.log("[DIFY DEBUG] Processing line:", JSON.stringify(line))
+
+					if (line.startsWith("data: ")) {
+						const data = line.slice(6).trim()
+						console.log("[DIFY DEBUG] Extracted data:", JSON.stringify(data))
+
+						if (data === "[DONE]") {
+							console.log("[DIFY DEBUG] Received [DONE] signal")
+							break
+						}
+
+						if (data === "") {
+							console.log("[DIFY DEBUG] Empty data line, skipping")
+							continue
+						}
+
+						try {
+							const parsed = JSON.parse(data)
+							console.log("[DIFY DEBUG] Parsed JSON:", parsed)
+							processedEvents.push(parsed.event || "unknown")
+							lastEventTime = Date.now()
+
+							// Capture conversation_id as soon as it's available
+							if (parsed.conversation_id && !this.conversationId) {
+								this.conversationId = parsed.conversation_id
+								console.log("[DIFY DEBUG] Captured conversation_id:", this.conversationId)
+							}
+
+							// Handle different Dify event types based on actual Dify API
+							if (parsed.event === "message") {
+								console.log("[DIFY DEBUG] Message event, answer:", parsed.answer)
+								// Dify sends the full text in each "answer" chunk, so we replace.
+								if (typeof parsed.answer === "string") {
+									fullText = parsed.answer
+									console.log("[DIFY DEBUG] Updated fullText length:", fullText.length)
+									yield {
+										type: "text",
+										text: fullText,
+									}
+									hasYieldedContent = true
+								}
+							} else if (parsed.event === "message_replace") {
+								console.log("[DIFY DEBUG] Replace message event:", parsed)
+								if (parsed.answer) {
+									fullText = parsed.answer // Replace instead of append
+									console.log("[DIFY DEBUG] Replaced fullText length:", fullText.length)
+									yield {
+										type: "text",
+										text: fullText,
+									}
+									hasYieldedContent = true
+								}
+							} else if (parsed.event === "message_end") {
+								console.log("[DIFY DEBUG] Message end event", parsed)
+								// Message completed. Yield final text if we have any.
+								if (fullText) {
+									yield {
+										type: "text",
+										text: fullText,
+									}
+									hasYieldedContent = true
+								}
+								// Yield usage data if available
+								if (parsed.usage) {
+									yield {
+										type: "usage",
+										inputTokens: parsed.usage.prompt_tokens || 0,
+										outputTokens: parsed.usage.completion_tokens || parsed.usage.total_tokens || 0,
+										totalCost: parsed.usage.total_price || 0,
+									}
+								}
+								return // End of stream
+							} else if (parsed.event === "error") {
+								console.error("[DIFY DEBUG] Error event:", parsed)
+								throw new Error(`Dify API error: ${parsed.message || "Unknown error"}`)
+							} else if (parsed.event === "workflow_started" || parsed.event === "workflow_finished") {
+								console.log("[DIFY DEBUG] Workflow event:", parsed.event)
+								// These are informational events, continue processing
+							} else if (parsed.event === "node_started" || parsed.event === "node_finished") {
+								console.log("[DIFY DEBUG] Node event:", parsed.event, parsed.data)
+								// These are informational events, continue processing
+							} else if (parsed.event === "ping") {
+								console.log("[DIFY DEBUG] Ping event received, keeping connection alive.")
+								// Ping event, do nothing
+							} else {
+								console.log("[DIFY DEBUG] Unknown event type:", parsed.event, "Full object:", parsed)
+								// Try to extract text from other possible fields
+								if (parsed.text) {
+									fullText += parsed.text
+									yield {
+										type: "text",
+										text: fullText,
+									}
+									hasYieldedContent = true
+								} else if (parsed.content) {
+									fullText += parsed.content
+									yield {
+										type: "text",
+										text: fullText,
+									}
+									hasYieldedContent = true
+								} else if (parsed.answer) {
+									// Fallback: some events might have answer field even if not "message" type
+									fullText += parsed.answer
+									yield {
+										type: "text",
+										text: fullText,
+									}
+									hasYieldedContent = true
+								}
+							}
+						} catch (e) {
+							console.warn("[DIFY DEBUG] Failed to parse JSON:", data, "Error:", e)
+						}
+					} else if (line.trim() !== "") {
+						console.log(
+							"[DIFY DEBUG] Non-data line (not starting with 'data:'), trying to parse as direct JSON:",
+							JSON.stringify(line),
+						)
+						// Try to parse as direct JSON (fallback for non-SSE responses, though Dify uses SSE)
+						try {
+							const parsed = JSON.parse(line.trim())
+							console.log("[DIFY DEBUG] Parsed direct JSON:", parsed)
+							processedEvents.push(parsed.event || "direct-json")
+
+							// Handle the same event types as above
+							if (parsed.event === "message" && parsed.answer) {
+								fullText += parsed.answer
+								yield {
+									type: "text",
+									text: fullText,
+								}
+								hasYieldedContent = true
+							} else if (parsed.event === "message_end") {
+								if (fullText) {
+									yield {
+										type: "text",
+										text: fullText,
+									}
+									hasYieldedContent = true
+								}
+								return
+							} else if (parsed.event === "error") {
+								console.error("[DIFY DEBUG] Direct JSON Error event:", parsed)
+								throw new Error(`Dify API error: ${parsed.message || "Unknown error"}`)
+							} else if (parsed.answer || parsed.text || parsed.content) {
+								// Fallback for any content in direct JSON
+								const content = parsed.answer || parsed.text || parsed.content
+								fullText += content
+								yield {
+									type: "text",
+									text: fullText,
+								}
+								hasYieldedContent = true
+							}
+						} catch (e) {
+							// Not JSON, continue
+							console.log("[DIFY DEBUG] Line is not direct JSON, continuing")
+						}
+					}
+				}
+			}
+
+			// Final check - if we haven't yielded any content, provide diagnostic information
+			if (!hasYieldedContent) {
+				const diagnosticInfo = {
+					processedEvents,
+					finalFullTextLength: fullText.length,
+					finalFullText: fullText,
+					streamDuration: Date.now() - lastEventTime,
+					conversationId: this.conversationId,
+				}
+				console.error("[DIFY DEBUG] No content was yielded! Diagnostic info:", diagnosticInfo)
+
+				// If we have any accumulated text at all, yield it as a fallback
+				if (fullText.trim()) {
+					console.log("[DIFY DEBUG] Yielding accumulated text as fallback:", fullText)
+					yield {
+						type: "text",
+						text: fullText,
+					}
+				} else {
+					// Provide a more informative error
+					throw new Error(
+						`Dify API did not provide any assistant messages. ` +
+							`Events processed: [${processedEvents.join(", ")}]. ` +
+							`Check your Dify application configuration and ensure it's properly set up to return responses. ` +
+							`API URL: ${fullUrl}. Conversation ID: ${this.conversationId || "none"}.`,
+					)
+				}
+			}
+		} finally {
+			reader.releaseLock()
+			console.log("[DIFY DEBUG] Stream reader released")
+		}
+	}
+
+	private convertMessagesToQuery(systemPrompt: string, messages: Anthropic.Messages.MessageParam[]): string {
+		// Dify's context is managed by `conversation_id`. The `query` should be the last user message.
+		// The system prompt is typically configured in the Dify App itself.
+		const lastUserMessage = messages.filter((m) => m.role === "user").pop()
+
+		if (!lastUserMessage) {
+			return "" // Should not happen in normal flow
+		}
+
+		const userQuery = Array.isArray(lastUserMessage.content)
+			? lastUserMessage.content.map((c) => ("text" in c ? c.text : "")).join("\n")
+			: (lastUserMessage.content as string)
+
+		// Only prepend the system prompt if it's the very first message of a new conversation.
+		if (!this.conversationId && systemPrompt) {
+			console.log("[DIFY DEBUG] Prepending system prompt for new conversation.")
+			return `${systemPrompt}\n\n---\n\n${userQuery}`
+		}
+
+		return userQuery
+	}
+
+	getModel(): { id: string; info: ModelInfo } {
+		return {
+			id: "dify-workflow",
+			info: {
+				maxTokens: 8192,
+				contextWindow: 128000,
+				supportsImages: true,
+				supportsPromptCache: false,
+				inputPrice: 0,
+				outputPrice: 0,
+				description: "Dify workflow - model selection is configured in your Dify application",
+			},
+		}
+	}
+
+	// Additional Dify API Methods
+
+	/**
+	 * Upload a file for use in conversations
+	 * @param file File buffer to upload
+	 * @param filename Name of the file
+	 * @param user User identifier (defaults to "cline-user")
+	 * @returns Promise with file upload response
+	 */
+	async uploadFile(file: Buffer, filename: string, user: string = "cline-user"): Promise<DifyFileResponse> {
+		const formData = new FormData()
+		formData.append("file", new Blob([new Uint8Array(file)]), filename)
+		formData.append("user", user)
+
+		const response = await fetch(`${this.baseUrl}/files/upload`, {
+			method: "POST",
+			headers: {
+				Authorization: `Bearer ${this.apiKey}`,
+			},
+			body: formData,
+		})
+
+		if (!response.ok) {
+			const errorText = await response.text()
+			throw new Error(`Dify file upload error: ${response.status} ${response.statusText} - ${errorText}`)
+		}
+
+		return response.json()
+	}
+
+	/**
+	 * Stop generation for a specific task
+	 * @param taskId Task ID from streaming response
+	 * @param user User identifier (defaults to "cline-user")
+	 * @returns Promise that resolves when generation is stopped
+	 */
+	async stopGeneration(taskId: string, user: string = "cline-user"): Promise<void> {
+		const response = await fetch(`${this.baseUrl}/chat-messages/${taskId}/stop`, {
+			method: "POST",
+			headers: {
+				Authorization: `Bearer ${this.apiKey}`,
+				"Content-Type": "application/json",
+			},
+			body: JSON.stringify({ user }),
+		})
+
+		if (!response.ok) {
+			const errorText = await response.text()
+			throw new Error(`Dify stop generation error: ${response.status} ${response.statusText} - ${errorText}`)
+		}
+	}
+
+	/**
+	 * Get conversation history messages with pagination
+	 * @param conversationId Conversation ID
+	 * @param user User identifier (defaults to "cline-user")
+	 * @param firstId First message ID for pagination (optional)
+	 * @param limit Number of messages to return (default: 20)
+	 * @returns Promise with conversation history
+	 */
+	async getConversationHistory(
+		conversationId: string,
+		user: string = "cline-user",
+		firstId?: string,
+		limit: number = 20,
+	): Promise<DifyHistoryResponse> {
+		const params = new URLSearchParams({ user, limit: limit.toString() })
+		if (firstId) {
+			params.append("first_id", firstId)
+		}
+
+		const response = await fetch(`${this.baseUrl}/conversations/${conversationId}/messages?${params}`, {
+			headers: {
+				Authorization: `Bearer ${this.apiKey}`,
+			},
+		})
+
+		if (!response.ok) {
+			const errorText = await response.text()
+			throw new Error(`Dify get conversation history error: ${response.status} ${response.statusText} - ${errorText}`)
+		}
+
+		return response.json()
+	}
+
+	/**
+	 * Get list of conversations for a user
+	 * @param user User identifier (defaults to "cline-user")
+	 * @param lastId Last conversation ID for pagination (optional)
+	 * @param limit Number of conversations to return (default: 20)
+	 * @param sortBy Sort field (default: "-updated_at")
+	 * @returns Promise with conversations list
+	 */
+	async getConversations(
+		user: string = "cline-user",
+		lastId?: string,
+		limit: number = 20,
+		sortBy: string = "-updated_at",
+	): Promise<DifyConversationsResponse> {
+		const params = new URLSearchParams({
+			user,
+			limit: limit.toString(),
+			sort_by: sortBy,
+		})
+		if (lastId) {
+			params.append("last_id", lastId)
+		}
+
+		const response = await fetch(`${this.baseUrl}/conversations?${params}`, {
+			headers: {
+				Authorization: `Bearer ${this.apiKey}`,
+			},
+		})
+
+		if (!response.ok) {
+			const errorText = await response.text()
+			throw new Error(`Dify get conversations error: ${response.status} ${response.statusText} - ${errorText}`)
+		}
+
+		return response.json()
+	}
+
+	/**
+	 * Delete a conversation
+	 * @param conversationId Conversation ID to delete
+	 * @param user User identifier (defaults to "cline-user")
+	 * @returns Promise that resolves when conversation is deleted
+	 */
+	async deleteConversation(conversationId: string, user: string = "cline-user"): Promise<void> {
+		const response = await fetch(`${this.baseUrl}/conversations/${conversationId}`, {
+			method: "DELETE",
+			headers: {
+				Authorization: `Bearer ${this.apiKey}`,
+				"Content-Type": "application/json",
+			},
+			body: JSON.stringify({ user }),
+		})
+
+		if (!response.ok) {
+			const errorText = await response.text()
+			throw new Error(`Dify delete conversation error: ${response.status} ${response.statusText} - ${errorText}`)
+		}
+	}
+
+	/**
+	 * Rename a conversation
+	 * @param conversationId Conversation ID to rename
+	 * @param user User identifier (defaults to "cline-user")
+	 * @param name New conversation name (optional if auto_generate is true)
+	 * @param autoGenerate Whether to auto-generate the name (default: false)
+	 * @returns Promise with updated conversation details
+	 */
+	async renameConversation(
+		conversationId: string,
+		user: string = "cline-user",
+		name?: string,
+		autoGenerate: boolean = false,
+	): Promise<DifyConversationResponse> {
+		const body: any = { user, auto_generate: autoGenerate }
+		if (name) {
+			body.name = name
+		}
+
+		const response = await fetch(`${this.baseUrl}/conversations/${conversationId}/name`, {
+			method: "POST",
+			headers: {
+				Authorization: `Bearer ${this.apiKey}`,
+				"Content-Type": "application/json",
+			},
+			body: JSON.stringify(body),
+		})
+
+		if (!response.ok) {
+			const errorText = await response.text()
+			throw new Error(`Dify rename conversation error: ${response.status} ${response.statusText} - ${errorText}`)
+		}
+
+		return response.json()
+	}
+
+	/**
+	 * Submit feedback for a message
+	 * @param messageId Message ID to provide feedback for
+	 * @param rating Rating: "like" or "dislike"
+	 * @param content Optional feedback content
+	 * @param user User identifier (defaults to "cline-user")
+	 * @returns Promise that resolves when feedback is submitted
+	 */
+	async submitMessageFeedback(
+		messageId: string,
+		rating: "like" | "dislike",
+		content?: string,
+		user: string = "cline-user",
+	): Promise<void> {
+		const body: any = { rating, user }
+		if (content) {
+			body.content = content
+		}
+
+		const response = await fetch(`${this.baseUrl}/messages/${messageId}/feedbacks`, {
+			method: "POST",
+			headers: {
+				Authorization: `Bearer ${this.apiKey}`,
+				"Content-Type": "application/json",
+			},
+			body: JSON.stringify(body),
+		})
+
+		if (!response.ok) {
+			const errorText = await response.text()
+			throw new Error(`Dify submit feedback error: ${response.status} ${response.statusText} - ${errorText}`)
+		}
+	}
+
+	/**
+	 * Get current conversation ID
+	 * @returns Current conversation ID or null
+	 */
+	getCurrentConversationId(): string | null {
+		return this.conversationId
+	}
+
+	/**
+	 * Set conversation ID for continuing existing conversations
+	 * @param conversationId Conversation ID to set
+	 */
+	setConversationId(conversationId: string): void {
+		this.conversationId = conversationId
+	}
+
+	/**
+	 * Reset conversation ID to start a new conversation
+	 */
+	resetConversation(): void {
+		this.conversationId = null
+		this.currentTaskId = null
+	}
+}

src/core/api/providers/ollama.ts

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

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

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

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

@@ -152,17 +152,6 @@ export async function createOpenRouterStream(
 			}
 	}
 
-	// Removes messages in the middle when close to context window limit. Should not be applied to models that support prompt caching since it would continuously break the cache.
-	let shouldApplyMiddleOutTransform = !model.info.supportsPromptCache
-	// except for deepseek (which we set supportsPromptCache to true for), where because the context window is so small our truncation algo might miss and we should use openrouter's middle-out transform as a fallback to ensure we don't exceed the context window (FIXME: once we have a more robust token estimator we should not rely on this)
-	if (model.id === "deepseek/deepseek-chat") {
-		shouldApplyMiddleOutTransform = true
-	}
-
-	if (isGPT5ModelFamily(model.id)) {
-		shouldApplyMiddleOutTransform = false
-	}
-
 	// hardcoded provider sorting for kimi-k2
 	const isKimiK2 = model.id === "moonshotai/kimi-k2"
 	openRouterProviderSorting = isKimiK2 ? undefined : openRouterProviderSorting
@@ -176,7 +165,6 @@ export async function createOpenRouterStream(
 		messages: openAiMessages,
 		stream: true,
 		stream_options: { include_usage: true },
-		transforms: shouldApplyMiddleOutTransform ? ["middle-out"] : undefined,
 		include_reasoning: true,
 		...(model.id.startsWith("openai/o") ? { reasoning_effort: reasoningEffort || "medium" } : {}),
 		...(reasoning ? { reasoning } : {}),

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

@@ -243,7 +243,7 @@ export class FileContextTracker {
 			const key = `pendingFileContextWarning_${this.taskId}`
 			// NOTE: Using 'as any' because dynamic keys like pendingFileContextWarning_${taskId}
 			// are legitimate workspace state keys but don't fit the strict LocalStateKey type system
-			this.controller.cacheService.setWorkspaceState(key as any, files)
+			this.controller.stateManager.setWorkspaceState(key as any, files)
 		} catch (error) {
 			console.error("Error storing pending file context warning:", error)
 		}
@@ -255,7 +255,7 @@ export class FileContextTracker {
 	async retrievePendingFileContextWarning(): Promise<string[] | undefined> {
 		try {
 			const key = `pendingFileContextWarning_${this.taskId}`
-			const files = this.controller.cacheService.getWorkspaceStateKey(key as any) as string[]
+			const files = this.controller.stateManager.getWorkspaceStateKey(key as any) as string[]
 			return files
 		} catch (error) {
 			console.error("Error retrieving pending file context warning:", error)
@@ -270,7 +270,7 @@ export class FileContextTracker {
 		try {
 			const files = await this.retrievePendingFileContextWarning()
 			if (files) {
-				this.controller.cacheService.setWorkspaceState(`pendingFileContextWarning_${this.taskId}` as any, undefined)
+				this.controller.stateManager.setWorkspaceState(`pendingFileContextWarning_${this.taskId}` as any, undefined)
 				return files
 			}
 		} catch (error) {

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

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

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

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

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

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

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

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

src/core/controller/account/authStateChanged.ts

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

src/core/controller/browser/discoverBrowser.ts

@@ -19,7 +19,7 @@ export async function discoverBrowser(controller: Controller, _request: EmptyReq
 			// This way we don't override the user's preference
 
 			// Test the connection to get the endpoint
-			const browserSettings = controller.cacheService.getGlobalStateKey("browserSettings")
+			const browserSettings = controller.stateManager.getGlobalStateKey("browserSettings")
 			const browserSession = new BrowserSession(controller.context, browserSettings)
 			const result = await browserSession.testConnection(discoveredHost)
 

src/core/controller/browser/getBrowserConnectionInfo.ts

@@ -11,7 +11,7 @@ import { Controller } from "../index"
 export async function getBrowserConnectionInfo(controller: Controller, _: EmptyRequest): Promise<BrowserConnectionInfo> {
 	try {
 		// Get browser settings from extension state
-		const browserSettings = controller.cacheService.getGlobalStateKey("browserSettings")
+		const browserSettings = controller.stateManager.getGlobalStateKey("browserSettings")
 
 		// Check if there's an active browser session by using the controller's handleWebviewMessage approach
 		// This is similar to what's done in controller/index.ts for the "getBrowserConnectionInfo" message

src/core/controller/browser/getDetectedChromePath.ts

@@ -11,7 +11,7 @@ import { Controller } from "../index"
  */
 export async function getDetectedChromePath(controller: Controller, _: EmptyRequest): Promise<ChromePath> {
 	try {
-		const browserSettings = controller.cacheService.getGlobalStateKey("browserSettings")
+		const browserSettings = controller.stateManager.getGlobalStateKey("browserSettings")
 		const browserSession = new BrowserSession(controller.context, browserSettings)
 		const result = await browserSession.getDetectedChromePath()
 

src/core/controller/browser/testBrowserConnection.ts

@@ -12,7 +12,7 @@ import { Controller } from "../index"
  */
 export async function testBrowserConnection(controller: Controller, request: StringRequest): Promise<BrowserConnection> {
 	try {
-		const browserSettings = controller.cacheService.getGlobalStateKey("browserSettings")
+		const browserSettings = controller.stateManager.getGlobalStateKey("browserSettings")
 		const browserSession = new BrowserSession(controller.context, browserSettings)
 		const text = request.value || ""
 

src/core/controller/browser/updateBrowserSettings.ts

@@ -12,7 +12,7 @@ import { Controller } from "../index"
 export async function updateBrowserSettings(controller: Controller, request: UpdateBrowserSettingsRequest): Promise<Boolean> {
 	try {
 		// Get current browser settings to preserve fields not in the request
-		const currentSettings = controller.cacheService.getGlobalStateKey("browserSettings")
+		const currentSettings = controller.stateManager.getGlobalStateKey("browserSettings")
 		const mergedWithDefaults = { ...DEFAULT_BROWSER_SETTINGS, ...currentSettings }
 
 		// Convert from protobuf format to shared format, merging with existing settings
@@ -39,7 +39,7 @@ export async function updateBrowserSettings(controller: Controller, request: Upd
 		}
 
 		// Update global state with new settings
-		controller.cacheService.setGlobalState("browserSettings", newBrowserSettings)
+		controller.stateManager.setGlobalState("browserSettings", newBrowserSettings)
 
 		// Update task browser settings if task exists
 		if (controller.task) {

src/core/controller/dictation/cancelRecording.ts

@@ -0,0 +1,35 @@
+import { telemetryService } from "@services/posthog/PostHogClientProvider"
+import { RecordingResult } from "@shared/proto/cline/dictation"
+import { audioRecordingService } from "@/services/dictation/AudioRecordingService"
+import { Controller } from ".."
+
+/**
+ * Cancels audio recording without saving or transcribing the audio
+ * @param controller The controller instance
+ * @returns RecordingResult indicating success or failure
+ */
+export const cancelRecording = async (controller: Controller): Promise<RecordingResult> => {
+	const taskId = controller.task?.taskId
+	const recordingStatus = audioRecordingService.getRecordingStatus()
+	const recordingDuration = recordingStatus.durationSeconds * 1000 // Convert to milliseconds
+
+	try {
+		const result = await audioRecordingService.cancelRecording()
+
+		telemetryService.captureVoiceRecordingStopped(taskId, recordingDuration, false, process.platform)
+
+		return RecordingResult.create({
+			success: result.success,
+			error: result.error || "",
+		})
+	} catch (error) {
+		console.error("Error canceling recording:", error)
+
+		telemetryService.captureVoiceRecordingStopped(taskId, recordingDuration, false, process.platform)
+
+		return RecordingResult.create({
+			success: false,
+			error: error instanceof Error ? error.message : "Unknown error occurred",
+		})
+	}
+}

src/core/controller/dictation/getRecordingStatus.ts

@@ -0,0 +1,25 @@
+import { RecordingStatus } from "@shared/proto/cline/dictation"
+import { audioRecordingService } from "@/services/dictation/AudioRecordingService"
+
+/**
+ * Gets the current recording status
+ * @returns RecordingStatus with current status
+ */
+export const getRecordingStatus = async (): Promise<RecordingStatus> => {
+	try {
+		const status = audioRecordingService.getRecordingStatus()
+
+		return RecordingStatus.create({
+			isRecording: status.isRecording,
+			durationSeconds: status.durationSeconds,
+			error: status.error || "",
+		})
+	} catch (error) {
+		console.error("Error getting recording status:", error)
+		return RecordingStatus.create({
+			isRecording: false,
+			durationSeconds: 0,
+			error: error instanceof Error ? error.message : "Unknown error occurred",
+		})
+	}
+}

src/core/controller/dictation/startRecording.ts

@@ -0,0 +1,52 @@
+import { RecordingResult } from "@shared/proto/cline/dictation"
+import { HostProvider } from "@/hosts/host-provider"
+import { audioRecordingService } from "@/services/dictation/AudioRecordingService"
+import { telemetryService } from "@/services/posthog/PostHogClientProvider"
+import { ShowMessageType } from "@/shared/proto/host/window"
+import { Controller } from ".."
+
+/**
+ * Starts audio recording using the Extension Host
+ * @param controller The controller instance
+ * @returns RecordingResult with success status
+ */
+export const startRecording = async (controller: Controller): Promise<RecordingResult> => {
+	const taskId = controller.task?.taskId
+
+	try {
+		const userInfo = controller.authService.getInfo()
+		if (!userInfo?.user?.uid) {
+			throw new Error("Please sign in to your Cline Account to use Dictation.")
+		}
+
+		const result = await audioRecordingService.startRecording()
+
+		if (result.success) {
+			telemetryService.captureVoiceRecordingStarted(taskId, process.platform)
+		}
+
+		return RecordingResult.create({
+			success: result.success,
+			error: result.error || "",
+		})
+	} catch (error) {
+		console.error("Error starting recording:", error)
+		const errorMessage = error instanceof Error ? error.message : "Unknown error occurred"
+
+		const signInAction = "Sign in to Cline"
+		const action = await HostProvider.window.showMessage({
+			type: ShowMessageType.ERROR,
+			message: `Voice recording error: ${errorMessage}`,
+			options: { items: [signInAction] },
+		})
+
+		if (action.selectedOption === signInAction) {
+			await controller.authService.createAuthRequest()
+		}
+
+		return RecordingResult.create({
+			success: false,
+			error: errorMessage,
+		})
+	}
+}

src/core/controller/dictation/stopRecording.ts

@@ -0,0 +1,37 @@
+import { telemetryService } from "@services/posthog/PostHogClientProvider"
+import { RecordedAudio } from "@shared/proto/cline/dictation"
+import { audioRecordingService } from "@/services/dictation/AudioRecordingService"
+import { Controller } from ".."
+
+/**
+ * Stops audio recording and returns the recorded audio
+ * @param controller The controller instance
+ * @returns RecordedAudio with audio data
+ */
+export const stopRecording = async (controller: Controller): Promise<RecordedAudio> => {
+	const taskId = controller.task?.taskId
+	const recordingStatus = audioRecordingService.getRecordingStatus()
+	const recordingDuration = recordingStatus.durationSeconds * 1000 // Convert to milliseconds
+
+	try {
+		const result = await audioRecordingService.stopRecording()
+
+		telemetryService.captureVoiceRecordingStopped(taskId, recordingDuration, result.success, process.platform)
+
+		return RecordedAudio.create({
+			success: result.success,
+			audioBase64: result.audioBase64 || "",
+			error: result.error || "",
+		})
+	} catch (error) {
+		console.error("Error stopping recording:", error)
+
+		telemetryService.captureVoiceRecordingStopped(taskId, recordingDuration, false, process.platform)
+
+		return RecordedAudio.create({
+			success: false,
+			audioBase64: "",
+			error: error instanceof Error ? error.message : "Unknown error occurred",
+		})
+	}
+}

src/core/controller/dictation/transcribeAudio.ts

@@ -0,0 +1,77 @@
+import { telemetryService } from "@services/posthog/PostHogClientProvider"
+import { TranscribeAudioRequest, Transcription } from "@shared/proto/cline/dictation"
+import { HostProvider } from "@/hosts/host-provider"
+import { voiceTranscriptionService } from "@/services/dictation/VoiceTranscriptionService"
+import { ShowMessageType } from "@/shared/proto/host/window"
+import { Controller } from ".."
+
+/**
+ * Transcribes audio using Cline transcription service
+ * @param controller The controller instance
+ * @param request TranscribeAudioRequest containing base64 audio data
+ * @returns Transcription with transcribed text or error
+ */
+export const transcribeAudio = async (controller: Controller, request: TranscribeAudioRequest): Promise<Transcription> => {
+	const taskId = controller.task?.taskId
+	const startTime = Date.now()
+
+	// Capture telemetry for transcription start
+	telemetryService.captureVoiceTranscriptionStarted(taskId, request.language || "en")
+
+	try {
+		// Transcribe the audio
+		const result = await voiceTranscriptionService.transcribeAudio(request.audioBase64, request.language || "en")
+		const durationMs = Date.now() - startTime
+
+		if (result.error) {
+			let errorType = "api_error"
+			if (result.error.includes("Authentication failed")) {
+				errorType = "invalid_jwt_token"
+			} else if (result.error.includes("Insufficient credits")) {
+				errorType = "insufficient_credits"
+			} else if (result.error.includes("Invalid audio format")) {
+				errorType = "invalid_audio_format"
+			} else if (result.error.includes("Cannot connect")) {
+				errorType = "connection_error"
+			} else if (result.error.includes("Network error")) {
+				errorType = "network_error"
+			}
+
+			telemetryService.captureVoiceTranscriptionError(taskId, errorType, result.error, durationMs)
+
+			let errorMessage = ""
+			if (result.error.includes("Authentication failed")) {
+				errorMessage = "Authentication failed. Please log in again."
+			} else if (result.error.includes("Insufficient credits")) {
+				errorMessage = "Insufficient credits for transcription service."
+			} else if (result.error.includes("Cannot connect")) {
+				errorMessage = "Cannot connect to transcription service."
+			} else {
+				errorMessage = `Voice transcription failed: ${result.error}`
+			}
+
+			HostProvider.window.showMessage({
+				type: ShowMessageType.ERROR,
+				message: errorMessage,
+			})
+		} else if (result.text) {
+			telemetryService.captureVoiceTranscriptionCompleted(taskId, result.text.length, durationMs, request.language || "en")
+		}
+
+		return Transcription.create({
+			text: result.text || "",
+			error: result.error || "",
+		})
+	} catch (error) {
+		console.error("Error transcribing audio:", error)
+		const durationMs = Date.now() - startTime
+		const errorMessage = error instanceof Error ? error.message : "Unknown error occurred"
+
+		telemetryService.captureVoiceTranscriptionError(taskId, "unexpected_error", errorMessage, durationMs)
+
+		return Transcription.create({
+			text: "",
+			error: errorMessage,
+		})
+	}
+}

src/core/controller/file/toggleClineRule.ts

@@ -24,13 +24,20 @@ export async function toggleClineRule(controller: Controller, request: ToggleCli
 
 	// This is the same core logic as in the original handler
 	if (isGlobal) {
-		const toggles = controller.cacheService.getGlobalStateKey("globalClineRulesToggles")
+		const toggles = controller.stateManager.getGlobalStateKey("globalClineRulesToggles")
 		toggles[rulePath] = enabled
-		controller.cacheService.setGlobalState("globalClineRulesToggles", toggles)
+		controller.stateManager.setGlobalState("globalClineRulesToggles", toggles)
 	} else {
-		const toggles = controller.cacheService.getWorkspaceStateKey("localClineRulesToggles")
+		const toggles = controller.stateManager.getWorkspaceStateKey("localClineRulesToggles")
 		toggles[rulePath] = enabled
-		controller.cacheService.setWorkspaceState("localClineRulesToggles", toggles)
+		controller.stateManager.setWorkspaceState("localClineRulesToggles", toggles)
+	}
+
+	// Track rule toggle telemetry with current task context
+	if (controller.task?.ulid) {
+		// Extract just the filename for privacy (no full paths)
+		const ruleFileName = path.basename(rulePath)
+		telemetryService.captureClineRuleToggled(controller.task.ulid, ruleFileName, enabled, isGlobal)
 	}
 
 	// Track rule toggle telemetry with current task context
@@ -41,8 +48,8 @@ export async function toggleClineRule(controller: Controller, request: ToggleCli
 	}
 
 	// Get the current state to return in the response
-	const globalToggles = controller.cacheService.getGlobalStateKey("globalClineRulesToggles")
-	const localToggles = controller.cacheService.getWorkspaceStateKey("localClineRulesToggles")
+	const globalToggles = controller.stateManager.getGlobalStateKey("globalClineRulesToggles")
+	const localToggles = controller.stateManager.getWorkspaceStateKey("localClineRulesToggles")
 
 	return ToggleClineRules.create({
 		globalClineRulesToggles: { toggles: globalToggles },

src/core/controller/file/toggleCursorRule.ts

@@ -20,12 +20,12 @@ export async function toggleCursorRule(controller: Controller, request: ToggleCu
 	}
 
 	// Update the toggles in workspace state
-	const toggles = controller.cacheService.getWorkspaceStateKey("localCursorRulesToggles")
+	const toggles = controller.stateManager.getWorkspaceStateKey("localCursorRulesToggles")
 	toggles[rulePath] = enabled
-	controller.cacheService.setWorkspaceState("localCursorRulesToggles", toggles)
+	controller.stateManager.setWorkspaceState("localCursorRulesToggles", toggles)
 
 	// Get the current state to return in the response
-	const cursorToggles = controller.cacheService.getWorkspaceStateKey("localCursorRulesToggles")
+	const cursorToggles = controller.stateManager.getWorkspaceStateKey("localCursorRulesToggles")
 
 	return ClineRulesToggles.create({
 		toggles: cursorToggles,

src/core/controller/file/toggleWindsurfRule.ts

@@ -1,3 +1,4 @@
+import { ClineRulesToggles as AppClineRulesToggles } from "@shared/cline-rules"
 import type { ToggleWindsurfRuleRequest } from "@shared/proto/cline/file"
 import { ClineRulesToggles } from "@shared/proto/cline/file"
 import type { Controller } from "../index"
@@ -20,9 +21,9 @@ export async function toggleWindsurfRule(controller: Controller, request: Toggle
 	}
 
 	// Update the toggles
-	const toggles = controller.cacheService.getWorkspaceStateKey("localWindsurfRulesToggles")
+	const toggles = controller.stateManager.getWorkspaceStateKey("localWindsurfRulesToggles")
 	toggles[rulePath] = enabled
-	controller.cacheService.setWorkspaceState("localWindsurfRulesToggles", toggles)
+	controller.stateManager.setWorkspaceState("localWindsurfRulesToggles", toggles)
 
 	// Return the toggles directly
 	return ClineRulesToggles.create({ toggles: toggles })

src/core/controller/file/toggleWorkflow.ts

@@ -21,18 +21,18 @@ export async function toggleWorkflow(controller: Controller, request: ToggleWork
 	// Update the toggles based on isGlobal flag
 	if (isGlobal) {
 		// Global workflows
-		const toggles = controller.cacheService.getGlobalStateKey("globalWorkflowToggles")
+		const toggles = controller.stateManager.getGlobalStateKey("globalWorkflowToggles")
 		toggles[workflowPath] = enabled
-		controller.cacheService.setGlobalState("globalWorkflowToggles", toggles)
+		controller.stateManager.setGlobalState("globalWorkflowToggles", toggles)
 		await controller.postStateToWebview()
 
 		// Return the global toggles
 		return ClineRulesToggles.create({ toggles: toggles })
 	} else {
 		// Workspace workflows
-		const toggles = controller.cacheService.getWorkspaceStateKey("workflowToggles")
+		const toggles = controller.stateManager.getWorkspaceStateKey("workflowToggles")
 		toggles[workflowPath] = enabled
-		controller.cacheService.setWorkspaceState("workflowToggles", toggles)
+		controller.stateManager.setWorkspaceState("workflowToggles", toggles)
 		await controller.postStateToWebview()
 
 		// Return the workspace toggles

src/core/controller/index.ts

@@ -25,8 +25,8 @@ import { PostHogClientProvider, telemetryService } from "@/services/posthog/Post
 import { ShowMessageType } from "@/shared/proto/host/window"
 import { getLatestAnnouncementId } from "@/utils/announcements"
 import { getCwd, getDesktopDir } from "@/utils/path"
-import { CacheService, PersistenceErrorEvent } from "../storage/CacheService"
 import { ensureMcpServersDirectoryExists, ensureSettingsDirectoryExists, GlobalFileNames } from "../storage/disk"
+import { PersistenceErrorEvent, StateManager } from "../storage/StateManager"
 import { Task } from "../task"
 import { sendMcpMarketplaceCatalogEvent } from "./mcp/subscribeToMcpMarketplaceCatalog"
 import { sendStateUpdate } from "./state/subscribeToState"
@@ -45,7 +45,7 @@ export class Controller {
 	mcpHub: McpHub
 	accountService: ClineAccountService
 	authService: AuthService
-	readonly cacheService: CacheService
+	readonly stateManager: StateManager
 
 	constructor(
 		readonly context: vscode.ExtensionContext,
@@ -55,24 +55,24 @@ export class Controller {
 
 		HostProvider.get().logToChannel("ClineProvider instantiated")
 		this.accountService = ClineAccountService.getInstance()
-		this.cacheService = new CacheService(context)
+		this.stateManager = new StateManager(context)
 		this.authService = AuthService.getInstance(this)
 
 		// Initialize cache service asynchronously - critical for extension functionality
-		this.cacheService
+		this.stateManager
 			.initialize()
 			.then(() => {
 				this.authService.restoreRefreshTokenAndRetrieveAuthInfo()
 			})
 			.catch((error) => {
-				console.error("CRITICAL: Failed to initialize CacheService - extension may not function properly:", error)
+				console.error("CRITICAL: Failed to initialize StateManager - extension may not function properly:", error)
 			})
 
 		// Set up persistence error recovery
-		this.cacheService.onPersistenceError = async ({ error }: PersistenceErrorEvent) => {
+		this.stateManager.onPersistenceError = async ({ error }: PersistenceErrorEvent) => {
 			console.error("Cache persistence failed, recovering:", error)
 			try {
-				await this.cacheService.reInitialize()
+				await this.stateManager.reInitialize()
 				await this.postStateToWebview()
 				HostProvider.window.showMessage({
 					type: ShowMessageType.WARNING,
@@ -101,7 +101,7 @@ export class Controller {
 	}
 
 	async getCurrentMode(): Promise<Mode> {
-		return this.cacheService.getGlobalStateKey("mode")
+		return this.stateManager.getGlobalStateKey("mode")
 	}
 
 	/*
@@ -126,17 +126,17 @@ export class Controller {
 	async handleSignOut() {
 		try {
 			// TODO: update to clineAccountId and then move clineApiKey to a clear function.
-			this.cacheService.setSecret("clineAccountId", undefined)
-			this.cacheService.setGlobalState("userInfo", undefined)
+			this.stateManager.setSecret("clineAccountId", undefined)
+			this.stateManager.setGlobalState("userInfo", undefined)
 
 			// Update API providers through cache service
-			const apiConfiguration = this.cacheService.getApiConfiguration()
+			const apiConfiguration = this.stateManager.getApiConfiguration()
 			const updatedConfig = {
 				...apiConfiguration,
 				planModeApiProvider: "openrouter" as ApiProvider,
 				actModeApiProvider: "openrouter" as ApiProvider,
 			}
-			this.cacheService.setApiConfiguration(updatedConfig)
+			this.stateManager.setApiConfiguration(updatedConfig)
 
 			await this.postStateToWebview()
 			HostProvider.window.showMessage({
@@ -152,35 +152,35 @@ export class Controller {
 	}
 
 	async setUserInfo(info?: UserInfo) {
-		this.cacheService.setGlobalState("userInfo", info)
+		this.stateManager.setGlobalState("userInfo", info)
 	}
 
 	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
 
-		const apiConfiguration = this.cacheService.getApiConfiguration()
-		const autoApprovalSettings = this.cacheService.getGlobalStateKey("autoApprovalSettings")
-		const browserSettings = this.cacheService.getGlobalStateKey("browserSettings")
-		const focusChainSettings = this.cacheService.getGlobalStateKey("focusChainSettings")
-		const focusChainFeatureFlagEnabled = this.cacheService.getGlobalStateKey("focusChainFeatureFlagEnabled")
-		const preferredLanguage = this.cacheService.getGlobalStateKey("preferredLanguage")
-		const openaiReasoningEffort = this.cacheService.getGlobalStateKey("openaiReasoningEffort")
-		const mode = this.cacheService.getGlobalStateKey("mode")
-		const shellIntegrationTimeout = this.cacheService.getGlobalStateKey("shellIntegrationTimeout")
-		const terminalReuseEnabled = this.cacheService.getGlobalStateKey("terminalReuseEnabled")
-		const terminalOutputLineLimit = this.cacheService.getGlobalStateKey("terminalOutputLineLimit")
-		const defaultTerminalProfile = this.cacheService.getGlobalStateKey("defaultTerminalProfile")
-		const enableCheckpointsSetting = this.cacheService.getGlobalStateKey("enableCheckpointsSetting")
-		const isNewUser = this.cacheService.getGlobalStateKey("isNewUser")
-		const taskHistory = this.cacheService.getGlobalStateKey("taskHistory")
-		const strictPlanModeEnabled = this.cacheService.getGlobalStateKey("strictPlanModeEnabled")
-		const useAutoCondense = this.cacheService.getGlobalStateKey("useAutoCondense")
+		const apiConfiguration = this.stateManager.getApiConfiguration()
+		const autoApprovalSettings = this.stateManager.getGlobalStateKey("autoApprovalSettings")
+		const browserSettings = this.stateManager.getGlobalStateKey("browserSettings")
+		const focusChainSettings = this.stateManager.getGlobalStateKey("focusChainSettings")
+		const focusChainFeatureFlagEnabled = this.stateManager.getGlobalStateKey("focusChainFeatureFlagEnabled")
+		const preferredLanguage = this.stateManager.getGlobalStateKey("preferredLanguage")
+		const openaiReasoningEffort = this.stateManager.getGlobalStateKey("openaiReasoningEffort")
+		const mode = this.stateManager.getGlobalStateKey("mode")
+		const shellIntegrationTimeout = this.stateManager.getGlobalStateKey("shellIntegrationTimeout")
+		const terminalReuseEnabled = this.stateManager.getGlobalStateKey("terminalReuseEnabled")
+		const terminalOutputLineLimit = this.stateManager.getGlobalStateKey("terminalOutputLineLimit")
+		const defaultTerminalProfile = this.stateManager.getGlobalStateKey("defaultTerminalProfile")
+		const enableCheckpointsSetting = this.stateManager.getGlobalStateKey("enableCheckpointsSetting")
+		const isNewUser = this.stateManager.getGlobalStateKey("isNewUser")
+		const taskHistory = this.stateManager.getGlobalStateKey("taskHistory")
+		const strictPlanModeEnabled = this.stateManager.getGlobalStateKey("strictPlanModeEnabled")
+		const useAutoCondense = this.stateManager.getGlobalStateKey("useAutoCondense")
 
 		const NEW_USER_TASK_COUNT_THRESHOLD = 10
 
 		// Check if the user has completed enough tasks to no longer be considered a "new user"
 		if (isNewUser && !historyItem && taskHistory && taskHistory.length >= NEW_USER_TASK_COUNT_THRESHOLD) {
-			this.cacheService.setGlobalState("isNewUser", false)
+			this.stateManager.setGlobalState("isNewUser", false)
 			await this.postStateToWebview()
 		}
 
@@ -189,7 +189,7 @@ export class Controller {
 				...autoApprovalSettings,
 				version: (autoApprovalSettings.version ?? 1) + 1,
 			}
-			this.cacheService.setGlobalState("autoApprovalSettings", updatedAutoApprovalSettings)
+			this.stateManager.setGlobalState("autoApprovalSettings", updatedAutoApprovalSettings)
 		}
 		// Apply remote feature flag gate to focus chain settings
 		const effectiveFocusChainSettings = {
@@ -219,7 +219,7 @@ export class Controller {
 			defaultTerminalProfile ?? "default",
 			enableCheckpointsSetting ?? true,
 			await getCwd(getDesktopDir()),
-			this.cacheService,
+			this.stateManager,
 			task,
 			images,
 			files,
@@ -235,7 +235,7 @@ export class Controller {
 	}
 
 	async updateTelemetrySetting(telemetrySetting: TelemetrySetting) {
-		this.cacheService.setGlobalState("telemetrySetting", telemetrySetting)
+		this.stateManager.setGlobalState("telemetrySetting", telemetrySetting)
 		const isOptedIn = telemetrySetting !== "disabled"
 		telemetryService.updateTelemetryState(isOptedIn)
 		await this.postStateToWebview()
@@ -245,14 +245,14 @@ export class Controller {
 		const didSwitchToActMode = modeToSwitchTo === "act"
 
 		// Store mode to global state
-		this.cacheService.setGlobalState("mode", modeToSwitchTo)
+		this.stateManager.setGlobalState("mode", modeToSwitchTo)
 
 		// Capture mode switch telemetry | Capture regardless of if we know the taskId
 		telemetryService.captureModeSwitch(this.task?.ulid ?? "0", modeToSwitchTo)
 
 		// Update API handler with new mode (buildApiHandler now selects provider based on mode)
 		if (this.task) {
-			const apiConfiguration = this.cacheService.getApiConfiguration()
+			const apiConfiguration = this.stateManager.getApiConfiguration()
 			this.task.api = buildApiHandler({ ...apiConfiguration, ulid: this.task.ulid }, modeToSwitchTo)
 		}
 
@@ -317,12 +317,12 @@ export class Controller {
 			const clineProvider: ApiProvider = "cline"
 
 			// Get current settings to determine how to update providers
-			const planActSeparateModelsSetting = this.cacheService.getGlobalStateKey("planActSeparateModelsSetting")
+			const planActSeparateModelsSetting = this.stateManager.getGlobalStateKey("planActSeparateModelsSetting")
 
 			const currentMode = await this.getCurrentMode()
 
 			// Get current API configuration from cache
-			const currentApiConfiguration = this.cacheService.getApiConfiguration()
+			const currentApiConfiguration = this.stateManager.getApiConfiguration()
 
 			const updatedConfig = { ...currentApiConfiguration }
 
@@ -340,10 +340,10 @@ export class Controller {
 			}
 
 			// Update the API configuration through cache service
-			this.cacheService.setApiConfiguration(updatedConfig)
+			this.stateManager.setApiConfiguration(updatedConfig)
 
 			// Mark welcome view as completed since user has successfully logged in
-			this.cacheService.setGlobalState("welcomeViewCompleted", true)
+			this.stateManager.setGlobalState("welcomeViewCompleted", true)
 
 			if (this.task) {
 				this.task.api = buildApiHandler({ ...updatedConfig, ulid: this.task.ulid }, currentMode)
@@ -384,7 +384,7 @@ export class Controller {
 			}
 
 			// Store in global state
-			this.cacheService.setGlobalState("mcpMarketplaceCatalog", catalog)
+			this.stateManager.setGlobalState("mcpMarketplaceCatalog", catalog)
 			return catalog
 		} catch (error) {
 			console.error("Failed to fetch MCP marketplace:", error)
@@ -422,7 +422,7 @@ export class Controller {
 			}
 
 			// Store in global state
-			this.cacheService.setGlobalState("mcpMarketplaceCatalog", catalog)
+			this.stateManager.setGlobalState("mcpMarketplaceCatalog", catalog)
 			return catalog
 		} catch (error) {
 			console.error("Failed to fetch MCP marketplace:", error)
@@ -479,14 +479,14 @@ export class Controller {
 		const currentMode = await this.getCurrentMode()
 
 		// Update API configuration through cache service
-		const currentApiConfiguration = this.cacheService.getApiConfiguration()
+		const currentApiConfiguration = this.stateManager.getApiConfiguration()
 		const updatedConfig = {
 			...currentApiConfiguration,
 			planModeApiProvider: openrouter,
 			actModeApiProvider: openrouter,
 			openRouterApiKey: apiKey,
 		}
-		this.cacheService.setApiConfiguration(updatedConfig)
+		this.stateManager.setApiConfiguration(updatedConfig)
 
 		await this.postStateToWebview()
 		if (this.task) {
@@ -537,7 +537,7 @@ export class Controller {
 		taskMetadataFilePath: string
 		apiConversationHistory: Anthropic.MessageParam[]
 	}> {
-		const history = this.cacheService.getGlobalStateKey("taskHistory")
+		const history = this.stateManager.getGlobalStateKey("taskHistory")
 		const historyItem = history.find((item) => item.id === id)
 		if (historyItem) {
 			const taskDirPath = path.join(this.context.globalStorageUri.fsPath, "tasks", id)
@@ -572,9 +572,9 @@ export class Controller {
 
 	async deleteTaskFromState(id: string) {
 		// Remove the task from history
-		const taskHistory = this.cacheService.getGlobalStateKey("taskHistory")
+		const taskHistory = this.stateManager.getGlobalStateKey("taskHistory")
 		const updatedTaskHistory = taskHistory.filter((task) => task.id !== id)
-		this.cacheService.setGlobalState("taskHistory", updatedTaskHistory)
+		this.stateManager.setGlobalState("taskHistory", updatedTaskHistory)
 
 		// Notify the webview that the task has been deleted
 		await this.postStateToWebview()
@@ -589,40 +589,41 @@ export class Controller {
 
 	async getStateToPostToWebview(): Promise<ExtensionState> {
 		// Get API configuration from cache for immediate access
-		const apiConfiguration = this.cacheService.getApiConfiguration()
-		const lastShownAnnouncementId = this.cacheService.getGlobalStateKey("lastShownAnnouncementId")
-		const taskHistory = this.cacheService.getGlobalStateKey("taskHistory")
-		const autoApprovalSettings = this.cacheService.getGlobalStateKey("autoApprovalSettings")
-		const browserSettings = this.cacheService.getGlobalStateKey("browserSettings")
-		const focusChainSettings = this.cacheService.getGlobalStateKey("focusChainSettings")
-		const focusChainFeatureFlagEnabled = this.cacheService.getGlobalStateKey("focusChainFeatureFlagEnabled")
-		const preferredLanguage = this.cacheService.getGlobalStateKey("preferredLanguage")
-		const openaiReasoningEffort = this.cacheService.getGlobalStateKey("openaiReasoningEffort")
-		const mode = this.cacheService.getGlobalStateKey("mode")
-		const strictPlanModeEnabled = this.cacheService.getGlobalStateKey("strictPlanModeEnabled")
-		const useAutoCondense = this.cacheService.getGlobalStateKey("useAutoCondense")
-		const userInfo = this.cacheService.getGlobalStateKey("userInfo")
-		const mcpMarketplaceEnabled = this.cacheService.getGlobalStateKey("mcpMarketplaceEnabled")
-		const mcpDisplayMode = this.cacheService.getGlobalStateKey("mcpDisplayMode")
-		const telemetrySetting = this.cacheService.getGlobalStateKey("telemetrySetting")
-		const planActSeparateModelsSetting = this.cacheService.getGlobalStateKey("planActSeparateModelsSetting")
-		const enableCheckpointsSetting = this.cacheService.getGlobalStateKey("enableCheckpointsSetting")
-		const globalClineRulesToggles = this.cacheService.getGlobalStateKey("globalClineRulesToggles")
-		const globalWorkflowToggles = this.cacheService.getGlobalStateKey("globalWorkflowToggles")
-		const shellIntegrationTimeout = this.cacheService.getGlobalStateKey("shellIntegrationTimeout")
-		const terminalReuseEnabled = this.cacheService.getGlobalStateKey("terminalReuseEnabled")
-		const defaultTerminalProfile = this.cacheService.getGlobalStateKey("defaultTerminalProfile")
-		const isNewUser = this.cacheService.getGlobalStateKey("isNewUser")
+		const apiConfiguration = this.stateManager.getApiConfiguration()
+		const lastShownAnnouncementId = this.stateManager.getGlobalStateKey("lastShownAnnouncementId")
+		const taskHistory = this.stateManager.getGlobalStateKey("taskHistory")
+		const autoApprovalSettings = this.stateManager.getGlobalStateKey("autoApprovalSettings")
+		const browserSettings = this.stateManager.getGlobalStateKey("browserSettings")
+		const dictationSettings = this.stateManager.getGlobalStateKey("dictationSettings")
+		const focusChainSettings = this.stateManager.getGlobalStateKey("focusChainSettings")
+		const focusChainFeatureFlagEnabled = this.stateManager.getGlobalStateKey("focusChainFeatureFlagEnabled")
+		const preferredLanguage = this.stateManager.getGlobalStateKey("preferredLanguage")
+		const openaiReasoningEffort = this.stateManager.getGlobalStateKey("openaiReasoningEffort")
+		const mode = this.stateManager.getGlobalStateKey("mode")
+		const strictPlanModeEnabled = this.stateManager.getGlobalStateKey("strictPlanModeEnabled")
+		const useAutoCondense = this.stateManager.getGlobalStateKey("useAutoCondense")
+		const userInfo = this.stateManager.getGlobalStateKey("userInfo")
+		const mcpMarketplaceEnabled = this.stateManager.getGlobalStateKey("mcpMarketplaceEnabled")
+		const mcpDisplayMode = this.stateManager.getGlobalStateKey("mcpDisplayMode")
+		const telemetrySetting = this.stateManager.getGlobalStateKey("telemetrySetting")
+		const planActSeparateModelsSetting = this.stateManager.getGlobalStateKey("planActSeparateModelsSetting")
+		const enableCheckpointsSetting = this.stateManager.getGlobalStateKey("enableCheckpointsSetting")
+		const globalClineRulesToggles = this.stateManager.getGlobalStateKey("globalClineRulesToggles")
+		const globalWorkflowToggles = this.stateManager.getGlobalStateKey("globalWorkflowToggles")
+		const shellIntegrationTimeout = this.stateManager.getGlobalStateKey("shellIntegrationTimeout")
+		const terminalReuseEnabled = this.stateManager.getGlobalStateKey("terminalReuseEnabled")
+		const defaultTerminalProfile = this.stateManager.getGlobalStateKey("defaultTerminalProfile")
+		const isNewUser = this.stateManager.getGlobalStateKey("isNewUser")
 		const welcomeViewCompleted = Boolean(
-			this.cacheService.getGlobalStateKey("welcomeViewCompleted") || this.authService.getInfo()?.user?.uid,
+			this.stateManager.getGlobalStateKey("welcomeViewCompleted") || this.authService.getInfo()?.user?.uid,
 		)
-		const customPrompt = this.cacheService.getGlobalStateKey("customPrompt")
-		const mcpResponsesCollapsed = this.cacheService.getGlobalStateKey("mcpResponsesCollapsed")
-		const terminalOutputLineLimit = this.cacheService.getGlobalStateKey("terminalOutputLineLimit")
-		const localClineRulesToggles = this.cacheService.getWorkspaceStateKey("localClineRulesToggles")
-		const localWindsurfRulesToggles = this.cacheService.getWorkspaceStateKey("localWindsurfRulesToggles")
-		const localCursorRulesToggles = this.cacheService.getWorkspaceStateKey("localCursorRulesToggles")
-		const workflowToggles = this.cacheService.getWorkspaceStateKey("workflowToggles")
+		const customPrompt = this.stateManager.getGlobalStateKey("customPrompt")
+		const mcpResponsesCollapsed = this.stateManager.getGlobalStateKey("mcpResponsesCollapsed")
+		const terminalOutputLineLimit = this.stateManager.getGlobalStateKey("terminalOutputLineLimit")
+		const localClineRulesToggles = this.stateManager.getWorkspaceStateKey("localClineRulesToggles")
+		const localWindsurfRulesToggles = this.stateManager.getWorkspaceStateKey("localWindsurfRulesToggles")
+		const localCursorRulesToggles = this.stateManager.getWorkspaceStateKey("localCursorRulesToggles")
+		const workflowToggles = this.stateManager.getWorkspaceStateKey("workflowToggles")
 
 		const currentTaskItem = this.task?.taskId ? (taskHistory || []).find((item) => item.id === this.task?.taskId) : undefined
 		const checkpointTrackerErrorMessage = this.task?.taskState.checkpointTrackerErrorMessage
@@ -653,6 +654,7 @@ export class Controller {
 			platform,
 			autoApprovalSettings,
 			browserSettings,
+			dictationSettings,
 			focusChainSettings,
 			focusChainFeatureFlagEnabled,
 			preferredLanguage,
@@ -710,14 +712,14 @@ export class Controller {
 	*/
 
 	async updateTaskHistory(item: HistoryItem): Promise<HistoryItem[]> {
-		const history = this.cacheService.getGlobalStateKey("taskHistory")
+		const history = this.stateManager.getGlobalStateKey("taskHistory")
 		const existingItemIndex = history.findIndex((h) => h.id === item.id)
 		if (existingItemIndex !== -1) {
 			history[existingItemIndex] = item
 		} else {
 			history.push(item)
 		}
-		this.cacheService.setGlobalState("taskHistory", history)
+		this.stateManager.setGlobalState("taskHistory", history)
 		return history
 	}
 }

src/core/controller/models/refreshBasetenModels.ts

@@ -22,7 +22,7 @@ export async function refreshBasetenModels(
 	const basetenModelsFilePath = path.join(await ensureCacheDirectoryExists(controller), GlobalFileNames.basetenModels)
 
 	// Get the Baseten API key from the controller's state
-	const basetenApiKey = controller.cacheService.getSecretKey("basetenApiKey")
+	const basetenApiKey = controller.stateManager.getSecretKey("basetenApiKey")
 
 	const models: Record<string, Partial<OpenRouterModelInfo>> = {}
 	try {

src/core/controller/models/refreshGroqModels.ts

@@ -18,7 +18,7 @@ import { Controller } from ".."
 export async function refreshGroqModels(controller: Controller, _request: EmptyRequest): Promise<OpenRouterCompatibleModelInfo> {
 	const groqModelsFilePath = path.join(await ensureCacheDirectoryExists(controller), GlobalFileNames.groqModels)
 
-	const groqApiKey = controller.cacheService.getSecretKey("groqApiKey")
+	const groqApiKey = controller.stateManager.getSecretKey("groqApiKey")
 
 	let models: Record<string, Partial<OpenRouterModelInfo>> = {}
 	try {

src/core/controller/models/refreshRequestyModels.ts

@@ -19,7 +19,7 @@ export async function refreshRequestyModels(controller: Controller, _: EmptyRequ
 
 	const models: Record<string, OpenRouterModelInfo> = {}
 	try {
-		const apiKey = controller.cacheService.getSecretKey("requestyApiKey")
+		const apiKey = controller.stateManager.getSecretKey("requestyApiKey")
 		const headers = {
 			Authorization: `Bearer ${apiKey}`,
 		}

src/core/controller/models/updateApiConfigurationProto.ts

@@ -24,7 +24,7 @@ export async function updateApiConfigurationProto(
 		const appApiConfiguration = convertProtoToApiConfiguration(request.apiConfiguration)
 
 		// Update the API configuration in storage
-		controller.cacheService.setApiConfiguration(appApiConfiguration)
+		controller.stateManager.setApiConfiguration(appApiConfiguration)
 
 		// Update the task's API handler if there's an active task
 		if (controller.task) {

src/core/controller/state/setWelcomeViewCompleted.ts

@@ -11,7 +11,7 @@ import type { Controller } from "../index"
 export async function setWelcomeViewCompleted(controller: Controller, request: BooleanRequest): Promise<Empty> {
 	try {
 		// Update the global state to set welcomeViewCompleted to the requested value
-		controller.cacheService.setGlobalState("welcomeViewCompleted", request.value)
+		controller.stateManager.setGlobalState("welcomeViewCompleted", request.value)
 
 		await controller.postStateToWebview()
 

src/core/controller/state/toggleFavoriteModel.ts

@@ -15,7 +15,7 @@ export async function toggleFavoriteModel(controller: Controller, request: Strin
 		}
 
 		const modelId = request.value
-		const apiConfiguration = controller.cacheService.getApiConfiguration()
+		const apiConfiguration = controller.stateManager.getApiConfiguration()
 
 		const favoritedModelIds = apiConfiguration.favoritedModelIds || []
 
@@ -29,7 +29,7 @@ export async function toggleFavoriteModel(controller: Controller, request: Strin
 			...apiConfiguration,
 			favoritedModelIds: updatedFavorites,
 		}
-		controller.cacheService.setApiConfiguration(updatedApiConfiguration)
+		controller.stateManager.setApiConfiguration(updatedApiConfiguration)
 
 		// Capture telemetry for model favorite toggle
 		const isFavorited = !favoritedModelIds.includes(modelId)

src/core/controller/state/updateAutoApprovalSettings.ts

@@ -18,7 +18,7 @@ export async function updateAutoApprovalSettings(controller: Controller, request
 	if (incomingVersion > currentVersion) {
 		const settings = convertProtoToAutoApprovalSettings(request)
 
-		controller.cacheService.setGlobalState("autoApprovalSettings", settings)
+		controller.stateManager.setGlobalState("autoApprovalSettings", settings)
 
 		if (controller.task) {
 			controller.task.updateAutoApprovalSettings(settings)

src/core/controller/state/updateDefaultTerminalProfile.ts

@@ -11,7 +11,7 @@ export async function updateDefaultTerminalProfile(
 	const profileId = request.value
 
 	// Update the terminal profile in the state
-	controller.cacheService.setGlobalState("defaultTerminalProfile", profileId)
+	controller.stateManager.setGlobalState("defaultTerminalProfile", profileId)
 
 	let closedCount = 0
 	let busyTerminals: TerminalInfo[] = []

src/core/controller/state/updateSettings.ts

@@ -1,4 +1,5 @@
 import { buildApiHandler } from "@core/api"
+import { telemetryService } from "@services/posthog/PostHogClientProvider"
 import { Empty } from "@shared/proto/cline/common"
 import {
 	PlanActMode,
@@ -7,10 +8,9 @@ import {
 	UpdateSettingsRequest,
 } from "@shared/proto/cline/state"
 import { convertProtoApiConfigurationToApiConfiguration } from "@shared/proto-conversions/state/settings-conversion"
-import { OpenaiReasoningEffort } from "@shared/storage/types"
-import { TelemetrySetting } from "@shared/TelemetrySetting"
 import { McpDisplayMode } from "@/shared/McpDisplayMode"
-import { telemetryService } from "../../../services/posthog/PostHogClientProvider"
+import { OpenaiReasoningEffort } from "@/shared/storage/types"
+import { TelemetrySetting } from "@/shared/TelemetrySetting"
 import { Controller } from ".."
 
 /**
@@ -24,7 +24,7 @@ export async function updateSettings(controller: Controller, request: UpdateSett
 		// Update API configuration
 		if (request.apiConfiguration) {
 			const apiConfiguration = convertProtoApiConfigurationToApiConfiguration(request.apiConfiguration)
-			controller.cacheService.setApiConfiguration(apiConfiguration)
+			controller.stateManager.setApiConfiguration(apiConfiguration)
 
 			if (controller.task) {
 				const currentMode = await controller.getCurrentMode()
@@ -39,22 +39,22 @@ export async function updateSettings(controller: Controller, request: UpdateSett
 
 		// Update plan/act separate models setting
 		if (request.planActSeparateModelsSetting !== undefined) {
-			controller.cacheService.setGlobalState("planActSeparateModelsSetting", request.planActSeparateModelsSetting)
+			controller.stateManager.setGlobalState("planActSeparateModelsSetting", request.planActSeparateModelsSetting)
 		}
 
 		// Update checkpoints setting
 		if (request.enableCheckpointsSetting !== undefined) {
-			controller.cacheService.setGlobalState("enableCheckpointsSetting", request.enableCheckpointsSetting)
+			controller.stateManager.setGlobalState("enableCheckpointsSetting", request.enableCheckpointsSetting)
 		}
 
 		// Update MCP marketplace setting
 		if (request.mcpMarketplaceEnabled !== undefined) {
-			controller.cacheService.setGlobalState("mcpMarketplaceEnabled", request.mcpMarketplaceEnabled)
+			controller.stateManager.setGlobalState("mcpMarketplaceEnabled", request.mcpMarketplaceEnabled)
 		}
 
 		// Update MCP responses collapsed setting
 		if (request.mcpResponsesCollapsed !== undefined) {
-			controller.cacheService.setGlobalState("mcpResponsesCollapsed", request.mcpResponsesCollapsed)
+			controller.stateManager.setGlobalState("mcpResponsesCollapsed", request.mcpResponsesCollapsed)
 		}
 
 		// Update MCP display mode setting
@@ -74,7 +74,7 @@ export async function updateSettings(controller: Controller, request: UpdateSett
 				default:
 					throw new Error(`Invalid MCP display mode value: ${request.mcpDisplayMode}`)
 			}
-			controller.cacheService.setGlobalState("mcpDisplayMode", displayMode)
+			controller.stateManager.setGlobalState("mcpDisplayMode", displayMode)
 		}
 
 		if (request.mode !== undefined) {
@@ -82,7 +82,7 @@ export async function updateSettings(controller: Controller, request: UpdateSett
 			if (controller.task) {
 				controller.task.updateMode(mode)
 			}
-			controller.cacheService.setGlobalState("mode", mode)
+			controller.stateManager.setGlobalState("mode", mode)
 		}
 
 		if (request.openaiReasoningEffort !== undefined) {
@@ -106,29 +106,29 @@ export async function updateSettings(controller: Controller, request: UpdateSett
 				controller.task.openaiReasoningEffort = reasoningEffort
 			}
 
-			controller.cacheService.setGlobalState("openaiReasoningEffort", reasoningEffort)
+			controller.stateManager.setGlobalState("openaiReasoningEffort", reasoningEffort)
 		}
 
 		if (request.preferredLanguage !== undefined) {
 			if (controller.task) {
 				controller.task.preferredLanguage = request.preferredLanguage
 			}
-			controller.cacheService.setGlobalState("preferredLanguage", request.preferredLanguage)
+			controller.stateManager.setGlobalState("preferredLanguage", request.preferredLanguage)
 		}
 
 		// Update terminal timeout setting
 		if (request.shellIntegrationTimeout !== undefined) {
-			controller.cacheService.setGlobalState("shellIntegrationTimeout", Number(request.shellIntegrationTimeout))
+			controller.stateManager.setGlobalState("shellIntegrationTimeout", Number(request.shellIntegrationTimeout))
 		}
 
 		// Update terminal reuse setting
 		if (request.terminalReuseEnabled !== undefined) {
-			controller.cacheService.setGlobalState("terminalReuseEnabled", request.terminalReuseEnabled)
+			controller.stateManager.setGlobalState("terminalReuseEnabled", request.terminalReuseEnabled)
 		}
 
 		// Update terminal output line limit
 		if (request.terminalOutputLineLimit !== undefined) {
-			controller.cacheService.setGlobalState("terminalOutputLineLimit", Number(request.terminalOutputLineLimit))
+			controller.stateManager.setGlobalState("terminalOutputLineLimit", Number(request.terminalOutputLineLimit))
 		}
 
 		// Update strict plan mode setting
@@ -136,7 +136,11 @@ export async function updateSettings(controller: Controller, request: UpdateSett
 			if (controller.task) {
 				controller.task.updateStrictPlanMode(request.strictPlanModeEnabled)
 			}
-			controller.cacheService.setGlobalState("strictPlanModeEnabled", request.strictPlanModeEnabled)
+			controller.stateManager.setGlobalState("strictPlanModeEnabled", request.strictPlanModeEnabled)
+		}
+
+		if (request.dictationSettings !== undefined) {
+			controller.stateManager.setGlobalState("dictationSettings", request.dictationSettings)
 		}
 
 		// Update auto-condense setting
@@ -144,16 +148,16 @@ export async function updateSettings(controller: Controller, request: UpdateSett
 			if (controller.task) {
 				controller.task.updateUseAutoCondense(request.useAutoCondense)
 			}
-			controller.cacheService.setGlobalState("useAutoCondense", request.useAutoCondense)
+			controller.stateManager.setGlobalState("useAutoCondense", request.useAutoCondense)
 		}
 
 		// Update focus chain settings
 		if (request.focusChainSettings !== undefined) {
-			const remoteEnabled = controller.cacheService.getGlobalStateKey("focusChainFeatureFlagEnabled")
+			const remoteEnabled = controller.stateManager.getGlobalStateKey("focusChainFeatureFlagEnabled")
 			if (remoteEnabled === false) {
 				// No-op when feature flag disabled
 			} else {
-				const currentSettings = controller.cacheService.getGlobalStateKey("focusChainSettings")
+				const currentSettings = controller.stateManager.getGlobalStateKey("focusChainSettings")
 				const wasEnabled = currentSettings?.enabled ?? false
 				const isEnabled = request.focusChainSettings.enabled
 
@@ -161,7 +165,7 @@ export async function updateSettings(controller: Controller, request: UpdateSett
 					enabled: isEnabled,
 					remindClineInterval: request.focusChainSettings.remindClineInterval,
 				}
-				controller.cacheService.setGlobalState("focusChainSettings", focusChainSettings)
+				controller.stateManager.setGlobalState("focusChainSettings", focusChainSettings)
 
 				// Capture telemetry when setting changes
 				if (wasEnabled !== isEnabled) {
@@ -173,7 +177,7 @@ export async function updateSettings(controller: Controller, request: UpdateSett
 		// Update custom prompt choice
 		if (request.customPrompt !== undefined) {
 			const value = request.customPrompt === "compact" ? "compact" : undefined
-			controller.cacheService.setGlobalState("customPrompt", value)
+			controller.stateManager.setGlobalState("customPrompt", value)
 		}
 
 		// Post updated state to webview

src/core/controller/state/updateTerminalConnectionTimeout.ts

@@ -8,7 +8,7 @@ export async function updateTerminalConnectionTimeout(
 	const timeoutMs = request.timeoutMs
 
 	// Update the terminal connection timeout setting in the state
-	controller.cacheService.setGlobalState("shellIntegrationTimeout", timeoutMs || 4000)
+	controller.stateManager.setGlobalState("shellIntegrationTimeout", timeoutMs || 4000)
 
 	// Broadcast state update to all webviews
 	await controller.postStateToWebview()

src/core/controller/state/updateTerminalReuseEnabled.ts

@@ -8,7 +8,7 @@ export async function updateTerminalReuseEnabled(
 	const enabled = request.value
 
 	// Update the terminal reuse setting in the state
-	controller.cacheService.setGlobalState("terminalReuseEnabled", enabled)
+	controller.stateManager.setGlobalState("terminalReuseEnabled", enabled)
 
 	// Broadcast state update to all webviews
 	await controller.postStateToWebview()

src/core/controller/task/deleteAllTaskHistory.ts

@@ -18,7 +18,7 @@ export async function deleteAllTaskHistory(controller: Controller): Promise<Dele
 		await controller.clearTask()
 
 		// Get existing task history
-		const taskHistory = controller.cacheService.getGlobalStateKey("taskHistory")
+		const taskHistory = controller.stateManager.getGlobalStateKey("taskHistory")
 		const totalTasks = taskHistory.length
 
 		const userChoice = (
@@ -47,7 +47,7 @@ export async function deleteAllTaskHistory(controller: Controller): Promise<Dele
 
 			// If there are favorited tasks, update state
 			if (favoritedTasks.length > 0) {
-				controller.cacheService.setGlobalState("taskHistory", favoritedTasks)
+				controller.stateManager.setGlobalState("taskHistory", favoritedTasks)
 
 				// Delete non-favorited task directories
 				const preserveTaskIds = favoritedTasks.map((task) => task.id)
@@ -87,7 +87,7 @@ export async function deleteAllTaskHistory(controller: Controller): Promise<Dele
 		}
 
 		// Delete everything (not preserving favorites)
-		controller.cacheService.setGlobalState("taskHistory", [])
+		controller.stateManager.setGlobalState("taskHistory", [])
 
 		try {
 			// Remove all contents of tasks directory

src/core/controller/task/getTaskHistory.ts

@@ -13,7 +13,7 @@ export async function getTaskHistory(controller: Controller, request: GetTaskHis
 		const { favoritesOnly, currentWorkspaceOnly, searchQuery, sortBy } = request
 
 		// Get task history from global state
-		const taskHistory = controller.cacheService.getGlobalStateKey("taskHistory")
+		const taskHistory = controller.stateManager.getGlobalStateKey("taskHistory")
 		const workspacePath = await getWorkspacePath()
 
 		// Apply filters

src/core/controller/task/showTaskWithId.ts

@@ -14,7 +14,7 @@ export async function showTaskWithId(controller: Controller, request: StringRequ
 		const id = request.value
 
 		// First check if task exists in global state for faster access
-		const taskHistory = controller.cacheService.getGlobalStateKey("taskHistory")
+		const taskHistory = controller.stateManager.getGlobalStateKey("taskHistory")
 		const historyItem = taskHistory.find((item) => item.id === id)
 
 		// We need to initialize the task before returning data

src/core/controller/task/toggleTaskFavorite.ts

@@ -12,7 +12,7 @@ export async function toggleTaskFavorite(controller: Controller, request: TaskFa
 	try {
 		// Update in-memory state only
 		try {
-			const history = controller.cacheService.getGlobalStateKey("taskHistory")
+			const history = controller.stateManager.getGlobalStateKey("taskHistory")
 
 			const taskIndex = history.findIndex((item) => item.id === request.taskId)
 
@@ -28,7 +28,7 @@ export async function toggleTaskFavorite(controller: Controller, request: TaskFa
 
 				// Update global state and wait for it to complete
 				try {
-					controller.cacheService.setGlobalState("taskHistory", updatedHistory)
+					controller.stateManager.setGlobalState("taskHistory", updatedHistory)
 				} catch (stateErr) {
 					console.error("Error updating global state:", stateErr)
 				}

src/core/controller/ui/initializeWebview.ts

@@ -29,8 +29,8 @@ export async function initializeWebview(controller: Controller, _request: EmptyR
 		refreshOpenRouterModels(controller, EmptyRequest.create()).then(async (response) => {
 			if (response && response.models) {
 				// Update model info in state (this needs to be done here since we don't want to update state while settings is open, and we may refresh models there)
-				const apiConfiguration = controller.cacheService.getApiConfiguration()
-				const planActSeparateModelsSetting = controller.cacheService.getGlobalStateKey("planActSeparateModelsSetting")
+				const apiConfiguration = controller.stateManager.getApiConfiguration()
+				const planActSeparateModelsSetting = controller.stateManager.getGlobalStateKey("planActSeparateModelsSetting")
 				const currentMode = await controller.getCurrentMode()
 
 				if (planActSeparateModelsSetting) {
@@ -44,7 +44,7 @@ export async function initializeWebview(controller: Controller, _request: EmptyR
 							...apiConfiguration,
 							[modelInfoField]: response.models[modelId],
 						}
-						controller.cacheService.setApiConfiguration(updatedConfig)
+						controller.stateManager.setApiConfiguration(updatedConfig)
 						await controller.postStateToWebview()
 					}
 				} else {
@@ -65,7 +65,7 @@ export async function initializeWebview(controller: Controller, _request: EmptyR
 
 					// Post state update if we updated any model info
 					if ((planModelId && response.models[planModelId]) || (actModelId && response.models[actModelId])) {
-						controller.cacheService.setApiConfiguration(updatedConfig)
+						controller.stateManager.setApiConfiguration(updatedConfig)
 						await controller.postStateToWebview()
 					}
 				}
@@ -75,8 +75,8 @@ export async function initializeWebview(controller: Controller, _request: EmptyR
 		refreshGroqModels(controller, EmptyRequest.create()).then(async (response) => {
 			if (response && response.models) {
 				// Update model info in state for Groq (this needs to be done here since we don't want to update state while settings is open, and we may refresh models there)
-				const apiConfiguration = controller.cacheService.getApiConfiguration()
-				const planActSeparateModelsSetting = controller.cacheService.getGlobalStateKey("planActSeparateModelsSetting")
+				const apiConfiguration = controller.stateManager.getApiConfiguration()
+				const planActSeparateModelsSetting = controller.stateManager.getGlobalStateKey("planActSeparateModelsSetting")
 				const currentMode = await controller.getCurrentMode()
 
 				if (planActSeparateModelsSetting) {
@@ -90,7 +90,7 @@ export async function initializeWebview(controller: Controller, _request: EmptyR
 							...apiConfiguration,
 							[modelInfoField]: response.models[modelId],
 						}
-						controller.cacheService.setApiConfiguration(updatedConfig)
+						controller.stateManager.setApiConfiguration(updatedConfig)
 						await controller.postStateToWebview()
 					}
 				} else {
@@ -111,7 +111,7 @@ export async function initializeWebview(controller: Controller, _request: EmptyR
 
 					// Post state update if we updated any model info
 					if ((planModelId && response.models[planModelId]) || (actModelId && response.models[actModelId])) {
-						controller.cacheService.setApiConfiguration(updatedConfig)
+						controller.stateManager.setApiConfiguration(updatedConfig)
 						await controller.postStateToWebview()
 					}
 				}
@@ -121,8 +121,8 @@ export async function initializeWebview(controller: Controller, _request: EmptyR
 		refreshBasetenModels(controller, EmptyRequest.create()).then(async (response) => {
 			if (response && response.models) {
 				// Update model info in state for Baseten (this needs to be done here since we don't want to update state while settings is open, and we may refresh models there)
-				const apiConfiguration = controller.cacheService.getApiConfiguration()
-				const planActSeparateModelsSetting = controller.cacheService.getGlobalStateKey("planActSeparateModelsSetting")
+				const apiConfiguration = controller.stateManager.getApiConfiguration()
+				const planActSeparateModelsSetting = controller.stateManager.getGlobalStateKey("planActSeparateModelsSetting")
 
 				const currentMode = await controller.getCurrentMode()
 
@@ -133,7 +133,7 @@ export async function initializeWebview(controller: Controller, _request: EmptyR
 					const modelId = apiConfiguration[modelIdField]
 
 					if (modelId && response.models[modelId]) {
-						controller.cacheService.setGlobalState(modelInfoField, response.models[modelId])
+						controller.stateManager.setGlobalState(modelInfoField, response.models[modelId])
 						await controller.postStateToWebview()
 					}
 				} else {
@@ -143,12 +143,12 @@ export async function initializeWebview(controller: Controller, _request: EmptyR
 
 					// Update plan mode model info if we have a model ID
 					if (planModelId && response.models[planModelId]) {
-						controller.cacheService.setGlobalState("planModeBasetenModelInfo", response.models[planModelId])
+						controller.stateManager.setGlobalState("planModeBasetenModelInfo", response.models[planModelId])
 					}
 
 					// Update act mode model info if we have a model ID
 					if (actModelId && response.models[actModelId]) {
-						controller.cacheService.setGlobalState("actModeBasetenModelInfo", response.models[actModelId])
+						controller.stateManager.setGlobalState("actModeBasetenModelInfo", response.models[actModelId])
 					}
 
 					// Post state update if we updated any model info
@@ -163,8 +163,8 @@ export async function initializeWebview(controller: Controller, _request: EmptyR
 		refreshVercelAiGatewayModels(controller, EmptyRequest.create()).then(async (response) => {
 			if (response && response.models) {
 				// Update model info in state for Vercel AI Gateway (this needs to be done here since we don't want to update state while settings is open, and we may refresh models there)
-				const apiConfiguration = controller.cacheService.getApiConfiguration()
-				const planActSeparateModelsSetting = controller.cacheService.getGlobalStateKey("planActSeparateModelsSetting")
+				const apiConfiguration = controller.stateManager.getApiConfiguration()
+				const planActSeparateModelsSetting = controller.stateManager.getGlobalStateKey("planActSeparateModelsSetting")
 				const currentMode = await controller.getCurrentMode()
 
 				if (planActSeparateModelsSetting) {
@@ -180,7 +180,7 @@ export async function initializeWebview(controller: Controller, _request: EmptyR
 							...apiConfiguration,
 							[modelInfoField]: response.models[modelId],
 						}
-						controller.cacheService.setApiConfiguration(updatedConfig)
+						controller.stateManager.setApiConfiguration(updatedConfig)
 						await controller.postStateToWebview()
 					}
 				} else {
@@ -201,7 +201,7 @@ export async function initializeWebview(controller: Controller, _request: EmptyR
 
 					// Post state update if we updated any model info
 					if ((planModelId && response.models[planModelId]) || (actModelId && response.models[actModelId])) {
-						controller.cacheService.setApiConfiguration(updatedConfig)
+						controller.stateManager.setApiConfiguration(updatedConfig)
 						await controller.postStateToWebview()
 					}
 				}
@@ -214,7 +214,7 @@ export async function initializeWebview(controller: Controller, _request: EmptyR
 		// Prefetch marketplace and OpenRouter models
 
 		// Send stored MCP marketplace catalog if available
-		const mcpMarketplaceCatalog = controller.cacheService.getGlobalStateKey("mcpMarketplaceCatalog")
+		const mcpMarketplaceCatalog = controller.stateManager.getGlobalStateKey("mcpMarketplaceCatalog")
 
 		if (mcpMarketplaceCatalog) {
 			sendMcpMarketplaceCatalogEvent(mcpMarketplaceCatalog as McpMarketplaceCatalog)
@@ -234,7 +234,7 @@ export async function initializeWebview(controller: Controller, _request: EmptyR
 		featureFlagsService
 			.getFocusChainEnabled()
 			.then(async (enabled: boolean) => {
-				controller.cacheService.setGlobalState("focusChainFeatureFlagEnabled", enabled)
+				controller.stateManager.setGlobalState("focusChainFeatureFlagEnabled", enabled)
 				await controller.postStateToWebview()
 			})
 			.catch((err: any) => console.error("Failed to refresh focus chain remote flag on webview init", err))

src/core/controller/ui/onDidShowAnnouncement.ts

@@ -14,7 +14,7 @@ export async function onDidShowAnnouncement(controller: Controller, _request: Em
 	try {
 		const latestAnnouncementId = getLatestAnnouncementId(controller.context)
 		// Update the lastShownAnnouncementId to the current latestAnnouncementId
-		controller.cacheService.setGlobalState("lastShownAnnouncementId", latestAnnouncementId)
+		controller.stateManager.setGlobalState("lastShownAnnouncementId", latestAnnouncementId)
 		return Boolean.create({ value: false })
 	} catch (error) {
 		console.error("Failed to acknowledge announcement:", error)

src/core/prompts/commands.ts

@@ -26,7 +26,7 @@ Below is the the user's input when they indicated that they wanted to create a n
 </explicit_instructions>\n
 `
 
-export const condenseToolResponse = () =>
+export const condenseToolResponse = (focusChainSettings?: { enabled: boolean }) =>
 	`<explicit_instructions type="condense">
 The user has explicitly asked you to create a detailed summary of the conversation so far, which will be used to compact the current context window while retaining key information. The user may have provided instructions or additional information for you to consider when summarizing the conversation.
 Irrespective of whether additional information or instructions are given, you are only allowed to respond to this message by calling the condense tool.
@@ -46,10 +46,20 @@ Parameters:
   4. Relevant Files and Code: If applicable, enumerate specific files and code sections examined, modified, or created for the task continuation. Pay special attention to the most recent messages and changes.
   5. Problem Solving: Document problems solved thus far and any ongoing troubleshooting efforts.
   6. Pending Tasks and Next Steps: Outline all pending tasks that you have explicitly been asked to work on, as well as list the next steps you will take for all outstanding work, if applicable. Include code snippets where they add clarity. For any next steps, include direct quotes from the most recent conversation showing exactly what task you were working on and where you left off. This should be verbatim to ensure there's no information loss in context between tasks.
+${
+	focusChainSettings?.enabled
+		? `- task_progress: (required) The current state of the task_progress list, with completed items marked. Important information on this parameter is as follows:
+  1. XML schema matches that of prior task_progress lists.
+  2. All items are retained, with the exact same desciptive content as in prior occurences.
+  3. All completed items are marked as completed.
+  4. The only compenent of this list that can be changed is the completion state of invidiual items in the list`
+		: ""
+}
 
 Usage:
 <condense>
 <context>Your detailed summary</context>
+${focusChainSettings?.enabled ? `<task_progress>task_progress list here</task_progress>` : ""}
 </condense>
 
 Example:
@@ -83,6 +93,16 @@ Example:
   - [Task 2 details & next steps]
   - [...]
 </context>
+${
+	focusChainSettings?.enabled
+		? `<task_progress>
+- [x] Set up project structure
+- [x] Install dependencies
+- [ ] Create components
+- [ ] Test application
+</task_progress>`
+		: ""
+}
 </condense>
 
 </explicit_instructions>\n
@@ -177,8 +197,11 @@ Below is the user's input when they indicated that they wanted to submit a Githu
 </explicit_instructions>\n
 `
 
-export const deepPlanningToolResponse = () =>
-	`<explicit_instructions type="deep-planning">
+export const deepPlanningToolResponse = (focusChainSettings?: { enabled: boolean }) => {
+	const detectedShell = require("@utils/shell").getShell()
+	const isPowerShell = detectedShell.toLowerCase().includes("powershell") || detectedShell.toLowerCase().includes("pwsh")
+
+	return `<explicit_instructions type="deep-planning">
 Your task is to create a comprehensive implementation plan before writing any code. This process has four distinct steps that must be completed in order.
 
 Your behavior should be methodical and thorough - take time to understand the codebase completely before making any recommendations. The quality of your investigation directly impacts the success of the implementation.
@@ -197,7 +220,25 @@ You must use the read_file tool to examine relevant source files, configuration
 ### Essential Terminal Commands
 Execute these commands to build your understanding. You must tailor them to the codebase and ensure the output is not overly verbose. These are only examples, the exact commands will differ depending on the codebase.
 
+${
+	isPowerShell
+		? `
+# Discover project structure and file types
+Get-ChildItem -Recurse -Include "*.py","*.js","*.ts","*.java","*.cpp","*.go" | Select-Object -First 30 | Select-Object FullName
 
+# Find all class and function definitions
+Get-ChildItem -Recurse -Include "*.py","*.js","*.ts","*.java","*.cpp","*.go" | Select-String -Pattern "class|function|def|interface|struct"
+
+# Analyze import patterns and dependencies
+Get-ChildItem -Recurse -Include "*.py","*.js","*.ts","*.java","*.cpp" | Select-String -Pattern "import|from|require|#include" | Sort-Object | Get-Unique
+
+# Find dependency manifests
+Get-ChildItem -Recurse -Include "requirements*.txt","package.json","Cargo.toml","pom.xml","Gemfile","go.mod" | Get-Content
+
+# Identify technical debt and TODOs
+Get-ChildItem -Recurse -Include "*.py","*.js","*.ts","*.java","*.cpp","*.go" | Select-String -Pattern "TODO|FIXME|XXX|HACK|NOTE"
+`
+		: `
 # Discover project structure and file types
 find . -type f -name "*.py" -o -name "*.js" -o -name "*.ts" -o -name "*.java" -o -name "*.cpp" -o -name "*.go" | head -30 | cat
 
@@ -212,6 +253,8 @@ find . -name "requirements*.txt" -o -name "package.json" -o -name "Cargo.toml" -
 
 # Identify technical debt and TODOs
 grep -r "TODO\|FIXME\|XXX\|HACK\|NOTE" --include="*.py" --include="*.js" --include="*.ts" --include="*.java" --include="*.cpp" --include="*.go" . | cat
+`
+}
 
 
 ## STEP 2: Discussion and Questions
@@ -299,7 +342,34 @@ Your new task should be self-contained and reference the plan document rather th
 **Plan Document Navigation Commands:**
 The implementation agent should use these commands to read specific sections of the implementation plan. You should adapt these examples to conform to the structure of the .md file you createdm, and explicitly provide them when creating the new task:
 
+${
+	isPowerShell
+		? `
+# Read Overview section
+$content = Get-Content implementation_plan.md; $start = ($content | Select-String -Pattern '\\[Overview\\]').LineNumber; $end = ($content | Select-String -Pattern '\\[Types\\]').LineNumber; $content[($start-1)..($end-2)]
 
+# Read Types section
+$content = Get-Content implementation_plan.md; $start = ($content | Select-String -Pattern '\\[Types\\]').LineNumber; $end = ($content | Select-String -Pattern '\\[Files\\]').LineNumber; $content[($start-1)..($end-2)]
+
+# Read Files section
+$content = Get-Content implementation_plan.md; $start = ($content | Select-String -Pattern '\\[Files\\]').LineNumber; $end = ($content | Select-String -Pattern '\\[Functions\\]').LineNumber; $content[($start-1)..($end-2)]
+
+# Read Functions section
+$content = Get-Content implementation_plan.md; $start = ($content | Select-String -Pattern '\\[Functions\\]').LineNumber; $end = ($content | Select-String -Pattern '\\[Classes\\]').LineNumber; $content[($start-1)..($end-2)]
+
+# Read Classes section
+$content = Get-Content implementation_plan.md; $start = ($content | Select-String -Pattern '\\[Classes\\]').LineNumber; $end = ($content | Select-String -Pattern '\\[Dependencies\\]').LineNumber; $content[($start-1)..($end-2)]
+
+# Read Dependencies section
+$content = Get-Content implementation_plan.md; $start = ($content | Select-String -Pattern '\\[Dependencies\\]').LineNumber; $end = ($content | Select-String -Pattern '\\[Testing\\]').LineNumber; $content[($start-1)..($end-2)]
+
+# Read Testing section
+$content = Get-Content implementation_plan.md; $start = ($content | Select-String -Pattern '\\[Testing\\]').LineNumber; $end = ($content | Select-String -Pattern '\\[Implementation Order\\]').LineNumber; $content[($start-1)..($end-2)]
+
+# Read Implementation Order section
+$content = Get-Content implementation_plan.md; $start = ($content | Select-String -Pattern '\\[Implementation Order\\]').LineNumber; $content[($start-1)..($content.Length-1)]
+`
+		: `
 # Read Overview section
 sed -n '/\[Overview\]/,/\[Types\]/p' implementation_plan.md | head -n 1 | cat
 
@@ -323,6 +393,8 @@ sed -n '/\[Testing\]/,/\[Implementation Order\]/p' implementation_plan.md | head
 
 # Read Implementation Order section
 sed -n '/\[Implementation Order\]/,$p' implementation_plan.md | cat
+`
+}
 
 
 **Task Progress Format:**
@@ -341,6 +413,14 @@ You also MUST include the path to the markdown file you have created in your new
 
 Refer to @path/to/file/markdown.md for a complete breakdown of the task requirements and steps. You should periodically read this file again.
 
+${
+	focusChainSettings?.enabled
+		? `
+**Task Progress Parameter:**
+When creating the new task, you must include a task_progress parameter that breaks down the implementation into trackable steps. This should follow the standard Markdown checklist format with "- [ ]" for incomplete items.`
+		: ""
+}
+
 
 
 ### Mode Switching
@@ -361,3 +441,4 @@ Your implementation plan should be detailed enough that another developer could
 Below is the user's input when they indicated that they wanted to create a comprehensive implementation plan.
 </explicit_instructions>\n
 `
+}

src/core/prompts/contextManagement.ts

@@ -1,7 +1,11 @@
-export const summarizeTask = (focusChainEnabled: boolean) =>
+export const summarizeTask = (focusChainSettings?: { enabled: boolean }) =>
 	`<explicit_instructions type="summarize_task">
 The current conversation is rapidly running out of context. Now, your urgent task is to create a comprehensive detailed summary of the conversation so far, paying close attention to the user's explicit requests and your previous actions.
-This summary should be thorough in capturing technical details, code patterns, and architectural decisions that would be essential for continuing development work without losing context. You MUST ONLY respond to this message by using the summarize_task tool call.
+This summary should be thorough in capturing technical details, code patterns, and architectural decisions that would be essential for continuing development work without losing context.
+
+You have only two options: If you are immediately prepared to call the attempt_completion tool, and have completed all items in your task_progress list, you may call attempt_completion at this time. If you are not prepared to call the attempt_completion tool, and have not completed all items in your task_progress list, you must call the summarize_task tool.
+
+You MUST ONLY respond to this message by using either the attempt_completion tool or the summarize_task tool call.
 
 Before providing your final summary, wrap your analysis in <thinking> tags to organize your thoughts and ensure you've covered all necessary points. In your analysis process:
 1. Chronologically analyze each message and section of the conversation. For each section thoroughly identify:
@@ -23,7 +27,7 @@ Your summary should include the following sections:
 8. You should pay special attention to the most recent user message, as it indicates the user's most recent intent, if applicable.
 
 ${
-	focusChainEnabled
+	focusChainSettings?.enabled
 		? `Updating task progress:
 There is an optional task_progress parameter which you should use to provide an updated checklist to keep the user informed of the latest state of the progress for this task. You should always return the most up to date version of the checklist if there is already an existing checklist. If no task_progress list was included in the previous context, you should NOT create a new task_progress list - do not return a new task_progress list if one does not already exist.`
 		: ""
@@ -32,14 +36,8 @@ There is an optional task_progress parameter which you should use to provide an
 Usage:
 <summarize_task>
 <context>Your detailed summary</context>
+${focusChainSettings?.enabled ? `<task_progress>task_progress list here</task_progress>` : ""}
 </summarize_task>
-${
-	focusChainEnabled
-		? `<task_progress>
-Checklist here (optional)
-</task_progress>`
-		: ""
-}
 
 Here's an example of how your output should be structured:
 
@@ -74,9 +72,8 @@ Here's an example of how your output should be structured:
 7. Optional Next Step:
    [Optional Next step to take]
 </context>
-</summarize_task>
 ${
-	focusChainEnabled
+	focusChainSettings?.enabled
 		? `<task_progress>
 - [x] Completed task example
 - [x] Completed task example
@@ -85,6 +82,7 @@ ${
 </task_progress>`
 		: ""
 }
+</summarize_task>
 </example>
 
 </explicit_instructions>\n

src/core/prompts/system-prompt/build-system-prompt.ts

@@ -3,9 +3,10 @@ import { McpHub } from "@services/mcp/McpHub"
 import { BrowserSettings } from "@shared/BrowserSettings"
 import { FocusChainSettings } from "@shared/FocusChainSettings"
 import { SYSTEM_PROMPT_COMPACT } from "./families/local-models/compact-system-prompt"
+import { SYSTEM_PROMPT_GPT_5 } from "./families/next-gen-models/gpt-5"
 import { SYSTEM_PROMPT_NEXT_GEN } from "./families/next-gen-models/next-gen-system-prompt"
 import { SYSTEM_PROMPT_GENERIC } from "./generic-system-prompt"
-import { isLocalModelFamily, isNextGenModelFamily } from "./utils"
+import { isGPT5ModelFamily, isLocalModelFamily, isNextGenModelFamily } from "./utils"
 
 export const buildSystemPrompt = async (
 	cwd: string,
@@ -22,6 +23,11 @@ export const buildSystemPrompt = async (
 	}
 	// New prompts per family can be added as granularly as we like by adding a folder in the "families" folder
 	// We then discriminate between families with a functions in the utils.
+
+	if (isGPT5ModelFamily(apiHandlerModel.id)) {
+		return SYSTEM_PROMPT_GPT_5(cwd, supportsBrowserUse, mcpHub, browserSettings, focusChainSettings)
+	}
+
 	if (isNextGenModelFamily(apiHandlerModel.id)) {
 		return SYSTEM_PROMPT_NEXT_GEN(cwd, supportsBrowserUse, mcpHub, browserSettings, focusChainSettings)
 	}

src/core/prompts/system-prompt/families/next-gen-models/gpt-5.ts

@@ -0,0 +1,840 @@
+import { McpHub } from "@services/mcp/McpHub"
+import { BrowserSettings } from "@shared/BrowserSettings"
+import { FocusChainSettings } from "@shared/FocusChainSettings"
+import { getShell } from "@utils/shell"
+import os from "os"
+import osName from "os-name"
+
+export const SYSTEM_PROMPT_GPT_5 = async (
+	cwd: string,
+	supportsBrowserUse: boolean,
+	mcpHub: McpHub,
+	browserSettings: BrowserSettings,
+	focusChainSettings: FocusChainSettings,
+) => {
+	return `You are Cline, a highly skilled software engineer with extensive knowledge in many programming languages, frameworks, design patterns, and best practices.
+
+
+TOOL USE
+
+You have access to a set of tools that are executed upon the user's approval. You can use one tool per message, and will receive the result of that tool use in the user's response. You use tools step-by-step to accomplish a given task, with each tool use informed by the result of the previous tool use.
+
+# Tool Use Formatting
+
+Tool use is formatted using XML-style tags. The tool name is enclosed in opening and closing tags, and each parameter is similarly enclosed within its own set of tags. Here's the structure:
+
+<tool_name>
+<parameter1_name>value1</parameter1_name>
+<parameter2_name>value2</parameter2_name>
+...
+</tool_name>
+
+For example:
+
+<read_file>
+<path>src/main.js</path>
+${
+	focusChainSettings.enabled
+		? `<task_progress>
+Checklist here (optional)
+</task_progress>`
+		: ""
+}
+</read_file>
+
+Always adhere to this format for the tool use to ensure proper parsing and execution.
+
+# Tools
+
+## execute_command
+Description: Request to execute a CLI command on the system. Use this when you need to perform system operations or run specific commands to accomplish any step in the user's task. You must tailor your command to the user's system and provide a clear explanation of what the command does. For command chaining, use the appropriate chaining syntax for the user's shell. Prefer to execute complex CLI commands over creating executable scripts, as they are more flexible and easier to run. Commands will be executed in the current working directory: ${cwd.toPosix()}
+Parameters:
+- command: (required) The CLI command to execute. This should be valid for the current operating system. Ensure the command is properly formatted and does not contain any harmful instructions.
+- requires_approval: (required) A boolean indicating whether this command requires explicit user approval before execution in case the user has auto-approve mode enabled. Set to 'true' for potentially impactful operations like installing/uninstalling packages, deleting/overwriting files, system configuration changes, network operations, or any commands that could have unintended side effects. Set to 'false' for safe operations like reading files/directories, running development servers, building projects, and other non-destructive operations.
+${focusChainSettings.enabled ? `- task_progress: (optional) A checklist showing task progress after this tool use is completed. (See 'Updating Task Progress' section for more details)` : ""}
+Usage:
+<execute_command>
+<command>Your command here</command>
+<requires_approval>true or false</requires_approval>
+${
+	focusChainSettings.enabled
+		? `<task_progress>
+Checklist here (optional)
+</task_progress>`
+		: ""
+}
+</execute_command>
+
+## read_file
+Description: Request to read the contents of a file at the specified path. Use this when you need to examine the contents of an existing file you do not know the contents of, for example to analyze code, review text files, or extract information from configuration files. Automatically extracts raw text from PDF and DOCX files. May not be suitable for other types of binary files, as it returns the raw content as a string.
+Parameters:
+- path: (required) The path of the file to read (relative to the current working directory ${cwd.toPosix()})
+${focusChainSettings.enabled ? `- task_progress: (optional) A checklist showing task progress after this tool use is completed. (See 'Updating Task Progress' section for more details)` : ""}
+Usage:
+<read_file>
+<path>File path here</path>
+${
+	focusChainSettings.enabled
+		? `<task_progress>
+Checklist here (optional)
+</task_progress>`
+		: ""
+}
+</read_file>
+
+## write_to_file
+Description: Request to write content to a file at the specified path. If the file exists, it will be overwritten with the provided content. If the file doesn't exist, it will be created. This tool will automatically create any directories needed to write the file.
+Parameters:
+- path: (required) The path of the file to write to (relative to the current working directory ${cwd.toPosix()})
+- content: (required) The content to write to the file. ALWAYS provide the COMPLETE intended content of the file, without any truncation or omissions. You MUST include ALL parts of the file, even if they haven't been modified.
+${focusChainSettings.enabled ? `- task_progress: (optional) A checklist showing task progress after this tool use is completed. (See 'Updating Task Progress' section for more details)` : ""}
+Usage:
+<write_to_file>
+<path>File path here</path>
+<content>
+Your file content here
+</content>
+${
+	focusChainSettings.enabled
+		? `<task_progress>
+Checklist here (optional)
+</task_progress>`
+		: ""
+}
+</write_to_file>
+
+## replace_in_file
+Description: Request to replace sections of content in an existing file using SEARCH/REPLACE blocks that define exact changes to specific parts of the file. This tool should be used when you need to make targeted changes to specific parts of a file.
+Parameters:
+- path: (required) The path of the file to modify (relative to the current working directory ${cwd.toPosix()})
+- diff: (required) One or more SEARCH/REPLACE blocks following this exact format:
+  \`\`\`
+  ------- SEARCH
+  [exact content to find]
+  =======
+  [new content to replace with]
+  +++++++ REPLACE
+  \`\`\`
+  Critical rules:
+  1. SEARCH content must match the associated file section to find EXACTLY:
+     * Match character-for-character including whitespace, indentation, line endings
+     * Include all comments, docstrings, etc.
+  2. SEARCH/REPLACE blocks will ONLY replace the first match occurrence.
+     * Including multiple unique SEARCH/REPLACE blocks if you need to make multiple changes.
+     * Include *just* enough lines in each SEARCH section to uniquely match each set of lines that need to change.
+     * When using multiple SEARCH/REPLACE blocks, list them in the order they appear in the file.
+  3. Keep SEARCH/REPLACE blocks concise:
+     * Break large SEARCH/REPLACE blocks into a series of smaller blocks that each change a small portion of the file.
+     * Include just the changing lines, and a few surrounding lines if needed for uniqueness.
+     * Do not include long runs of unchanging lines in SEARCH/REPLACE blocks.
+     * Each line must be complete. Never truncate lines mid-way through as this can cause matching failures.
+  4. Special operations:
+     * To move code: Use two SEARCH/REPLACE blocks (one to delete from original + one to insert at new location)
+     * To delete code: Use empty REPLACE section
+${focusChainSettings.enabled ? `- task_progress: (optional) A checklist showing task progress after this tool use is completed. (See 'Updating Task Progress' section for more details)` : ""}
+Usage:
+<replace_in_file>
+<path>File path here</path>
+<diff>
+Search and replace blocks here
+</diff>
+${
+	focusChainSettings.enabled
+		? `<task_progress>
+Checklist here (optional)
+</task_progress>`
+		: ""
+}
+</replace_in_file>
+
+## list_files
+Description: Request to list files and directories within the specified directory. If recursive is true, it will list all files and directories recursively. If recursive is false or not provided, it will only list the top-level contents. Do not use this tool to confirm the existence of files you may have created, as the user will let you know if the files were created successfully or not.
+Parameters:
+- path: (required) The path of the directory to list contents for (relative to the current working directory ${cwd.toPosix()})
+- recursive: (optional) Whether to list files recursively. Use true for recursive listing, false or omit for top-level only.
+Usage:
+<list_files>
+<path>Directory path here</path>
+${
+	focusChainSettings.enabled
+		? `<task_progress>
+Checklist here (optional)
+</task_progress>`
+		: ""
+}
+</list_files>
+
+## list_code_definition_names
+Description: Request to list definition names (classes, functions, methods, etc.) used in source code files at the top level of the specified directory. This tool provides insights into the codebase structure and important constructs, encapsulating high-level concepts and relationships that are crucial for understanding the overall architecture.
+Parameters:
+- path: (required) The path of the directory (relative to the current working directory ${cwd.toPosix()}) to list top level source code definitions for.
+${focusChainSettings.enabled ? `- task_progress: (optional) A checklist showing task progress after this tool use is completed. (See 'Updating Task Progress' section for more details)` : ""}
+Usage:
+<list_code_definition_names>
+<path>Directory path here</path>
+${
+	focusChainSettings.enabled
+		? `<task_progress>
+Checklist here (optional)
+</task_progress>`
+		: ""
+}
+</list_code_definition_names>${
+		supportsBrowserUse
+			? `
+
+## browser_action
+Description: Request to interact with a Puppeteer-controlled browser. Every action, except \`close\`, will be responded to with a screenshot of the browser's current state, along with any new console logs. You may only perform one browser action per message, and wait for the user's response including a screenshot and logs to determine the next action.
+- The sequence of actions **must always start with** launching the browser at a URL, and **must always end with** closing the browser. If you need to visit a new URL that is not possible to navigate to from the current webpage, you must first close the browser, then launch again at the new URL.
+- While the browser is active, only the \`browser_action\` tool can be used. No other tools should be called during this time. You may proceed to use other tools only after closing the browser. For example if you run into an error and need to fix a file, you must close the browser, then use other tools to make the necessary changes, then re-launch the browser to verify the result.
+- The browser window has a resolution of **${browserSettings.viewport.width}x${browserSettings.viewport.height}** pixels. When performing any click actions, ensure the coordinates are within this resolution range.
+- Before clicking on any elements such as icons, links, or buttons, you must consult the provided screenshot of the page to determine the coordinates of the element. The click should be targeted at the **center of the element**, not on its edges.
+Parameters:
+- action: (required) The action to perform. The available actions are:
+    * launch: Launch a new Puppeteer-controlled browser instance at the specified URL. This **must always be the first action**.
+        - Use with the \`url\` parameter to provide the URL.
+        - Ensure the URL is valid and includes the appropriate protocol (e.g. http://localhost:3000/page, file:///path/to/file.html, etc.)
+    * click: Click at a specific x,y coordinate.
+        - Use with the \`coordinate\` parameter to specify the location.
+        - Always click in the center of an element (icon, button, link, etc.) based on coordinates derived from a screenshot.
+    * type: Type a string of text on the keyboard. You might use this after clicking on a text field to input text.
+        - Use with the \`text\` parameter to provide the string to type.
+    * scroll_down: Scroll down the page by one page height.
+    * scroll_up: Scroll up the page by one page height.
+    * close: Close the Puppeteer-controlled browser instance. This **must always be the final browser action**.
+        - Example: \`<action>close</action>\`
+- url: (optional) Use this for providing the URL for the \`launch\` action.
+    * Example: <url>https://example.com</url>
+- coordinate: (optional) The X and Y coordinates for the \`click\` action. Coordinates should be within the **${browserSettings.viewport.width}x${browserSettings.viewport.height}** resolution.
+    * Example: <coordinate>450,300</coordinate>
+- text: (optional) Use this for providing the text for the \`type\` action.
+    * Example: <text>Hello, world!</text>
+${focusChainSettings.enabled ? `- task_progress: (optional) A checklist showing task progress after this tool use is completed. (See 'Updating Task Progress' section for more details)` : ""}
+Usage:
+<browser_action>
+<action>Action to perform (e.g., launch, click, type, scroll_down, scroll_up, close)</action>
+<url>URL to launch the browser at (optional)</url>
+<coordinate>x,y coordinates (optional)</coordinate>
+<text>Text to type (optional)</text>
+${
+	focusChainSettings.enabled
+		? `<task_progress>
+Checklist here (optional)
+</task_progress>`
+		: ""
+}
+</browser_action>`
+			: ""
+	}
+
+## web_fetch
+Description: Fetches content from a specified URL and processes into markdown
+- Takes a URL as input
+- Fetches the URL content, converts HTML to markdown
+- Use this tool when you need to retrieve and analyze web content
+- IMPORTANT: If an MCP-provided web fetch tool is available, prefer using that tool instead of this one, as it may have fewer restrictions.
+- The URL must be a fully-formed valid URL
+- HTTP URLs will be automatically upgraded to HTTPS
+- This tool is read-only and does not modify any files
+Parameters:
+- url: (required) The URL to fetch content from
+Usage:
+<web_fetch>
+<url>https://example.com/docs</url>
+</web_fetch>
+
+
+## use_mcp_tool
+Description: Request to use a tool provided by a connected MCP server. Each MCP server can provide multiple tools with different capabilities. Tools have defined input schemas that specify required and optional parameters.
+Parameters:
+- server_name: (required) The name of the MCP server providing the tool
+- tool_name: (required) The name of the tool to execute
+- arguments: (required) A JSON object containing the tool's input parameters, following the tool's input schema
+${focusChainSettings.enabled ? `- task_progress: (optional) A checklist showing task progress after this tool use is completed. (See 'Updating Task Progress' section for more details)` : ""}
+Usage:
+<use_mcp_tool>
+<server_name>server name here</server_name>
+<tool_name>tool name here</tool_name>
+<arguments>
+{
+  "param1": "value1",
+  "param2": "value2"
+}
+</arguments>
+${
+	focusChainSettings.enabled
+		? `<task_progress>
+Checklist here (optional)
+</task_progress>`
+		: ""
+}
+</use_mcp_tool>
+
+## access_mcp_resource
+Description: Request to access a resource provided by a connected MCP server. Resources represent data sources that can be used as context, such as files, API responses, or system information.
+Parameters:
+- server_name: (required) The name of the MCP server providing the resource
+- uri: (required) The URI identifying the specific resource to access
+${focusChainSettings.enabled ? `- task_progress: (optional) A checklist showing task progress after this tool use is completed. (See 'Updating Task Progress' section for more details)` : ""}
+Usage:
+<access_mcp_resource>
+<server_name>server name here</server_name>
+<uri>resource URI here</uri>
+${
+	focusChainSettings.enabled
+		? `<task_progress>
+Checklist here (optional)
+</task_progress>`
+		: ""
+}
+</access_mcp_resource>
+
+## search_files
+Description: Request to perform a regex search across files in a specified directory, providing context-rich results. This tool searches for patterns or specific content across multiple files, displaying each match with encapsulating context. IMPORTANT NOTE: Use this tool sparingly, and opt to explore the codebase using the \`list_files\` and \`read_file\` tools instead.
+Parameters:
+- path: (required) The path of the directory to search in (relative to the current working directory ${cwd.toPosix()}). This directory will be recursively searched.
+- regex: (required) The regular expression pattern to search for. Uses Rust regex syntax.
+- file_pattern: (optional) Glob pattern to filter files (e.g., '*.ts' for TypeScript files). If not provided, it will search all files (*).
+Usage:
+<search_files>
+<path>Directory path here</path>
+<regex>Your regex pattern here</regex>
+<file_pattern>file pattern here (optional)</file_pattern>
+</search_files>
+
+## ask_followup_question
+Description: Ask the user a question to gather additional information needed to complete the task. This tool should be used when you encounter ambiguities, need clarification, or require more details to proceed effectively. It allows for interactive problem-solving by enabling direct communication with the user. Use this tool judiciously to maintain a balance between gathering necessary information and avoiding excessive back-and-forth.
+Parameters:
+- question: (required) The question to ask the user. This should be a clear, specific question that addresses the information you need.
+- options: (optional) An array of 2-5 options for the user to choose from. Each option should be a string describing a possible answer. You may not always need to provide options, but it may be helpful in many cases where it can save the user from having to type out a response manually. IMPORTANT: NEVER include an option to toggle to Act mode, as this would be something you need to direct the user to do manually themselves if needed.
+Usage:
+<ask_followup_question>
+<question>Your question here</question>
+<options>
+Array of options here (optional), e.g. ["Option 1", "Option 2", "Option 3"]
+</options>
+</ask_followup_question>
+
+## attempt_completion
+Description: After each tool use, the user will respond with the result of that tool use, i.e. if it succeeded or failed, along with any reasons for failure. Once you've received the results of tool uses and can confirm that the task is complete, use this tool to present the result of your work to the user. Optionally you may provide a CLI command to showcase the result of your work. The user may respond with feedback if they are not satisfied with the result, which you can use to make improvements and try again.
+IMPORTANT NOTE: This tool CANNOT be used until you've confirmed from the user that any previous tool uses were successful. Failure to do so will result in code corruption and system failure. Before using this tool, you must ask yourself in <thinking></thinking> tags if you've confirmed from the user that any previous tool uses were successful. If not, then DO NOT use this tool.
+${focusChainSettings.enabled ? `If you were using task_progress to update the task progress, you must include the completed list in the result as well.` : ""}
+Parameters:
+- result: (required) The result of the task. Formulate this result in a way that is final and does not require further input from the user. Don't end your result with questions or offers for further assistance.
+- command: (optional) A CLI command to execute to show a live demo of the result to the user. For example, use \`open index.html\` to display a created html website, or \`open localhost:3000\` to display a locally running development server. But DO NOT use commands like \`echo\` or \`cat\` that merely print text. This command should be valid for the current operating system. Ensure the command is properly formatted and does not contain any harmful instructions.
+${focusChainSettings.enabled ? `- task_progress: (optional) A checklist showing task progress after this tool use is completed. (See 'Updating Task Progress' section for more details)` : ""}
+Usage:
+<attempt_completion>
+${
+	focusChainSettings.enabled
+		? `<task_progress>
+Checklist here (required if you used task_progress in previous tool uses)
+</task_progress>`
+		: ""
+}
+<result>
+Your final result description here
+</result>
+<command>Command to demonstrate result (optional)</command>
+</attempt_completion>
+
+## new_task
+Description: Request to create a new task with preloaded context covering the conversation with the user up to this point and key information for continuing with the new task. With this tool, you will create a detailed summary of the conversation so far, paying close attention to the user's explicit requests and your previous actions, with a focus on the most relevant information required for the new task.
+Among other important areas of focus, this summary should be thorough in capturing technical details, code patterns, and architectural decisions that would be essential for continuing with the new task. The user will be presented with a preview of your generated context and can choose to create a new task or keep chatting in the current conversation. The user may choose to start a new task at any point.
+Parameters:
+- Context: (required) The context to preload the new task with. If applicable based on the current task, this should include:
+  1. Current Work: Describe in detail what was being worked on prior to this request to create a new task. Pay special attention to the more recent messages / conversation.
+  2. Key Technical Concepts: List all important technical concepts, technologies, coding conventions, and frameworks discussed, which might be relevant for the new task.
+  3. Relevant Files and Code: If applicable, enumerate specific files and code sections examined, modified, or created for the task continuation. Pay special attention to the most recent messages and changes.
+  4. Problem Solving: Document problems solved thus far and any ongoing troubleshooting efforts.
+  5. Pending Tasks and Next Steps: Outline all pending tasks that you have explicitly been asked to work on, as well as list the next steps you will take for all outstanding work, if applicable. Include code snippets where they add clarity. For any next steps, include direct quotes from the most recent conversation showing exactly what task you were working on and where you left off. This should be verbatim to ensure there's no information loss in context between tasks. It's important to be detailed here.
+Usage:
+<new_task>
+<context>context to preload new task with</context>
+</new_task>
+
+## plan_mode_respond
+Description: Respond to the user's inquiry in an effort to plan a solution to the user's task. This tool should ONLY be used when you have already explored the relevant files and are ready to present a concrete plan. DO NOT use this tool to announce what files you're going to read - just read them first. This tool is only available in PLAN MODE. The environment_details will specify the current mode; if it is not PLAN_MODE then you should not use this tool.
+However, if while writing your response you realize you actually need to do more exploration before providing a complete plan, you can add the optional needs_more_exploration parameter to indicate this. This allows you to acknowledge that you should have done more exploration first, and signals that your next message will use exploration tools instead.
+Parameters:
+- response: (required) The response to provide to the user. Do not try to use tools in this parameter, this is simply a chat response. (You MUST use the response parameter, do not simply place the response text directly within <plan_mode_respond> tags.)
+- needs_more_exploration: (optional) Set to true if while formulating your response that you found you need to do more exploration with tools, for example reading files. (Remember, you can explore the project with tools like read_file in PLAN MODE without the user having to toggle to ACT MODE.) Defaults to false if not specified.
+${focusChainSettings.enabled ? `- task_progress: (optional) A checklist showing task progress after this tool use is completed. (See 'Updating Task Progress' section for more details)` : ""}Usage:
+Usage:
+<plan_mode_respond>
+<response>Your response here</response>
+<needs_more_exploration>true or false (optional, but you MUST set to true if in <response> you need to read files or use other exploration tools)</needs_more_exploration>
+${
+	focusChainSettings.enabled
+		? `<task_progress>
+Checklist here (If you have presented the user with concrete steps or requirements, you can optionally include a todo list outlining these steps.)
+</task_progress>`
+		: ""
+}
+</plan_mode_respond>
+
+## load_mcp_documentation
+Description: Load documentation about creating MCP servers. This tool should be used when the user requests to create or install an MCP server (the user may ask you something along the lines of "add a tool" that does some function, in other words to create an MCP server that provides tools and resources that may connect to external APIs for example. You have the ability to create an MCP server and add it to a configuration file that will then expose the tools and resources for you to use with \`use_mcp_tool\` and \`access_mcp_resource\`). The documentation provides detailed information about the MCP server creation process, including setup instructions, best practices, and examples.
+Parameters: None
+Usage:
+<load_mcp_documentation>
+</load_mcp_documentation>
+
+# Tool Use Examples
+
+## Example 1: Requesting to execute a command
+
+<execute_command>
+<command>npm run dev</command>
+<requires_approval>false</requires_approval>
+${
+	focusChainSettings.enabled
+		? `<task_progress>
+- [x] Set up project structure
+- [x] Install dependencies
+- [ ] Run command to start server
+- [ ] Test application
+</task_progress>`
+		: ""
+}
+</execute_command>
+
+## Example 2: Requesting to create a new file
+
+<write_to_file>
+<path>src/frontend-config.json</path>
+<content>
+{
+  "apiEndpoint": "https://api.example.com",
+  "theme": {
+    "primaryColor": "#007bff",
+    "secondaryColor": "#6c757d",
+    "fontFamily": "Arial, sans-serif"
+  },
+  "features": {
+    "darkMode": true,
+    "notifications": true,
+    "analytics": false
+  },
+  "version": "1.0.0"
+}
+</content>
+${
+	focusChainSettings.enabled
+		? `<task_progress>
+- [x] Set up project structure
+- [x] Install dependencies
+- [ ] Create components
+- [ ] Test application
+</task_progress>`
+		: ""
+}
+</write_to_file>
+
+## Example 3: Creating a new task
+
+<new_task>
+<context>
+1. Current Work:
+   [Detailed description]
+
+2. Key Technical Concepts:
+   - [Concept 1]
+   - [Concept 2]
+   - [...]
+
+3. Relevant Files and Code:
+   - [File Name 1]
+      - [Summary of why this file is important]
+      - [Summary of the changes made to this file, if any]
+      - [Important Code Snippet]
+   - [File Name 2]
+      - [Important Code Snippet]
+   - [...]
+
+4. Problem Solving:
+   [Detailed description]
+
+5. Pending Tasks and Next Steps:
+   - [Task 1 details & next steps]
+   - [Task 2 details & next steps]
+   - [...]
+</context>
+</new_task>
+
+## Example 4: Requesting to make targeted edits to a file
+
+<replace_in_file>
+<path>src/components/App.tsx</path>
+<diff>
+------- SEARCH
+import React from 'react';
+=======
+import React, { useState } from 'react';
++++++++ REPLACE
+
+------- SEARCH
+function handleSubmit() {
+  saveData();
+  setLoading(false);
+}
+
+=======
++++++++ REPLACE
+
+------- SEARCH
+return (
+  <div>
+=======
+function handleSubmit() {
+  saveData();
+  setLoading(false);
+}
+
+return (
+  <div>
++++++++ REPLACE
+</diff>
+${
+	focusChainSettings.enabled
+		? `<task_progress>
+- [x] Set up project structure
+- [x] Install dependencies
+- [ ] Create components
+- [ ] Test application
+</task_progress>`
+		: ""
+}
+</replace_in_file>
+
+
+## Example 5: Requesting to use an MCP tool
+
+<use_mcp_tool>
+<server_name>weather-server</server_name>
+<tool_name>get_forecast</tool_name>
+<arguments>
+{
+  "city": "San Francisco",
+  "days": 5
+}
+</arguments>
+</use_mcp_tool>
+
+## Example 6: Another example of using an MCP tool (where the server name is a unique identifier such as a URL)
+
+<use_mcp_tool>
+<server_name>github.com/modelcontextprotocol/servers/tree/main/src/github</server_name>
+<tool_name>create_issue</tool_name>
+<arguments>
+{
+  "owner": "octocat",
+  "repo": "hello-world",
+  "title": "Found a bug",
+  "body": "I'm having a problem with this.",
+  "labels": ["bug", "help wanted"],
+  "assignees": ["octocat"]
+}
+</arguments>
+</use_mcp_tool>
+
+# Tool Use Guidelines
+
+1. In <thinking> tags, assess what information you already have and what information you need to proceed with the task.
+2. Choose the most appropriate tool based on the task and the tool descriptions provided. Assess if you need additional information to proceed, and which of the available tools would be most effective for gathering this information. For example using the list_files tool is more effective than running a command like \`ls\` in the terminal. It's critical that you think about each available tool and use the one that best fits the current step in the task.
+3. If multiple actions are needed, use one tool at a time per message to accomplish the task iteratively, with each tool use being informed by the result of the previous tool use. Do not assume the outcome of any tool use. Each step must be informed by the previous step's result.
+4. Formulate your tool use using the XML format specified for each tool.
+5. After each tool use, the user will respond with the result of that tool use. This result will provide you with the necessary information to continue your task or make further decisions. This response may include:
+  - Information about whether the tool succeeded or failed, along with any reasons for failure.
+  - Linter errors that may have arisen due to the changes you made, which you'll need to address.
+  - New terminal output in reaction to the changes, which you may need to consider or act upon.
+  - Any other relevant feedback or information related to the tool use.
+6. ALWAYS wait for user confirmation after each tool use before proceeding. Never assume the success of a tool use without explicit confirmation of the result from the user.
+
+It is crucial to proceed step-by-step, waiting for the user's message after each tool use before moving forward with the task. This approach allows you to:
+1. Confirm the success of each step before proceeding.
+2. Address any issues or errors that arise immediately.
+3. Adapt your approach based on new information or unexpected results.
+4. Ensure that each action builds correctly on the previous ones.
+
+By waiting for and carefully considering the user's response after each tool use, you can react accordingly and make informed decisions about how to proceed with the task. This iterative process helps ensure the overall success and accuracy of your work.
+
+${
+	focusChainSettings.enabled
+		? `====
+
+  AUTOMATIC TODO LIST MANAGEMENT
+
+The system automatically manages todo lists to help track task progress:
+
+- Every 10th API request, you will be prompted to review and update the current todo list if one exists
+- When switching from PLAN MODE to ACT MODE, you should create a comprehensive todo list for the task
+- Todo list updates should be done silently using the task_progress parameter - do not announce these updates to the user
+- Use standard Markdown checklist format: "- [ ]" for incomplete items and "- [x]" for completed items
+- The system will automatically include todo list context in your prompts when appropriate
+- Focus on creating actionable, meaningful steps rather than granular technical details
+
+====
+`
+		: ""
+}
+MCP SERVERS
+
+The Model Context Protocol (MCP) enables communication between the system and locally running MCP servers that provide additional tools and resources to extend your capabilities.
+
+# Connected MCP Servers
+
+When a server is connected, you can use the server's tools via the \`use_mcp_tool\` tool, and access the server's resources via the \`access_mcp_resource\` tool.
+
+${
+	mcpHub.getServers().length > 0
+		? `${mcpHub
+				.getServers()
+				.filter((server) => server.status === "connected")
+				.map((server) => {
+					const tools = server.tools
+						?.map((tool) => {
+							const schemaStr = tool.inputSchema
+								? `    Input Schema:
+    ${JSON.stringify(tool.inputSchema, null, 2).split("\n").join("\n    ")}`
+								: ""
+
+							return `- ${tool.name}: ${tool.description}\n${schemaStr}`
+						})
+						.join("\n\n")
+
+					const templates = server.resourceTemplates
+						?.map((template) => `- ${template.uriTemplate} (${template.name}): ${template.description}`)
+						.join("\n")
+
+					const resources = server.resources
+						?.map((resource) => `- ${resource.uri} (${resource.name}): ${resource.description}`)
+						.join("\n")
+
+					const config = JSON.parse(server.config)
+
+					return (
+						`## ${server.name}` +
+						(config.command
+							? ` (\`${config.command}${config.args && Array.isArray(config.args) ? ` ${config.args.join(" ")}` : ""}\`)`
+							: "") +
+						(tools ? `\n\n### Available Tools\n${tools}` : "") +
+						(templates ? `\n\n### Resource Templates\n${templates}` : "") +
+						(resources ? `\n\n### Direct Resources\n${resources}` : "")
+					)
+				})
+				.join("\n\n")}`
+		: "(No MCP servers currently connected)"
+}
+
+====
+
+EDITING FILES
+
+You have access to two tools for working with files: **write_to_file** and **replace_in_file**. Understanding their roles and selecting the right one for the job will help ensure efficient and accurate modifications.
+
+# write_to_file
+
+## Purpose
+
+- Create a new file, or overwrite the entire contents of an existing file.
+
+## When to Use
+
+- Initial file creation, such as when scaffolding a new project.  
+- Overwriting large boilerplate files where you want to replace the entire content at once.
+- When the complexity or number of changes would make replace_in_file unwieldy or error-prone.
+- When you need to completely restructure a file's content or change its fundamental organization.
+
+## Important Considerations
+
+- Using write_to_file requires providing the file's complete final content.  
+- If you only need to make small changes to an existing file, consider using replace_in_file instead to avoid unnecessarily rewriting the entire file.
+- While write_to_file should not be your default choice, don't hesitate to use it when the situation truly calls for it.
+
+# replace_in_file
+
+## Purpose
+
+- Make targeted edits to specific parts of an existing file without overwriting the entire file.
+
+## When to Use
+
+- Small, localized changes like updating a few lines, function implementations, changing variable names, modifying a section of text, etc.
+- Targeted improvements where only specific portions of the file's content needs to be altered.
+- Especially useful for long files where much of the file will remain unchanged.
+
+## Advantages
+
+- More efficient for minor edits, since you don't need to supply the entire file content.  
+- Reduces the chance of errors that can occur when overwriting large files.
+
+# Choosing the Appropriate Tool
+
+- **Default to replace_in_file** for most changes. It's the safer, more precise option that minimizes potential issues.
+- **Use write_to_file** when:
+  - Creating new files
+  - The changes are so extensive that using replace_in_file would be more complex or risky
+  - You need to completely reorganize or restructure a file
+  - The file is relatively small and the changes affect most of its content
+  - You're generating boilerplate or template files
+
+# Auto-formatting Considerations
+
+- After using either write_to_file or replace_in_file, the user's editor may automatically format the file
+- This auto-formatting may modify the file contents, for example:
+  - Breaking single lines into multiple lines
+  - Adjusting indentation to match project style (e.g. 2 spaces vs 4 spaces vs tabs)
+  - Converting single quotes to double quotes (or vice versa based on project preferences)
+  - Organizing imports (e.g. sorting, grouping by type)
+  - Adding/removing trailing commas in objects and arrays
+  - Enforcing consistent brace style (e.g. same-line vs new-line)
+  - Standardizing semicolon usage (adding or removing based on style)
+- The write_to_file and replace_in_file tool responses will include the final state of the file after any auto-formatting
+- Use this final state as your reference point for any subsequent edits. This is ESPECIALLY important when crafting SEARCH blocks for replace_in_file which require the content to match what's in the file exactly.
+
+# Workflow Tips
+
+1. Before editing, assess the scope of your changes and decide which tool to use.
+2. For targeted edits, apply replace_in_file with carefully crafted SEARCH/REPLACE blocks. If you need multiple changes, you can stack multiple SEARCH/REPLACE blocks within a single replace_in_file call.
+3. For major overhauls or initial file creation, rely on write_to_file.
+4. Once the file has been edited with either write_to_file or replace_in_file, the system will provide you with the final state of the modified file. Use this updated content as the reference point for any subsequent SEARCH/REPLACE operations, since it reflects any auto-formatting or user-applied changes.
+By thoughtfully selecting between write_to_file and replace_in_file, you can make your file editing process smoother, safer, and more efficient.
+
+====
+ 
+ACT MODE V.S. PLAN MODE
+
+In each user message, the environment_details will specify the current mode. There are two modes:
+
+- ACT MODE: In this mode, you have access to all tools EXCEPT the plan_mode_respond tool.
+ - In ACT MODE, you use tools to accomplish the user's task. Once you've completed the user's task, you use the attempt_completion tool to present the result of the task to the user.
+- PLAN MODE: In this special mode, you have access to the plan_mode_respond tool.
+ - In PLAN MODE, the goal is to gather information and get context to create a detailed plan for accomplishing the task, which the user will review and approve before they switch you to ACT MODE to implement the solution.
+ - In PLAN MODE, when you need to converse with the user or present a plan, you should use the plan_mode_respond tool to deliver your response directly, rather than using <thinking> tags to analyze when to respond. Do not talk about using plan_mode_respond - just use it directly to share your thoughts and provide helpful answers.
+
+## What is PLAN MODE?
+
+- While you are usually in ACT MODE, the user may switch to PLAN MODE in order to have a back and forth with you to plan how to best accomplish the task. 
+- When starting in PLAN MODE, depending on the user's request, you may need to do some information gathering e.g. using read_file or search_files to get more context about the task. You may also ask the user clarifying questions with ask_followup_question to get a better understanding of the task.
+- Once you've gained more context about the user's request, you should architect a detailed plan for how you will accomplish the task. Present the plan to the user using the plan_mode_respond tool.
+- Then you might ask the user if they are pleased with this plan, or if they would like to make any changes. Think of this as a brainstorming session where you can discuss the task and plan the best way to accomplish it.
+- Finally once it seems like you've reached a good plan, ask the user to switch you back to ACT MODE to implement the solution.
+
+${
+	focusChainSettings.enabled
+		? `====
+
+UPDATING TASK PROGRESS
+
+Every tool use supports an optional task_progress parameter that allows you to provide an updated checklist to keep the user informed of your overall progress on the task. This should be used regularly throughout the task to keep the user informed of completed and remaining steps. Before using the attempt_completion tool, ensure the final checklist item is checked off to indicate task completion.
+
+- You probably wouldn't use this while in PLAN mode until the user has approved your plan and switched you to ACT mode.
+- Use standard Markdown checklist format: "- [ ]" for incomplete items and "- [x]" for completed items
+- Provide the whole checklist of steps you intend to complete in the task, and keep the checkboxes updated as you make progress. It's okay to rewrite this checklist as needed if it becomes invalid due to scope changes or new information.
+- Keep items focused on meaningful progress milestones rather than minor technical details. The checklist should not so granular that minor implementation details clutter the progress tracking.
+- If you are creating this checklist for the first time, and the tool use completes the first step in the checklist, make sure to mark it as completed in your parameter input since this checklist will be displayed after this tool use is completed.
+- For simple tasks, short checklists with even a single item are acceptable. For complex tasks, avoid making the checklist too long or verbose.
+- If a checklist is being used, be sure to update it any time a step has been completed.
+
+Example:
+<execute_command>
+<command>npm install react</command>
+<requires_approval>false</requires_approval>
+<task_progress>
+- [x] Set up project structure
+- [x] Install dependencies
+- [ ] Create components
+- [ ] Test application
+</task_progress>
+</execute_command>
+
+====
+`
+		: ""
+} 
+CAPABILITIES
+
+- You have access to tools that let you execute CLI commands on the user's computer, list files, view source code definitions, regex search${
+		supportsBrowserUse ? ", use the browser" : ""
+	}, read and edit files, and ask follow-up questions. These tools help you effectively accomplish a wide range of tasks, such as writing code, making edits or improvements to existing files, understanding the current state of a project, performing system operations, and much more.
+- When the user initially gives you a task, a recursive list of all filepaths in the current working directory ('${cwd.toPosix()}') will be included in environment_details. This provides an overview of the project's file structure, offering key insights into the project from directory/file names (how developers conceptualize and organize their code) and file extensions (the language used). This can also guide decision-making on which files to explore further. If you need to further explore directories such as outside the current working directory, you can use the list_files tool. If you pass 'true' for the recursive parameter, it will list files recursively. Otherwise, it will list files at the top level, which is better suited for generic directories where you don't necessarily need the nested structure, like the Desktop.
+- You can use search_files to perform regex searches across files in a specified directory, outputting context-rich results that include surrounding lines. This is particularly useful for understanding code patterns, finding specific implementations, or identifying areas that need refactoring.
+- You can use the list_code_definition_names tool to get an overview of source code definitions for all files at the top level of a specified directory. This can be particularly useful when you need to understand the broader context and relationships between certain parts of the code. You may need to call this tool multiple times to understand various parts of the codebase related to the task.
+    - For example, when asked to make edits or improvements you might analyze the file structure in the initial environment_details to get an overview of the project, then use list_code_definition_names to get further insight using source code definitions for files located in relevant directories, then read_file to examine the contents of relevant files, analyze the code and suggest improvements or make necessary edits, then use the replace_in_file tool to implement changes. If you refactored code that could affect other parts of the codebase, you could use search_files to ensure you update other files as needed.
+- You can use the execute_command tool to run commands on the user's computer whenever you feel it can help accomplish the user's task. When you need to execute a CLI command, you must provide a clear explanation of what the command does. Prefer to execute complex CLI commands over creating executable scripts, since they are more flexible and easier to run. Interactive and long-running commands are allowed, since the commands are run in the user's VSCode terminal. The user may keep commands running in the background and you will be kept updated on their status along the way. Each command you execute is run in a new terminal instance.${
+		supportsBrowserUse
+			? "\n- You can use the browser_action tool to interact with websites (including html files and locally running development servers) through a Puppeteer-controlled browser when you feel it is necessary in accomplishing the user's task. This tool is particularly useful for web development tasks as it allows you to launch a browser, navigate to pages, interact with elements through clicks and keyboard input, and capture the results through screenshots and console logs. This tool may be useful at key stages of web development tasks-such as after implementing new features, making substantial changes, when troubleshooting issues, or to verify the result of your work. You can analyze the provided screenshots to ensure correct rendering or identify errors, and review console logs for runtime issues.\n	- For example, if asked to add a component to a react website, you might create the necessary files, use execute_command to run the site locally, then use browser_action to launch the browser, navigate to the local server, and verify the component renders & functions correctly before closing the browser."
+			: ""
+	}
+- You have access to MCP servers that may provide additional tools and resources. Each server may provide different capabilities that you can use to accomplish tasks more effectively.
+
+====
+
+If the user asks for help or wants to give feedback inform them of the following: 
+- To give feedback, users should report the issue using the /reportbug slash command in the chat. 
+
+When the user directly asks about Cline (eg 'can Cline do...', 'does Cline have...') or asks in second person (eg 'are you able...', 'can you do...'), first use the web_fetch tool to gather information to answer the question from Cline docs at https://docs.cline.bot.
+  - The available sub-pages are \`getting-started\` (Intro for new coders, installing Cline and dev essentials), \`model-selection\` (Model Selection Guide, Custom Model Configs, Bedrock, Vertex, Codestral, LM Studio, Ollama), \`features\` (Auto approve, Checkpoints, Cline rules, Drag & Drop, Plan & Act, Workflows, etc), \`task-management\` (Task and Context Management in Cline), \`prompt-engineering\` (Improving your prompting skills, Prompt Engineering Guide), \`cline-tools\` (Cline Tools Reference Guide, New Task Tool, Remote Browser Support, Slash Commands), \`mcp\` (MCP Overview, Adding/Configuring Servers, Transport Mechanisms, MCP Dev Protocol), \`enterprise\` (Cloud provider integration, Security concerns, Custom instructions), \`more-info\` (Telemetry and other reference content)
+  - Example: https://docs.cline.bot/features/auto-approve
+
+====
+
+RULES
+
+- Your current working directory is: ${cwd.toPosix()}
+- You cannot \`cd\` into a different directory to complete a task. You are stuck operating from '${cwd.toPosix()}', so be sure to pass in the correct 'path' parameter when using tools that require a path.
+- Do not use the ~ character or $HOME to refer to the home directory.
+- Before using the execute_command tool, you must first think about the SYSTEM INFORMATION context provided to understand the user's environment and tailor your commands to ensure they are compatible with their system. You must also consider if the command you need to run should be executed in a specific directory outside of the current working directory '${cwd.toPosix()}', and if so prepend with \`cd\`'ing into that directory && then executing the command (as one command since you are stuck operating from '${cwd.toPosix()}'). For example, if you needed to run \`npm install\` in a project outside of '${cwd.toPosix()}', you would need to prepend with a \`cd\` i.e. pseudocode for this would be \`cd (path to project) && (command, in this case npm install)\`.
+- When using the search_files tool, craft your regex patterns carefully to balance specificity and flexibility. Based on the user's task you may use it to find code patterns, TODO comments, function definitions, or any text-based information across the project. The results include context, so analyze the surrounding code to better understand the matches. Leverage the search_files tool in combination with other tools for more comprehensive analysis. For example, use it to find specific code patterns, then use read_file to examine the full context of interesting matches before using replace_in_file to make informed changes.
+- When creating a new project (such as an app, website, or any software project), organize all new files within a dedicated project directory unless the user specifies otherwise. Use appropriate file paths when creating files, as the write_to_file tool will automatically create any necessary directories. Structure the project logically, adhering to best practices for the specific type of project being created. Unless otherwise specified, new projects should be easily run without additional setup, for example most projects can be built in HTML, CSS, and JavaScript - which you can open in a browser.
+- Be sure to consider the type of project (e.g. Python, JavaScript, web application) when determining the appropriate structure and files to include. Also consider what files may be most relevant to accomplishing the task, for example looking at a project's manifest file would help you understand the project's dependencies, which you could incorporate into any code you write.
+- When making changes to code, always consider the context in which the code is being used. Ensure that your changes are compatible with the existing codebase and that they follow the project's coding standards and best practices.
+- When you want to modify a file, use the replace_in_file or write_to_file tool directly with the desired changes. You do not need to display the changes before using the tool.
+- Use Markdown **only where semantically correct** (e.g., \`inline code\`, \`\`\`code fences\`\`\`, lists, tables). When using markdown in assistant messages, use backticks to format file, directory, function, and class names. Use \( and \) for inline math, \[ and \] for block math.
+- Do not ask for more information than necessary. Use the tools provided to accomplish the user's request efficiently and effectively. When you've completed your task, you must use the attempt_completion tool to present the result to the user. The user may provide feedback, which you can use to make improvements and try again.
+- You are only allowed to ask the user questions using the ask_followup_question tool. Use this tool only when you need additional details to complete a task, and be sure to use a clear and concise question that will help you move forward with the task. However if you can use the available tools to avoid having to ask the user questions, you should do so. For example, if the user mentions a file that may be in an outside directory like the Desktop, you should use the list_files tool to list the files in the Desktop and check if the file they are talking about is there, rather than asking the user to provide the file path themselves.
+- When the user is being vague, you should be proactive about asking clarifying questions using the ask_followup_question tool to ensure you understand their request. However, if you can infer the user's intent based on the context and available tools, you should proceed without asking unnecessary questions
+- When executing commands, if you don't see the expected output, assume the terminal executed the command successfully and proceed with the task. The user's terminal may be unable to stream the output back properly. If you absolutely need to see the actual terminal output, use the ask_followup_question tool to request the user to copy and paste it back to you.
+- The user may provide a file's contents directly in their message, in which case you shouldn't use the read_file tool to get the file contents again since you already have it.
+- Your goal is to try to accomplish the user's task, NOT engage in a back and forth conversation.${
+		supportsBrowserUse
+			? `\n- The user may ask generic non-development tasks, such as "what\'s the latest news" or "look up the weather in San Diego", in which case you might use the browser_action tool to complete the task if it makes sense to do so, rather than trying to create a website or using curl to answer the question. However, if an available MCP server tool or resource can be used instead, you should prefer to use it over browser_action.`
+			: ""
+	}
+- NEVER end attempt_completion result with a question or request to engage in further conversation! Formulate the end of your result in a way that is final and does not require further input from the user.
+- You are STRICTLY FORBIDDEN from starting your messages with "Great", "Certainly", "Okay", "Sure". You should NOT be conversational in your responses, but rather direct and to the point. For example you should NOT say "Great, I've updated the CSS" but instead something like "I've updated the CSS". It is important you be clear and technical in your messages.
+- When presented with images, utilize your vision capabilities to thoroughly examine them and extract meaningful information. Incorporate these insights into your thought process as you accomplish the user's task.
+- At the end of each user message, you will automatically receive environment_details. This information is not written by the user themselves, but is auto-generated to provide potentially relevant context about the project structure and environment. While this information can be valuable for understanding the project context, do not treat it as a direct part of the user's request or response. Use it to inform your actions and decisions, but don't assume the user is explicitly asking about or referring to this information unless they clearly do so in their message. When using environment_details, explain your actions clearly to ensure the user understands, as they may not be aware of these details.
+- Before executing commands, check the "Actively Running Terminals" section in environment_details. If present, consider how these active processes might impact your task. For example, if a local development server is already running, you wouldn't need to start it again. If no active terminals are listed, proceed with command execution as normal.
+- When using the replace_in_file tool, you must include complete lines in your SEARCH blocks, not partial lines. The system requires exact line matches and cannot match partial lines. For example, if you want to match a line containing "const x = 5;", your SEARCH block must include the entire line, not just "x = 5" or other fragments.
+- When using the replace_in_file tool, if you use multiple SEARCH/REPLACE blocks, list them in the order they appear in the file. For example if you need to make changes to both line 10 and line 50, first include the SEARCH/REPLACE block for line 10, followed by the SEARCH/REPLACE block for line 50.
+- When using the replace_in_file tool, Do NOT add extra characters to the markers (e.g., ------- SEARCH> is INVALID). Do NOT forget to use the closing +++++++ REPLACE marker. Do NOT modify the marker format in any way. Malformed XML will cause complete tool failure and break the entire editing process.
+- It is critical you wait for the user's response after each tool use, in order to confirm the success of the tool use. For example, if asked to make a todo app, you would create a file, wait for the user's response it was created successfully, then create another file if needed, wait for the user's response it was created successfully, etc.${
+		supportsBrowserUse
+			? " Then if you want to test your work, you might use browser_action to launch the site, wait for the user's response confirming the site was launched along with a screenshot, then perhaps e.g., click a button to test functionality if needed, wait for the user's response confirming the button was clicked along with a screenshot of the new state, before finally closing the browser."
+			: ""
+	}
+- MCP operations should be used one at a time, similar to other tool usage. Wait for confirmation of success before proceeding with additional operations.
+
+====
+
+SYSTEM INFORMATION
+
+Operating System: ${osName()}
+Default Shell: ${getShell()}
+Home Directory: ${os.homedir().toPosix()}
+Current Working Directory: ${cwd.toPosix()}
+
+====
+
+OBJECTIVE
+
+You accomplish a given task iteratively, breaking it down into clear steps and working through them methodically.
+
+1. Analyze the user's task and set clear, achievable goals to accomplish it. Prioritize these goals in a logical order.
+2. Work through these goals sequentially, utilizing available tools one at a time as necessary. Each goal should correspond to a distinct step in your problem-solving process. You will be informed on the work completed and what's remaining as you go.
+3. Remember, you have extensive capabilities with access to a wide range of tools that can be used in powerful and clever ways as necessary to accomplish each goal. Before calling a tool, do some analysis within <thinking></thinking> tags. First, analyze the file structure provided in environment_details to gain context and insights for proceeding effectively. Then, think about which of the provided tools is the most relevant tool to accomplish the user's task. Next, go through each of the required parameters of the relevant tool and determine if the user has directly provided or given enough information to infer a value. When deciding if the parameter can be inferred, carefully consider all the context to see if it supports a specific value. If all of the required parameters are present or can be reasonably inferred, close the thinking tag and proceed with the tool use. BUT, if one of the values for a required parameter is missing, DO NOT invoke the tool (not even with fillers for the missing params) and instead, ask the user to provide the missing parameters using the ask_followup_question tool. DO NOT ask for more information on optional parameters if it is not provided.
+4. Once you've completed the user's task, you must use the attempt_completion tool to present the result of the task to the user. You may also provide a CLI command to showcase the result of your task; this can be particularly useful for web development tasks, where you can run e.g. \`open index.html\` to show the website you've built.
+5. The user may provide feedback, which you can use to make improvements and try again. But DO NOT continue in pointless back and forth conversations, i.e. don't end your responses with questions or offers for further assistance.`
+}

src/core/slash-commands/index.ts

@@ -18,16 +18,17 @@ export async function parseSlashCommands(
 	localWorkflowToggles: ClineRulesToggles,
 	globalWorkflowToggles: ClineRulesToggles,
 	ulid: string,
+	focusChainSettings?: { enabled: boolean },
 ): Promise<{ processedText: string; needsClinerulesFileCheck: boolean }> {
 	const SUPPORTED_DEFAULT_COMMANDS = ["newtask", "smol", "compact", "newrule", "reportbug", "deep-planning"]
 
 	const commandReplacements: Record<string, string> = {
 		newtask: newTaskToolResponse(),
-		smol: condenseToolResponse(),
-		compact: condenseToolResponse(),
+		smol: condenseToolResponse(focusChainSettings),
+		compact: condenseToolResponse(focusChainSettings),
 		newrule: newRuleToolResponse(),
 		reportbug: reportBugToolResponse(),
-		"deep-planning": deepPlanningToolResponse(),
+		"deep-planning": deepPlanningToolResponse(focusChainSettings),
 	}
 
 	// this currently allows matching prepended whitespace prior to /slash-command

src/core/storage/StateManager.ts

@@ -2,7 +2,8 @@ import { ApiConfiguration, fireworksDefaultModelId } from "@shared/api"
 import { DEFAULT_FOCUS_CHAIN_SETTINGS } from "@shared/FocusChainSettings"
 import type { ExtensionContext } from "vscode"
 import { DEFAULT_AUTO_APPROVAL_SETTINGS } from "@/shared/AutoApprovalSettings"
-import { CACHE_SERVICE_NOT_INITIALIZED } from "./error-messages"
+import { DEFAULT_DICTATION_SETTINGS } from "@/shared/DictationSettings"
+import { STATE_MANAGER_NOT_INITIALIZED } from "./error-messages"
 import { GlobalState, GlobalStateKey, LocalState, LocalStateKey, SecretKey, Secrets } from "./state-keys"
 import { readStateFromDisk } from "./utils/state-helpers"
 
@@ -14,10 +15,10 @@ export interface PersistenceErrorEvent {
 }
 
 /**
- * In-memory cache service for fast state access
+ * In-memory state manager for fast state access
  * Provides immediate reads/writes with async disk persistence
  */
-export class CacheService {
+export class StateManager {
 	private globalStateCache: GlobalState = {} as GlobalState
 	private secretsCache: Secrets = {} as Secrets
 	private workspaceStateCache: LocalState = {} as LocalState
@@ -54,7 +55,7 @@ export class CacheService {
 
 			this.isInitialized = true
 		} catch (error) {
-			console.error("Failed to initialize CacheService:", error)
+			console.error("Failed to initialize StateManager:", error)
 			throw error
 		}
 	}
@@ -64,7 +65,7 @@ export class CacheService {
 	 */
 	setGlobalState<K extends keyof GlobalState>(key: K, value: GlobalState[K]): void {
 		if (!this.isInitialized) {
-			throw new Error(CACHE_SERVICE_NOT_INITIALIZED)
+			throw new Error(STATE_MANAGER_NOT_INITIALIZED)
 		}
 
 		// Update cache immediately for instant access
@@ -80,7 +81,7 @@ export class CacheService {
 	 */
 	setGlobalStateBatch(updates: Partial<GlobalState>): void {
 		if (!this.isInitialized) {
-			throw new Error(CACHE_SERVICE_NOT_INITIALIZED)
+			throw new Error(STATE_MANAGER_NOT_INITIALIZED)
 		}
 
 		// Update cache in one go
@@ -101,7 +102,7 @@ export class CacheService {
 	 */
 	setSecret<K extends keyof Secrets>(key: K, value: Secrets[K]): void {
 		if (!this.isInitialized) {
-			throw new Error(CACHE_SERVICE_NOT_INITIALIZED)
+			throw new Error(STATE_MANAGER_NOT_INITIALIZED)
 		}
 
 		// Update cache immediately for instant access
@@ -117,7 +118,7 @@ export class CacheService {
 	 */
 	setSecretsBatch(updates: Partial<Secrets>): void {
 		if (!this.isInitialized) {
-			throw new Error(CACHE_SERVICE_NOT_INITIALIZED)
+			throw new Error(STATE_MANAGER_NOT_INITIALIZED)
 		}
 
 		// Update cache immediately for all keys
@@ -135,7 +136,7 @@ export class CacheService {
 	 */
 	setWorkspaceState<K extends keyof LocalState>(key: K, value: LocalState[K]): void {
 		if (!this.isInitialized) {
-			throw new Error(CACHE_SERVICE_NOT_INITIALIZED)
+			throw new Error(STATE_MANAGER_NOT_INITIALIZED)
 		}
 
 		// Update cache immediately for instant access
@@ -151,7 +152,7 @@ export class CacheService {
 	 */
 	setWorkspaceStateBatch(updates: Partial<LocalState>): void {
 		if (!this.isInitialized) {
-			throw new Error(CACHE_SERVICE_NOT_INITIALIZED)
+			throw new Error(STATE_MANAGER_NOT_INITIALIZED)
 		}
 
 		// Update cache immediately for all keys
@@ -170,7 +171,7 @@ export class CacheService {
 	 */
 	getApiConfiguration(): ApiConfiguration {
 		if (!this.isInitialized) {
-			throw new Error(CACHE_SERVICE_NOT_INITIALIZED)
+			throw new Error(STATE_MANAGER_NOT_INITIALIZED)
 		}
 
 		// Construct API configuration from cached component keys
@@ -182,7 +183,7 @@ export class CacheService {
 	 */
 	setApiConfiguration(apiConfiguration: ApiConfiguration): void {
 		if (!this.isInitialized) {
-			throw new Error(CACHE_SERVICE_NOT_INITIALIZED)
+			throw new Error(STATE_MANAGER_NOT_INITIALIZED)
 		}
 
 		const {
@@ -251,6 +252,8 @@ export class CacheService {
 			basetenApiKey,
 			huggingFaceApiKey,
 			huaweiCloudMaasApiKey,
+			difyApiKey,
+			difyBaseUrl,
 			vercelAiGatewayApiKey,
 			zaiApiKey,
 			requestTimeoutMs,
@@ -419,6 +422,7 @@ export class CacheService {
 			sapAiCoreTokenUrl,
 			sapAiResourceGroup,
 			claudeCodePath,
+			difyBaseUrl,
 			qwenCodeOauthPath,
 		})
 
@@ -455,6 +459,7 @@ export class CacheService {
 			basetenApiKey,
 			huggingFaceApiKey,
 			huaweiCloudMaasApiKey,
+			difyApiKey,
 			vercelAiGatewayApiKey,
 			zaiApiKey,
 		})
@@ -465,7 +470,7 @@ export class CacheService {
 	 */
 	getGlobalStateKey<K extends keyof GlobalState>(key: K): GlobalState[K] {
 		if (!this.isInitialized) {
-			throw new Error(CACHE_SERVICE_NOT_INITIALIZED)
+			throw new Error(STATE_MANAGER_NOT_INITIALIZED)
 		}
 		return this.globalStateCache[key]
 	}
@@ -475,7 +480,7 @@ export class CacheService {
 	 */
 	getSecretKey<K extends keyof Secrets>(key: K): Secrets[K] {
 		if (!this.isInitialized) {
-			throw new Error(CACHE_SERVICE_NOT_INITIALIZED)
+			throw new Error(STATE_MANAGER_NOT_INITIALIZED)
 		}
 		return this.secretsCache[key]
 	}
@@ -485,13 +490,13 @@ export class CacheService {
 	 */
 	getWorkspaceStateKey<K extends keyof LocalState>(key: K): LocalState[K] {
 		if (!this.isInitialized) {
-			throw new Error(CACHE_SERVICE_NOT_INITIALIZED)
+			throw new Error(STATE_MANAGER_NOT_INITIALIZED)
 		}
 		return this.workspaceStateCache[key]
 	}
 
 	/**
-	 * Reinitialize the cache service by clearing all state and reloading from disk
+	 * Reinitialize the state manager by clearing all state and reloading from disk
 	 * Used for error recovery when write operations fail
 	 */
 	async reInitialize(): Promise<void> {
@@ -503,7 +508,7 @@ export class CacheService {
 	}
 
 	/**
-	 * Dispose of the cache service
+	 * Dispose of the state manager
 	 */
 	private dispose(): void {
 		if (this.persistenceTimeout) {
@@ -682,6 +687,8 @@ export class CacheService {
 			qwenCodeOauthPath,
 			huggingFaceApiKey,
 			huaweiCloudMaasApiKey,
+			difyApiKey,
+			difyBaseUrl,
 			vercelAiGatewayApiKey,
 			zaiApiKey,
 			requestTimeoutMs,
@@ -760,6 +767,7 @@ export class CacheService {
 			autoApprovalSettings: state.autoApprovalSettings || DEFAULT_AUTO_APPROVAL_SETTINGS,
 			globalClineRulesToggles: state.globalClineRulesToggles,
 			browserSettings: state.browserSettings,
+			dictationSettings: state.dictationSettings || DEFAULT_DICTATION_SETTINGS,
 			focusChainSettings: state.focusChainSettings || DEFAULT_FOCUS_CHAIN_SETTINGS,
 			focusChainFeatureFlagEnabled: state.focusChainFeatureFlagEnabled,
 			preferredLanguage: state.preferredLanguage,
@@ -882,6 +890,7 @@ export class CacheService {
 			sapAiCoreTokenUrl,
 			sapAiResourceGroup,
 			claudeCodePath,
+			difyBaseUrl,
 			qwenCodeOauthPath,
 		} satisfies GlobalState
 
@@ -922,6 +931,7 @@ export class CacheService {
 			authNonce,
 			huggingFaceApiKey,
 			huaweiCloudMaasApiKey,
+			difyApiKey,
 			vercelAiGatewayApiKey,
 			zaiApiKey,
 		} satisfies Secrets
@@ -977,6 +987,7 @@ export class CacheService {
 			sapAiCoreClientSecret: this.secretsCache["sapAiCoreClientSecret"],
 			huggingFaceApiKey: this.secretsCache["huggingFaceApiKey"],
 			huaweiCloudMaasApiKey: this.secretsCache["huaweiCloudMaasApiKey"],
+			difyApiKey: this.secretsCache["difyApiKey"],
 			vercelAiGatewayApiKey: this.secretsCache["vercelAiGatewayApiKey"],
 			zaiApiKey: this.secretsCache["zaiApiKey"],
 
@@ -1016,6 +1027,7 @@ export class CacheService {
 			sapAiResourceGroup: this.globalStateCache["sapAiResourceGroup"],
 			claudeCodePath: this.globalStateCache["claudeCodePath"],
 			qwenCodeOauthPath: this.globalStateCache["qwenCodeOauthPath"],
+			difyBaseUrl: this.globalStateCache["difyBaseUrl"],
 
 			// Plan mode configurations
 			planModeApiProvider: this.globalStateCache["planModeApiProvider"],

src/core/storage/error-messages.ts

@@ -1 +1 @@
-export const CACHE_SERVICE_NOT_INITIALIZED = "CacheService must be initialized before attempting to access state."
+export const STATE_MANAGER_NOT_INITIALIZED = "StateManager must be initialized before attempting to access state."

src/core/storage/state-keys.ts

@@ -4,6 +4,7 @@ import { LanguageModelChatSelector } from "vscode"
 import { AutoApprovalSettings } from "@/shared/AutoApprovalSettings"
 import { BrowserSettings } from "@/shared/BrowserSettings"
 import { ClineRulesToggles } from "@/shared/cline-rules"
+import { DictationSettings } from "@/shared/DictationSettings"
 import { HistoryItem } from "@/shared/HistoryItem"
 import { McpDisplayMode } from "@/shared/McpDisplayMode"
 import { McpMarketplaceCatalog } from "@/shared/mcp"
@@ -11,172 +12,11 @@ import { Mode, OpenaiReasoningEffort } from "@/shared/storage/types"
 import { TelemetrySetting } from "@/shared/TelemetrySetting"
 import { UserInfo } from "@/shared/UserInfo"
 
-export type SecretKey =
-	| "apiKey"
-	| "clineAccountId"
-	| "openRouterApiKey"
-	| "awsAccessKey"
-	| "awsSecretKey"
-	| "awsSessionToken"
-	| "awsBedrockApiKey"
-	| "openAiApiKey"
-	| "ollamaApiKey"
-	| "geminiApiKey"
-	| "openAiNativeApiKey"
-	| "deepSeekApiKey"
-	| "requestyApiKey"
-	| "togetherApiKey"
-	| "fireworksApiKey"
-	| "qwenApiKey"
-	| "doubaoApiKey"
-	| "mistralApiKey"
-	| "liteLlmApiKey"
-	| "authNonce"
-	| "asksageApiKey"
-	| "xaiApiKey"
-	| "moonshotApiKey"
-	| "huggingFaceApiKey"
-	| "nebiusApiKey"
-	| "sambanovaApiKey"
-	| "cerebrasApiKey"
-	| "sapAiCoreClientId"
-	| "sapAiCoreClientSecret"
-	| "groqApiKey"
-	| "huaweiCloudMaasApiKey"
-	| "basetenApiKey"
-	| "vercelAiGatewayApiKey"
-	| "zaiApiKey"
+export type SecretKey = keyof Secrets
 
-export type GlobalStateKey =
-	| "awsRegion"
-	| "awsUseCrossRegionInference"
-	| "awsBedrockUsePromptCache"
-	| "awsBedrockEndpoint"
-	| "awsProfile"
-	| "awsBedrockApiKey"
-	| "awsAuthentication"
-	| "awsUseProfile"
-	| "vertexProjectId"
-	| "vertexRegion"
-	| "lastShownAnnouncementId"
-	| "taskHistory"
-	| "requestyBaseUrl"
-	| "openAiBaseUrl"
-	| "openAiHeaders"
-	| "ollamaBaseUrl"
-	| "ollamaApiOptionsCtxNum"
-	| "lmStudioBaseUrl"
-	| "lmStudioMaxTokens"
-	| "anthropicBaseUrl"
-	| "geminiBaseUrl"
-	| "azureApiVersion"
-	| "openRouterProviderSorting"
-	| "autoApprovalSettings"
-	| "globalClineRulesToggles"
-	| "globalWorkflowToggles"
-	| "browserSettings"
-	| "userInfo"
-	| "liteLlmBaseUrl"
-	| "liteLlmUsePromptCache"
-	| "fireworksModelMaxCompletionTokens"
-	| "fireworksModelMaxTokens"
-	| "qwenApiLine"
-	| "moonshotApiLine"
-	| "zaiApiLine"
-	| "mcpMarketplaceCatalog"
-	| "telemetrySetting"
-	| "asksageApiUrl"
-	| "planActSeparateModelsSetting"
-	| "enableCheckpointsSetting"
-	| "mcpMarketplaceEnabled"
-	| "favoritedModelIds"
-	| "requestTimeoutMs"
-	| "shellIntegrationTimeout"
-	| "mcpResponsesCollapsed"
-	| "terminalReuseEnabled"
-	| "defaultTerminalProfile"
-	| "isNewUser"
-	| "welcomeViewCompleted"
-	| "terminalOutputLineLimit"
-	| "mcpDisplayMode"
-	| "sapAiCoreTokenUrl"
-	| "sapAiCoreBaseUrl"
-	| "sapAiResourceGroup"
-	| "claudeCodePath"
-	| "qwenCodeOauthPath"
-	| "strictPlanModeEnabled"
-	| "useAutoCondense"
-	| "focusChainSettings"
-	| "focusChainFeatureFlagEnabled"
-	// Settings around plan/act and ephemeral model configuration
-	| "preferredLanguage"
-	| "openaiReasoningEffort"
-	| "mode"
-	| "customPrompt"
-	// Plan mode configurations
-	| "planModeApiProvider"
-	| "planModeApiModelId"
-	| "planModeThinkingBudgetTokens"
-	| "planModeReasoningEffort"
-	| "planModeVsCodeLmModelSelector"
-	| "planModeAwsBedrockCustomSelected"
-	| "planModeAwsBedrockCustomModelBaseId"
-	| "planModeOpenRouterModelId"
-	| "planModeOpenRouterModelInfo"
-	| "planModeOpenAiModelId"
-	| "planModeOpenAiModelInfo"
-	| "planModeOllamaModelId"
-	| "planModeLmStudioModelId"
-	| "planModeLiteLlmModelId"
-	| "planModeLiteLlmModelInfo"
-	| "planModeRequestyModelId"
-	| "planModeRequestyModelInfo"
-	| "planModeTogetherModelId"
-	| "planModeFireworksModelId"
-	| "planModeSapAiCoreModelId"
-	| "planModeGroqModelId"
-	| "planModeGroqModelInfo"
-	| "planModeBasetenModelId"
-	| "planModeBasetenModelInfo"
-	| "planModeHuggingFaceModelId"
-	| "planModeHuggingFaceModelInfo"
-	| "planModeHuaweiCloudMaasModelId"
-	| "planModeHuaweiCloudMaasModelInfo"
-	| "planModeVercelAiGatewayModelId"
-	| "planModeVercelAiGatewayModelInfo"
-	// Act mode configurations
-	| "actModeApiProvider"
-	| "actModeApiModelId"
-	| "actModeThinkingBudgetTokens"
-	| "actModeReasoningEffort"
-	| "actModeVsCodeLmModelSelector"
-	| "actModeAwsBedrockCustomSelected"
-	| "actModeAwsBedrockCustomModelBaseId"
-	| "actModeOpenRouterModelId"
-	| "actModeOpenRouterModelInfo"
-	| "actModeOpenAiModelId"
-	| "actModeOpenAiModelInfo"
-	| "actModeOllamaModelId"
-	| "actModeLmStudioModelId"
-	| "actModeLiteLlmModelId"
-	| "actModeLiteLlmModelInfo"
-	| "actModeRequestyModelId"
-	| "actModeRequestyModelInfo"
-	| "actModeTogetherModelId"
-	| "actModeFireworksModelId"
-	| "actModeSapAiCoreModelId"
-	| "actModeGroqModelId"
-	| "actModeGroqModelInfo"
-	| "actModeBasetenModelId"
-	| "actModeBasetenModelInfo"
-	| "actModeHuggingFaceModelId"
-	| "actModeHuggingFaceModelInfo"
-	| "actModeHuaweiCloudMaasModelId"
-	| "actModeHuaweiCloudMaasModelInfo"
-	| "actModeVercelAiGatewayModelId"
-	| "actModeVercelAiGatewayModelInfo"
+export type GlobalStateKey = keyof GlobalState
 
-export type LocalStateKey = "localClineRulesToggles" | "localCursorRulesToggles" | "localWindsurfRulesToggles" | "workflowToggles"
+export type LocalStateKey = keyof LocalState
 
 export interface GlobalState {
 	awsRegion: string | undefined
@@ -240,9 +80,11 @@ export interface GlobalState {
 	preferredLanguage: string
 	openaiReasoningEffort: OpenaiReasoningEffort
 	mode: Mode
+	dictationSettings: DictationSettings
 	focusChainSettings: FocusChainSettings
 	focusChainFeatureFlagEnabled: boolean
 	customPrompt: "compact" | undefined
+	difyBaseUrl: string | undefined
 
 	// Plan mode configurations
 	planModeApiProvider: ApiProvider
@@ -343,6 +185,7 @@ export interface Secrets {
 	huaweiCloudMaasApiKey: string | undefined
 	basetenApiKey: string | undefined
 	vercelAiGatewayApiKey: string | undefined
+	difyApiKey: string | undefined
 }
 
 export interface LocalState {

src/core/storage/state-migrations.ts

@@ -545,6 +545,7 @@ export async function migrateWelcomeViewCompleted(context: vscode.ExtensionConte
 						config.xaiApiKey,
 						config.sambanovaApiKey,
 						config.sapAiCoreClientId,
+						config.difyApiKey,
 					].some((key) => key !== undefined)
 				: false
 

src/core/storage/utils/state-helpers.ts

@@ -4,6 +4,8 @@ import { Controller } from "@/core/controller"
 import { AutoApprovalSettings, DEFAULT_AUTO_APPROVAL_SETTINGS } from "@/shared/AutoApprovalSettings"
 import { BrowserSettings, DEFAULT_BROWSER_SETTINGS } from "@/shared/BrowserSettings"
 import { ClineRulesToggles } from "@/shared/cline-rules"
+import { DEFAULT_DICTATION_SETTINGS, DictationSettings } from "@/shared/DictationSettings"
+import { DEFAULT_FOCUS_CHAIN_SETTINGS, FocusChainSettings } from "@/shared/FocusChainSettings"
 import { HistoryItem } from "@/shared/HistoryItem"
 import { DEFAULT_MCP_DISPLAY_MODE, McpDisplayMode } from "@/shared/McpDisplayMode"
 import { Mode, OpenaiReasoningEffort } from "@/shared/storage/types"
@@ -68,8 +70,12 @@ export async function readStateFromDisk(context: ExtensionContext) {
 	const sapAiCoreTokenUrl = context.globalState.get("sapAiCoreTokenUrl") as string | undefined
 	const sapAiResourceGroup = context.globalState.get("sapAiResourceGroup") as string | undefined
 	const claudeCodePath = context.globalState.get("claudeCodePath") as string | undefined
+	const difyBaseUrl = context.globalState.get("difyBaseUrl") as string | undefined
 	const openaiReasoningEffort = context.globalState.get("openaiReasoningEffort") as OpenaiReasoningEffort | undefined
 	const preferredLanguage = context.globalState.get("preferredLanguage") as string | undefined
+	const dictationSettings = context.globalState.get("dictationSettings") as DictationSettings | undefined
+	const focusChainSettings = context.globalState.get("focusChainSettings") as FocusChainSettings | undefined
+	const focusChainFeatureFlagEnabled = context.globalState.get("focusChainFeatureFlagEnabled") as boolean | undefined
 
 	// Get all secret values
 	const [
@@ -106,6 +112,7 @@ export async function readStateFromDisk(context: ExtensionContext) {
 		zaiApiKey,
 		ollamaApiKey,
 		vercelAiGatewayApiKey,
+		difyApiKey,
 	] = await Promise.all([
 		context.secrets.get("apiKey") as Promise<string | undefined>,
 		context.secrets.get("openRouterApiKey") as Promise<string | undefined>,
@@ -140,6 +147,7 @@ export async function readStateFromDisk(context: ExtensionContext) {
 		context.secrets.get("zaiApiKey") as Promise<string | undefined>,
 		context.secrets.get("ollamaApiKey") as Promise<string | undefined>,
 		context.secrets.get("vercelAiGatewayApiKey") as Promise<string | undefined>,
+		context.secrets.get("difyApiKey") as Promise<string | undefined>,
 	])
 
 	const localClineRulesToggles = context.workspaceState.get("localClineRulesToggles") as ClineRulesToggles | undefined
@@ -321,6 +329,8 @@ export async function readStateFromDisk(context: ExtensionContext) {
 			zaiApiKey,
 			ollamaApiKey,
 			vercelAiGatewayApiKey,
+			difyApiKey,
+			difyBaseUrl,
 			// Plan mode configurations
 			planModeApiProvider: planModeApiProvider || apiProvider,
 			planModeApiModelId,
@@ -384,6 +394,9 @@ export async function readStateFromDisk(context: ExtensionContext) {
 			actModeVercelAiGatewayModelId,
 			actModeVercelAiGatewayModelInfo,
 		},
+		focusChainSettings: focusChainSettings || DEFAULT_FOCUS_CHAIN_SETTINGS,
+		focusChainFeatureFlagEnabled: focusChainFeatureFlagEnabled ?? false,
+		dictationSettings: dictationSettings || DEFAULT_DICTATION_SETTINGS,
 		strictPlanModeEnabled: strictPlanModeEnabled ?? true,
 		useAutoCondense: useAutoCondense ?? true,
 		isNewUser: isNewUser ?? true,
@@ -419,7 +432,7 @@ export async function resetWorkspaceState(controller: Controller) {
 	const context = controller.context
 	await Promise.all(context.workspaceState.keys().map((key) => controller.context.workspaceState.update(key, undefined)))
 
-	await controller.cacheService.reInitialize()
+	await controller.stateManager.reInitialize()
 }
 
 export async function resetGlobalState(controller: Controller) {
@@ -459,7 +472,8 @@ export async function resetGlobalState(controller: Controller) {
 		"huaweiCloudMaasApiKey",
 		"vercelAiGatewayApiKey",
 		"zaiApiKey",
+		"difyApiKey",
 	]
 	await Promise.all(secretKeys.map((key) => context.secrets.delete(key)))
-	await controller.cacheService.reInitialize()
+	await controller.stateManager.reInitialize()
 }

src/core/task/ToolExecutor.ts

@@ -50,8 +50,8 @@ import { ContextManager } from "../context/context-management/ContextManager"
 import { continuationPrompt } from "../prompts/contextManagement"
 import { loadMcpDocumentation } from "../prompts/loadMcpDocumentation"
 import { formatResponse } from "../prompts/responses"
-import { CacheService } from "../storage/CacheService"
 import { ensureTaskDirectoryExists } from "../storage/disk"
+import { StateManager } from "../storage/StateManager"
 import { ToolResponse } from "."
 import { MessageStateHandler } from "./message-state"
 import { TaskState } from "./TaskState"
@@ -86,7 +86,7 @@ export class ToolExecutor {
 		private fileContextTracker: FileContextTracker,
 		private clineIgnoreController: ClineIgnoreController,
 		private contextManager: ContextManager,
-		private cacheService: CacheService,
+		private stateManager: StateManager,
 
 		// Configuration & Settings
 		private autoApprovalSettings: AutoApprovalSettings,
@@ -1906,7 +1906,7 @@ export class ToolExecutor {
 						const host = await HostProvider.env.getHostVersion({})
 						const systemInfo = `${host.platform}: ${host.version}, Node.js: ${process.version}, Architecture: ${os.arch()}`
 						const currentMode = this.mode
-						const apiConfig = this.cacheService.getApiConfiguration()
+						const apiConfig = this.stateManager.getApiConfiguration()
 						const apiProvider = currentMode === "plan" ? apiConfig.planModeApiProvider : apiConfig.actModeApiProvider
 						const providerAndModel = `${apiProvider} / ${this.api.getModel().id}`
 

src/core/task/focus-chain/index.ts

@@ -7,8 +7,8 @@ import { ClineSay } from "../../../shared/ExtensionMessage"
 import { FileChangeEvent_ChangeType, SubscribeToFileRequest } from "../../../shared/proto/host/watch"
 import { Mode } from "../../../shared/storage/types"
 import { writeFile } from "../../../utils/fs"
-import { CacheService } from "../../storage/CacheService"
 import { ensureTaskDirectoryExists } from "../../storage/disk"
+import { StateManager } from "../../storage/StateManager"
 import { TaskState } from "../TaskState"
 import {
 	createFocusChainMarkdownContent,
@@ -23,7 +23,7 @@ export interface FocusChainDependencies {
 	taskState: TaskState
 	mode: Mode
 	context: vscode.ExtensionContext
-	cacheService: CacheService
+	stateManager: StateManager
 	postStateToWebview: () => Promise<void>
 	say: (type: ClineSay, text?: string, images?: string[], files?: string[], partial?: boolean) => Promise<undefined>
 	focusChainSettings: FocusChainSettings
@@ -34,7 +34,7 @@ export class FocusChainManager {
 	private taskState: TaskState
 	private mode: Mode
 	private context: vscode.ExtensionContext
-	private cacheService: CacheService
+	private stateManager: StateManager
 	private postStateToWebview: () => Promise<void>
 	private say: (type: ClineSay, text?: string, images?: string[], files?: string[], partial?: boolean) => Promise<undefined>
 	private focusChainFileWatcherCancel?: () => void
@@ -47,7 +47,7 @@ export class FocusChainManager {
 		this.taskState = dependencies.taskState
 		this.mode = dependencies.mode
 		this.context = dependencies.context
-		this.cacheService = dependencies.cacheService
+		this.stateManager = dependencies.stateManager
 		this.postStateToWebview = dependencies.postStateToWebview
 		this.say = dependencies.say
 		this.focusChainSettings = dependencies.focusChainSettings
@@ -66,7 +66,7 @@ export class FocusChainManager {
 	private async initializeRemoteFeatureFlags(): Promise<void> {
 		try {
 			const enabled = await featureFlagsService.getFocusChainEnabled()
-			this.cacheService.setGlobalState("focusChainFeatureFlagEnabled", enabled)
+			this.stateManager.setGlobalState("focusChainFeatureFlagEnabled", enabled)
 			await this.postStateToWebview()
 		} catch (error) {
 			console.error("Error initializing focus chain remote feature flags:", error)

src/core/task/index.ts

@@ -77,7 +77,7 @@ import { refreshWorkflowToggles } from "../context/instructions/user-instruction
 import { Controller } from "../controller"
 import { addUserInstructions } from "../prompts/system-prompt/user-instructions/addUserInstructions"
 import { isNextGenModelFamily } from "../prompts/system-prompt/utils"
-import { CacheService } from "../storage/CacheService"
+import { StateManager } from "../storage/StateManager"
 import { FocusChainManager } from "./focus-chain"
 import { MessageStateHandler } from "./message-state"
 import { showChangedFilesDiff } from "./multifile-diff"
@@ -94,6 +94,7 @@ export class Task {
 	readonly ulid: string
 	private taskIsFavorited?: boolean
 	private cwd: string
+	private taskInitializationStartTime: number
 
 	taskState: TaskState
 
@@ -132,7 +133,7 @@ export class Task {
 	private cancelTask: () => Promise<void>
 
 	// Cache service
-	private cacheService: CacheService
+	private stateManager: StateManager
 
 	// User chat state
 	autoApprovalSettings: AutoApprovalSettings
@@ -166,12 +167,13 @@ export class Task {
 		defaultTerminalProfile: string,
 		enableCheckpointsSetting: boolean,
 		cwd: string,
-		cacheService: CacheService,
+		stateManager: StateManager,
 		task?: string,
 		images?: string[],
 		files?: string[],
 		historyItem?: HistoryItem,
 	) {
+		this.taskInitializationStartTime = performance.now()
 		this.taskState = new TaskState()
 		this.controller = controller
 		this.mcpHub = mcpHub
@@ -210,7 +212,7 @@ export class Task {
 		this.mode = mode
 		this.enableCheckpoints = enableCheckpointsSetting
 		this.cwd = cwd
-		this.cacheService = cacheService
+		this.stateManager = stateManager
 		this.useAutoCondense = useAutoCondense
 
 		// Set up MCP notification callback for real-time notifications
@@ -255,7 +257,7 @@ export class Task {
 				taskState: this.taskState,
 				mode: this.mode,
 				context: this.getContext(),
-				cacheService: this.cacheService,
+				stateManager: this.stateManager,
 				postStateToWebview: this.postStateToWebview,
 				say: this.say.bind(this),
 				focusChainSettings: this.focusChainSettings,
@@ -351,7 +353,7 @@ export class Task {
 			this.fileContextTracker,
 			this.clineIgnoreController,
 			this.contextManager,
-			this.cacheService,
+			this.stateManager,
 			this.autoApprovalSettings,
 			this.browserSettings,
 			this.focusChainSettings,
@@ -1626,9 +1628,9 @@ export class Task {
 
 	private getCurrentProviderInfo(): ApiProviderInfo {
 		const modelId = this.api.getModel()?.id
-		const apiConfig = this.cacheService.getApiConfiguration()
+		const apiConfig = this.stateManager.getApiConfiguration()
 		const providerId = (this.mode === "plan" ? apiConfig.planModeApiProvider : apiConfig.actModeApiProvider) as string
-		const customPrompt = this.cacheService.getGlobalStateKey("customPrompt")
+		const customPrompt = this.stateManager.getGlobalStateKey("customPrompt")
 		return { modelId, providerId, customPrompt }
 	}
 
@@ -2239,7 +2241,7 @@ export class Task {
 			}
 
 			if (shouldCompact) {
-				userContent.push({ type: "text", text: summarizeTask(this.focusChainSettings.enabled) })
+				userContent.push({ type: "text", text: summarizeTask(this.focusChainSettings) })
 			}
 		} else {
 			const [parsedUserContent, environmentDetails, clinerulesError] = await this.loadContext(
@@ -2267,6 +2269,12 @@ export class Task {
 
 		telemetryService.captureConversationTurnEvent(this.ulid, providerId, modelId, "user")
 
+		// Capture task initialization timing telemetry for the first API request
+		if (isFirstRequest) {
+			const durationMs = Math.round(performance.now() - this.taskInitializationStartTime)
+			telemetryService.captureTaskInitialization(this.ulid, this.taskId, durationMs, this.enableCheckpoints)
+		}
+
 		// since we sent off a placeholder api_req_started message to update the webview while waiting to actually start the API request (to load potential details for example), we need to update the text of that message
 		const lastApiReqIndex = findLastIndex(this.messageStateHandler.getClineMessages(), (m) => m.say === "api_req_started")
 		await this.messageStateHandler.updateClineMessage(lastApiReqIndex, {
@@ -2613,6 +2621,7 @@ export class Task {
 								localWorkflowToggles,
 								globalWorkflowToggles,
 								this.ulid,
+								this.focusChainSettings,
 							)
 
 							if (needsCheck) {

src/integrations/dify/dify-integration.ts

@@ -0,0 +1,271 @@
+import { DifyHandler } from "../../core/api/providers/dify"
+
+/**
+ * Dify Integration Utilities
+ *
+ * This module provides helper functions to integrate Dify's additional APIs
+ * with Cline's existing systems like file handling, conversation management,
+ * and feedback collection.
+ */
+
+export interface DifyIntegrationOptions {
+	difyHandler: DifyHandler
+	onConversationChange?: (conversationId: string) => void
+	onFileUploaded?: (fileId: string, filename: string) => void
+	onFeedbackSubmitted?: (messageId: string, rating: "like" | "dislike") => void
+}
+
+export class DifyIntegration {
+	private handler: DifyHandler
+	private onConversationChange?: (conversationId: string) => void
+	private onFileUploaded?: (fileId: string, filename: string) => void
+	private onFeedbackSubmitted?: (messageId: string, rating: "like" | "dislike") => void
+
+	constructor(options: DifyIntegrationOptions) {
+		this.handler = options.difyHandler
+		this.onConversationChange = options.onConversationChange
+		this.onFileUploaded = options.onFileUploaded
+		this.onFeedbackSubmitted = options.onFeedbackSubmitted
+	}
+
+	/**
+	 * Upload multiple files and return their IDs for use in conversations
+	 * @param files Array of file data with name and content
+	 * @param user User identifier (defaults to "cline-user")
+	 * @returns Array of uploaded file IDs
+	 */
+	async uploadFiles(files: Array<{ name: string; content: Buffer }>, user?: string): Promise<string[]> {
+		const uploadedFileIds: string[] = []
+
+		for (const file of files) {
+			try {
+				const response = await this.handler.uploadFile(file.content, file.name, user)
+				uploadedFileIds.push(response.id)
+
+				// Notify about successful upload
+				if (this.onFileUploaded) {
+					this.onFileUploaded(response.id, file.name)
+				}
+			} catch (error) {
+				console.error(`Failed to upload file ${file.name}:`, error)
+				throw new Error(`File upload failed for ${file.name}: ${error instanceof Error ? error.message : String(error)}`)
+			}
+		}
+
+		return uploadedFileIds
+	}
+
+	/**
+	 * Enhanced conversation management with callbacks
+	 * @param conversationId Conversation ID to switch to
+	 */
+	async switchToConversation(conversationId: string): Promise<void> {
+		this.handler.setConversationId(conversationId)
+
+		if (this.onConversationChange) {
+			this.onConversationChange(conversationId)
+		}
+	}
+
+	/**
+	 * Start a new conversation and notify listeners
+	 */
+	async startNewConversation(): Promise<void> {
+		this.handler.resetConversation()
+
+		if (this.onConversationChange) {
+			this.onConversationChange("new")
+		}
+	}
+
+	/**
+	 * Get conversation history with error handling and formatting
+	 * @param conversationId Conversation ID (uses current if not provided)
+	 * @param user User identifier
+	 * @param limit Number of messages to fetch
+	 * @returns Formatted conversation history
+	 */
+	async getFormattedConversationHistory(
+		conversationId?: string,
+		user?: string,
+		limit: number = 20,
+	): Promise<Array<{ role: "user" | "assistant"; content: string; timestamp: number; id: string }>> {
+		const currentConversationId = conversationId || this.handler.getCurrentConversationId()
+
+		if (!currentConversationId) {
+			throw new Error("No conversation ID available")
+		}
+
+		try {
+			const history = await this.handler.getConversationHistory(currentConversationId, user, undefined, limit)
+
+			return history.data
+				.map((message) => ({
+					role: "user" as const, // Dify messages are typically user queries
+					content: message.query || message.answer || "",
+					timestamp: message.created_at,
+					id: message.id,
+				}))
+				.reverse() // Reverse to get chronological order
+		} catch (error) {
+			console.error("Failed to get conversation history:", error)
+			throw new Error(`Failed to retrieve conversation history: ${error instanceof Error ? error.message : String(error)}`)
+		}
+	}
+
+	/**
+	 * Submit feedback with enhanced error handling
+	 * @param messageId Message ID to provide feedback for
+	 * @param rating Rating: "like" or "dislike"
+	 * @param content Optional feedback content
+	 * @param user User identifier
+	 */
+	async submitFeedback(messageId: string, rating: "like" | "dislike", content?: string, user?: string): Promise<void> {
+		try {
+			await this.handler.submitMessageFeedback(messageId, rating, content, user)
+
+			if (this.onFeedbackSubmitted) {
+				this.onFeedbackSubmitted(messageId, rating)
+			}
+		} catch (error) {
+			console.error("Failed to submit feedback:", error)
+			throw new Error(`Failed to submit feedback: ${error instanceof Error ? error.message : String(error)}`)
+		}
+	}
+
+	/**
+	 * Get all conversations for the user with enhanced formatting
+	 * @param user User identifier
+	 * @param limit Number of conversations to fetch
+	 * @returns Formatted conversation list
+	 */
+	async getConversationList(
+		user?: string,
+		limit: number = 20,
+	): Promise<
+		Array<{
+			id: string
+			name: string
+			lastUpdated: number
+			status: string
+			messageCount?: number
+		}>
+	> {
+		try {
+			const conversations = await this.handler.getConversations(user, undefined, limit)
+
+			return conversations.data.map((conv) => ({
+				id: conv.id,
+				name: conv.name || "Untitled Conversation",
+				lastUpdated: conv.updated_at,
+				status: conv.status,
+			}))
+		} catch (error) {
+			console.error("Failed to get conversation list:", error)
+			throw new Error(`Failed to retrieve conversations: ${error instanceof Error ? error.message : String(error)}`)
+		}
+	}
+
+	/**
+	 * Auto-rename conversation based on content
+	 * @param conversationId Conversation ID to rename
+	 * @param user User identifier
+	 * @returns New conversation name
+	 */
+	async autoRenameConversation(conversationId?: string, user?: string): Promise<string> {
+		const targetConversationId = conversationId || this.handler.getCurrentConversationId()
+
+		if (!targetConversationId) {
+			throw new Error("No conversation ID available for renaming")
+		}
+
+		try {
+			const result = await this.handler.renameConversation(targetConversationId, user, undefined, true)
+			return result.name
+		} catch (error) {
+			console.error("Failed to auto-rename conversation:", error)
+			throw new Error(`Failed to rename conversation: ${error instanceof Error ? error.message : String(error)}`)
+		}
+	}
+
+	/**
+	 * Delete conversation with confirmation
+	 * @param conversationId Conversation ID to delete
+	 * @param user User identifier
+	 */
+	async deleteConversation(conversationId: string, user?: string): Promise<void> {
+		try {
+			await this.handler.deleteConversation(conversationId, user)
+		} catch (error) {
+			console.error("Failed to delete conversation:", error)
+			throw new Error(`Failed to delete conversation: ${error instanceof Error ? error.message : String(error)}`)
+		}
+	}
+
+	/**
+	 * Stop current generation if task ID is available
+	 * @param taskId Task ID to stop
+	 * @param user User identifier
+	 */
+	async stopCurrentGeneration(taskId: string, user?: string): Promise<void> {
+		try {
+			await this.handler.stopGeneration(taskId, user)
+		} catch (error) {
+			console.error("Failed to stop generation:", error)
+			throw new Error(`Failed to stop generation: ${error instanceof Error ? error.message : String(error)}`)
+		}
+	}
+
+	/**
+	 * Get the underlying Dify handler for direct access
+	 * @returns DifyHandler instance
+	 */
+	getHandler(): DifyHandler {
+		return this.handler
+	}
+}
+
+/**
+ * Helper function to create a Dify integration instance
+ * @param handler DifyHandler instance
+ * @param callbacks Optional callback functions
+ * @returns DifyIntegration instance
+ */
+export function createDifyIntegration(
+	handler: DifyHandler,
+	callbacks?: {
+		onConversationChange?: (conversationId: string) => void
+		onFileUploaded?: (fileId: string, filename: string) => void
+		onFeedbackSubmitted?: (messageId: string, rating: "like" | "dislike") => void
+	},
+): DifyIntegration {
+	return new DifyIntegration({
+		difyHandler: handler,
+		...callbacks,
+	})
+}
+
+/**
+ * Utility function to convert Cline file objects to Dify upload format
+ * @param files Array of file paths or file objects from Cline
+ * @returns Promise with array of file data ready for upload
+ */
+export async function prepareClineFilesForDify(files: string[]): Promise<Array<{ name: string; content: Buffer }>> {
+	const fs = await import("fs")
+	const path = await import("path")
+
+	const fileData: Array<{ name: string; content: Buffer }> = []
+
+	for (const filePath of files) {
+		try {
+			const content = fs.readFileSync(filePath)
+			const name = path.basename(filePath)
+			fileData.push({ name, content })
+		} catch (error) {
+			console.error(`Failed to read file ${filePath}:`, error)
+			throw new Error(`Failed to read file ${filePath}: ${error instanceof Error ? error.message : String(error)}`)
+		}
+	}
+
+	return fileData
+}

src/services/account/ClineAccountService.ts

@@ -240,4 +240,22 @@ export class ClineAccountService {
 			await this._authService.restoreRefreshTokenAndRetrieveAuthInfo()
 		}
 	}
+
+	/**
+	 * Transcribes audio using the Cline transcription service
+	 * @param audioBase64 - Base64 encoded audio data
+	 * @param language - Optional language hint for transcription
+	 * @returns Promise with transcribed text or error
+	 */
+	async transcribeAudio(audioBase64: string, language?: string): Promise<{ text: string }> {
+		const response = await this.authenticatedRequest<{ text: string }>(`/api/v1/chat/transcriptions`, {
+			method: "POST",
+			data: {
+				audioData: audioBase64,
+				language: language || "en",
+			},
+		})
+
+		return response
+	}
 }

src/services/auth/AuthService.ts

@@ -240,7 +240,7 @@ export class AuthService {
 	 * This is typically called when the user logs out.
 	 */
 	async clearAuthToken(): Promise<void> {
-		this._controller.cacheService.setSecret("clineAccountId", undefined)
+		this._controller.stateManager.setSecret("clineAccountId", undefined)
 	}
 
 	/**

src/services/auth/providers/FirebaseAuthProvider.ts

@@ -41,7 +41,7 @@ export class FirebaseAuthProvider {
 	 * @throws {Error} Throws an error if the restoration fails.
 	 */
 	async retrieveClineAuthInfo(controller: Controller): Promise<ClineAuthInfo | null> {
-		const userRefreshToken = controller.cacheService.getSecretKey("clineAccountId")
+		const userRefreshToken = controller.stateManager.getSecretKey("clineAccountId")
 		if (!userRefreshToken) {
 			console.error("No stored authentication credential found.")
 			return null
@@ -122,7 +122,7 @@ export class FirebaseAuthProvider {
 
 			// store the long-lived refresh token in secret storage
 			try {
-				controller.cacheService.setSecret("clineAccountId", userCredential.refreshToken)
+				controller.stateManager.setSecret("clineAccountId", userCredential.refreshToken)
 			} catch (error) {
 				errorService.logMessage("Firebase store token error", "error")
 				errorService.logException(error)

src/services/dictation/AudioRecordingService.ts

@@ -0,0 +1,272 @@
+import { Logger } from "@services/logging/Logger"
+import { ChildProcess, spawn } from "child_process"
+import * as fs from "fs"
+import * as os from "os"
+import * as path from "path"
+import { AUDIO_PROGRAM_CONFIG } from "@/shared/audioProgramConstants"
+
+function isExecutable(filePath: string): boolean {
+	try {
+		fs.accessSync(filePath, fs.constants.X_OK)
+		return true
+	} catch (e) {
+		return false
+	}
+}
+
+export class AudioRecordingService {
+	private recordingProcess: ChildProcess | null = null
+	private isRecording: boolean = false
+	private startTime: number = 0
+	private outputFile: string = ""
+
+	constructor() {}
+
+	async startRecording(): Promise<{ success: boolean; error?: string }> {
+		try {
+			if (this.isRecording) {
+				return { success: false, error: "Already recording" }
+			}
+
+			// Check if recording software is available
+			const checkResult = this.checkRecordingDependencies()
+			if (!checkResult.available) {
+				return { success: false, error: checkResult.error }
+			}
+
+			// Create temporary file for audio output
+			const tempDir = os.tmpdir()
+			this.outputFile = path.join(tempDir, `cline_recording_${Date.now()}.webm`)
+
+			Logger.info("Starting audio recording...")
+
+			// Get the recording program path
+			const recordProgram = this.getRecordProgram()
+			if (!recordProgram) {
+				return { success: false, error: "Recording program not found" }
+			}
+			Logger.info(`Using recording program: ${recordProgram.path}`)
+
+			// Set up recording arguments
+			const args = recordProgram.getArgs(this.outputFile)
+
+			// Spawn the recording process
+			this.recordingProcess = spawn(recordProgram.path, args)
+			this.isRecording = true
+			this.startTime = Date.now()
+
+			// Handle process errors
+			this.recordingProcess.on("error", (error) => {
+				Logger.error(`Recording process error: ${error.message}`)
+				this.isRecording = false
+			})
+
+			// Handle process exit
+			this.recordingProcess.on("exit", (code) => {
+				if (code !== 0 && code !== null) {
+					Logger.warn(`Recording process exited with code: ${code}`)
+				}
+			})
+
+			this.recordingProcess.stderr?.on("data", (data) => {
+				const message = data.toString().trim()
+				if (message && !message.includes("In:") && !message.includes("Out:")) {
+					Logger.info(`Recording stderr: ${message}`)
+				}
+			})
+
+			Logger.info("Audio recording started successfully")
+			return { success: true }
+		} catch (error) {
+			const errorMessage = error instanceof Error ? error.message : String(error)
+			Logger.error("Failed to start audio recording: " + errorMessage)
+			return { success: false, error: `Failed to start recording: ${errorMessage}` }
+		}
+	}
+
+	async stopRecording(): Promise<{ success: boolean; audioBase64?: string; error?: string }> {
+		try {
+			if (!this.isRecording || !this.recordingProcess) {
+				return { success: false, error: "Not currently recording" }
+			}
+
+			Logger.info("Stopping audio recording...")
+
+			// Send SIGINT to stop recording gracefully (like Ctrl+C)
+			this.recordingProcess.kill("SIGINT")
+
+			// Wait for the process to finish
+			await new Promise<void>((resolve) => {
+				if (this.recordingProcess) {
+					// Timeout after 5 seconds
+					const timeoutId = setTimeout(() => {
+						resolve()
+					}, 5000)
+
+					this.recordingProcess.on("exit", (code) => {
+						clearTimeout(timeoutId) // Clear the timeout since process exited
+						resolve()
+					})
+				} else {
+					resolve()
+				}
+			})
+
+			this.recordingProcess = null
+			this.isRecording = false
+
+			// Wait a moment for file to be fully written
+			await new Promise((resolve) => setTimeout(resolve, 500))
+
+			// Read the audio file and convert to base64
+			if (!fs.existsSync(this.outputFile)) {
+				return { success: false, error: "Recording file not found" }
+			}
+
+			const audioBuffer = fs.readFileSync(this.outputFile)
+			const audioBase64 = audioBuffer.toString("base64")
+
+			// Clean up temporary file
+			try {
+				fs.unlinkSync(this.outputFile)
+			} catch (cleanupError) {
+				Logger.warn(
+					"Failed to cleanup temporary audio file: " +
+						(cleanupError instanceof Error ? cleanupError.message : String(cleanupError)),
+				)
+			}
+
+			Logger.info("Audio recording stopped and converted to base64")
+			return { success: true, audioBase64 }
+		} catch (error) {
+			const errorMessage = error instanceof Error ? error.message : String(error)
+			Logger.error("Failed to stop audio recording: " + errorMessage)
+			return { success: false, error: `Failed to stop recording: ${errorMessage}` }
+		}
+	}
+
+	async cancelRecording(): Promise<{ success: boolean; error?: string }> {
+		try {
+			if (!this.isRecording || !this.recordingProcess) {
+				return { success: false, error: "Not currently recording" }
+			}
+
+			Logger.info("Canceling audio recording...")
+
+			// Send SIGINT to stop recording gracefully (like Ctrl+C)
+			this.recordingProcess.kill("SIGINT")
+
+			// Wait for the process to finish
+			await new Promise<void>((resolve) => {
+				if (this.recordingProcess) {
+					// Timeout after 5 seconds
+					const timeoutId = setTimeout(() => {
+						resolve()
+					}, 5000)
+
+					this.recordingProcess.on("exit", (code) => {
+						clearTimeout(timeoutId) // Clear the timeout since process exited
+						resolve()
+					})
+				} else {
+					resolve()
+				}
+			})
+
+			this.recordingProcess = null
+			this.isRecording = false
+
+			// Clean up temporary file without reading it
+			if (this.outputFile && fs.existsSync(this.outputFile)) {
+				try {
+					fs.unlinkSync(this.outputFile)
+				} catch (cleanupError) {
+					Logger.warn(
+						"Failed to cleanup temporary audio file during cancel: " +
+							(cleanupError instanceof Error ? cleanupError.message : String(cleanupError)),
+					)
+				}
+			}
+
+			Logger.info("Audio recording canceled successfully")
+			return { success: true }
+		} catch (error) {
+			const errorMessage = error instanceof Error ? error.message : String(error)
+			Logger.error("Failed to cancel audio recording: " + errorMessage)
+			return { success: false, error: `Failed to cancel recording: ${errorMessage}` }
+		}
+	}
+
+	getRecordingStatus(): { isRecording: boolean; durationSeconds: number; error?: string } {
+		const durationSeconds = this.isRecording ? (Date.now() - this.startTime) / 1000 : 0
+		return {
+			isRecording: this.isRecording,
+			durationSeconds,
+		}
+	}
+
+	private checkRecordingDependencies(): { available: boolean; error?: string } {
+		const program = this.getRecordProgram()
+		if (!program) {
+			const platform = os.platform() as keyof typeof AUDIO_PROGRAM_CONFIG
+			const config = AUDIO_PROGRAM_CONFIG[platform]
+			const error = config ? config.error : `Audio recording is not supported on platform: ${platform}`
+			return { available: false, error }
+		}
+		return { available: true }
+	}
+
+	private getRecordProgram(): { path: string; getArgs: (outputFile: string) => string[] } | undefined {
+		const platform = os.platform() as keyof typeof AUDIO_PROGRAM_CONFIG
+		const config = AUDIO_PROGRAM_CONFIG[platform]
+
+		if (!config) {
+			return undefined
+		}
+
+		// 1. Check if the command is in the system's PATH
+		const pathDirs = (process.env.PATH || "").split(path.delimiter)
+		for (const dir of pathDirs) {
+			const fullPath = path.join(dir, config.command)
+			if (fs.existsSync(fullPath) && isExecutable(fullPath)) {
+				return { path: fullPath, getArgs: config.getArgs }
+			}
+		}
+
+		// 2. Check fallback paths if not in PATH
+		for (const p of config.fallbackPaths) {
+			if (fs.existsSync(p) && isExecutable(p)) {
+				return { path: p, getArgs: config.getArgs }
+			}
+		}
+
+		return undefined
+	}
+
+	// Cleanup method
+	cleanup(): void {
+		if (this.isRecording && this.recordingProcess) {
+			try {
+				this.recordingProcess.kill("SIGINT")
+				this.recordingProcess = null
+				this.isRecording = false
+			} catch (error) {
+				Logger.error("Error during cleanup: " + (error instanceof Error ? error.message : String(error)))
+			}
+		}
+
+		// Clean up any leftover temp files
+		if (this.outputFile && fs.existsSync(this.outputFile)) {
+			try {
+				fs.unlinkSync(this.outputFile)
+			} catch (error) {
+				Logger.warn(
+					"Failed to cleanup temp file during service cleanup: " +
+						(error instanceof Error ? error.message : String(error)),
+				)
+			}
+		}
+	}
+}
+
+export const audioRecordingService = new AudioRecordingService()

src/services/dictation/VoiceTranscriptionService.ts

@@ -0,0 +1,54 @@
+import { Logger } from "@services/logging/Logger"
+import axios from "axios"
+import { ClineAccountService } from "@/services/account/ClineAccountService"
+
+export class VoiceTranscriptionService {
+	private clineAccountService: ClineAccountService
+
+	constructor() {
+		this.clineAccountService = ClineAccountService.getInstance()
+	}
+
+	async transcribeAudio(audioBase64: string, language?: string): Promise<{ text?: string; error?: string }> {
+		try {
+			Logger.info("Transcribing audio with Cline transcription service...")
+
+			const result = await this.clineAccountService.transcribeAudio(audioBase64, language)
+
+			Logger.info("Transcription successful")
+
+			return { text: result.text }
+		} catch (error) {
+			Logger.error("Voice transcription error:", error)
+
+			// Handle axios errors with proper status code mapping
+			if (axios.isAxiosError(error)) {
+				const status = error.response?.status
+				const message = error.response?.data?.message || error.message
+
+				switch (status) {
+					case 401:
+						return { error: "Authentication failed. Please reauthenticate your Cline account" }
+					case 402:
+						return { error: "Insufficient credits for transcription service." }
+					case 400:
+						return { error: "Invalid audio format or request data." }
+					case 500:
+						return { error: "Transcription server error. Please try again later." }
+					default:
+						return { error: `Transcription failed: ${message}` }
+				}
+			}
+
+			// Handle network errors
+			const errorMessage = error instanceof Error ? error.message : String(error)
+			if (errorMessage.includes("ECONNREFUSED") || errorMessage.includes("Network Error")) {
+				return { error: "Cannot connect to transcription service." }
+			}
+
+			return { error: `Network error: ${errorMessage}` }
+		}
+	}
+}
+
+export const voiceTranscriptionService = new VoiceTranscriptionService()

src/services/posthog/telemetry/TelemetryService.ts

@@ -19,7 +19,7 @@ import type { PostHogClientProvider } from "../PostHogClientProvider"
  * When adding a new category, add it both here and to the initial values in telemetryCategoryEnabled
  * Ensure `if (!this.isCategoryEnabled('<category_name>')` is added to the capture method
  */
-type TelemetryCategory = "checkpoints" | "browser" | "focus_chain"
+type TelemetryCategory = "checkpoints" | "browser" | "dictation" | "focus_chain"
 
 /**
  * Maximum length for error messages to prevent excessive data
@@ -29,8 +29,9 @@ const MAX_ERROR_MESSAGE_LENGTH = 500
 export class TelemetryService {
 	// Map to control specific telemetry categories (event types)
 	private telemetryCategoryEnabled: Map<TelemetryCategory, boolean> = new Map([
-		["checkpoints", false], // Checkpoints telemetry disabled
+		["checkpoints", true], // Checkpoints telemetry enabled
 		["browser", true], // Browser telemetry enabled
+		["dictation", true], // Dictation telemetry enabled
 		["focus_chain", true], // Focus Chain telemetry enabled
 	])
 
@@ -106,6 +107,8 @@ export class TelemetryService {
 			RULE_TOGGLED: "task.rule_toggled",
 			// Tracks when auto condense setting is toggled on/off
 			AUTO_CONDENSE_TOGGLED: "task.auto_condense_toggled",
+			// Tracks task initialization timing
+			INITIALIZATION: "task.initialization",
 		},
 		// UI interaction events for tracking user engagement
 		UI: {
@@ -118,6 +121,19 @@ export class TelemetryService {
 			// Tracks when the rules menu button is clicked
 			RULES_MENU_OPENED: "ui.rules_menu_opened",
 		},
+		DICTATION: {
+			// Tracks when voice recording is started
+			RECORDING_STARTED: "voice.recording_started",
+			// Tracks when voice recording is stopped
+			RECORDING_STOPPED: "voice.recording_stopped",
+			// Tracks when voice transcription is started
+			TRANSCRIPTION_STARTED: "voice.transcription_started",
+			// Tracks when voice transcription is completed successfully
+			TRANSCRIPTION_COMPLETED: "voice.transcription_completed",
+			// Tracks when voice transcription fails
+			TRANSCRIPTION_ERROR: "voice.transcription_error",
+			// Tracks when voice feature is enabled or disabled in settings
+		},
 	}
 
 	/** Current version of the extension */
@@ -674,6 +690,143 @@ export class TelemetryService {
 		})
 	}
 
+	// Dictation events
+	/**
+	 * Records when voice recording is started
+	 * @param taskId Optional task identifier if recording was started during a task
+	 * @param platform The platform where recording is happening (macOS, Windows, Linux)
+	 * @param collect If true, collect event instead of sending
+	 */
+	public captureVoiceRecordingStarted(taskId?: string, platform?: string, collect: boolean = false) {
+		if (!this.isCategoryEnabled("dictation")) {
+			return
+		}
+
+		this.capture({
+			event: TelemetryService.EVENTS.DICTATION.RECORDING_STARTED,
+			properties: {
+				taskId,
+				platform: platform || process.platform,
+				timestamp: new Date().toISOString(),
+			},
+		})
+	}
+
+	/**
+	 * Records when voice recording is stopped
+	 * @param taskId Optional task identifier if recording was stopped during a task
+	 * @param durationMs Duration of the recording in milliseconds
+	 * @param success Whether the recording was successful
+	 * @param platform The platform where recording happened
+	 * @param collect If true, collect event instead of sending
+	 */
+	public captureVoiceRecordingStopped(
+		taskId?: string,
+		durationMs?: number,
+		success?: boolean,
+		platform?: string,
+		collect?: boolean,
+	) {
+		if (!this.isCategoryEnabled("dictation")) {
+			return
+		}
+
+		this.capture({
+			event: TelemetryService.EVENTS.DICTATION.RECORDING_STOPPED,
+			properties: {
+				taskId,
+				durationMs,
+				success,
+				platform: platform || process.platform,
+				timestamp: new Date().toISOString(),
+			},
+		})
+	}
+
+	/**
+	 * Records when voice transcription is started
+	 * @param taskId Optional task identifier if transcription was started during a task
+	 * @param audioSizeBytes Size of the audio data being transcribed
+	 * @param language Language hint provided for transcription
+	 * @param collect If true, collect event instead of sending
+	 */
+	public captureVoiceTranscriptionStarted(taskId?: string, language?: string, collect: boolean = false) {
+		if (!this.isCategoryEnabled("dictation")) {
+			return
+		}
+
+		this.capture({
+			event: TelemetryService.EVENTS.DICTATION.TRANSCRIPTION_STARTED,
+			properties: {
+				taskId,
+				language,
+				timestamp: new Date().toISOString(),
+			},
+		})
+	}
+
+	/**
+	 * Records when voice transcription is completed successfully
+	 * @param taskId Optional task identifier if transcription was completed during a task
+	 * @param transcriptionLength Length of the transcribed text
+	 * @param durationMs Time taken for transcription in milliseconds
+	 * @param language Language used for transcription
+	 * @param collect If true, collect event instead of sending
+	 */
+	public captureVoiceTranscriptionCompleted(
+		taskId?: string,
+		transcriptionLength?: number,
+		durationMs?: number,
+		language?: string,
+		collect: boolean = false,
+	) {
+		if (!this.isCategoryEnabled("dictation")) {
+			return
+		}
+
+		this.capture({
+			event: TelemetryService.EVENTS.DICTATION.TRANSCRIPTION_COMPLETED,
+			properties: {
+				taskId,
+				transcriptionLength,
+				durationMs,
+				language,
+				timestamp: new Date().toISOString(),
+			},
+		})
+	}
+
+	/**
+	 * Records when voice transcription fails
+	 * @param taskId Optional task identifier if transcription failed during a task
+	 * @param errorType Type of error that occurred (e.g., "no_openai_key", "api_error", "network_error")
+	 * @param errorMessage The error message
+	 * @param durationMs Time taken before failure in milliseconds
+	 * @param collect If true, collect event instead of sending
+	 */
+	public captureVoiceTranscriptionError(
+		taskId?: string,
+		errorType?: string,
+		errorMessage?: string,
+		durationMs?: number,
+		collect: boolean = false,
+	) {
+		if (!this.isCategoryEnabled("dictation")) {
+			return
+		}
+
+		this.capture({
+			event: TelemetryService.EVENTS.DICTATION.TRANSCRIPTION_ERROR,
+			properties: {
+				taskId,
+				errorType,
+				errorMessage,
+				durationMs,
+				timestamp: new Date().toISOString(),
+			},
+		})
+	}
+
 	/**
 	 * Records when focus chain is enabled/disabled by the user
 	 * @param enabled Whether focus chain was enabled (true) or disabled (false)
@@ -851,6 +1004,25 @@ export class TelemetryService {
 		})
 	}
 
+	/**
+	 * Records task initialization timing and metadata
+	 * @param ulid Unique identifier for the task
+	 * @param taskId Task ID (timestamp in milliseconds when task was created)
+	 * @param durationMs Duration of initialization in milliseconds
+	 * @param hasCheckpoints Whether checkpoints are enabled for this task
+	 */
+	public captureTaskInitialization(ulid: string, taskId: string, durationMs: number, hasCheckpoints: boolean) {
+		this.capture({
+			event: TelemetryService.EVENTS.TASK.INITIALIZATION,
+			properties: {
+				ulid,
+				taskId,
+				durationMs,
+				hasCheckpoints,
+			},
+		})
+	}
+
 	/**
 	 * Records when the rules menu button is clicked to open the rules/workflows modal
 	 */

src/services/test/TestServer.ts

@@ -48,7 +48,7 @@ let messageCatcherDisposable: vscode.Disposable | undefined
  */
 async function updateAutoApprovalSettings(_context: vscode.ExtensionContext, controller?: Controller) {
 	try {
-		const autoApprovalSettings = controller?.cacheService.getGlobalStateKey("autoApprovalSettings")
+		const autoApprovalSettings = controller?.stateManager.getGlobalStateKey("autoApprovalSettings")
 
 		// Enable all actions
 		const updatedSettings: AutoApprovalSettings = {
@@ -67,7 +67,7 @@ async function updateAutoApprovalSettings(_context: vscode.ExtensionContext, con
 			maxRequests: 10000, // Increase max requests for tests
 		}
 
-		controller?.cacheService.setGlobalState("autoApprovalSettings", updatedSettings)
+		controller?.stateManager.setGlobalState("autoApprovalSettings", updatedSettings)
 		Logger.log("Auto approval settings updated for test mode")
 
 		// Update the webview with the new state
@@ -208,7 +208,7 @@ export function createTestServer(controller: Controller): http.Server {
 						Logger.log("API key provided, updating API configuration")
 
 						// Get current API configuration
-						const apiConfiguration = visibleWebview.controller.cacheService.getApiConfiguration()
+						const apiConfiguration = visibleWebview.controller.stateManager.getApiConfiguration()
 
 						// Update API configuration with API key
 						const updatedConfig = {
@@ -218,13 +218,13 @@ export function createTestServer(controller: Controller): http.Server {
 						}
 
 						// Store the API key securely
-						visibleWebview.controller.cacheService.setSecret("clineAccountId", apiKey)
+						visibleWebview.controller.stateManager.setSecret("clineAccountId", apiKey)
 
-						visibleWebview.controller.cacheService.setApiConfiguration(updatedConfig)
+						visibleWebview.controller.stateManager.setApiConfiguration(updatedConfig)
 
 						// Update cache service to use cline provider
-						const currentConfig = visibleWebview.controller.cacheService.getApiConfiguration()
-						visibleWebview.controller.cacheService.setApiConfiguration({
+						const currentConfig = visibleWebview.controller.stateManager.getApiConfiguration()
+						visibleWebview.controller.stateManager.setApiConfiguration({
 							...currentConfig,
 							planModeApiProvider: "cline",
 							actModeApiProvider: "cline",

src/shared/DictationSettings.ts

@@ -0,0 +1,74 @@
+export interface DictationSettings {
+	dictationEnabled: boolean
+	dictationLanguage: string
+}
+
+export const DEFAULT_DICTATION_SETTINGS: DictationSettings = {
+	dictationEnabled: false, // While this service is in Experimental status, we should default to false
+	dictationLanguage: "en",
+}
+
+export interface LanguageItem {
+	name: string
+	code: string
+}
+
+export const SUPPORTED_DICTATION_LANGUAGES: LanguageItem[] = [
+	{ name: "English", code: "en" },
+	{ name: "Spanish (Español)", code: "es" },
+	{ name: "Chinese (中文)", code: "zh" },
+	{ name: "Japanese (日本語)", code: "ja" },
+	{ name: "Afrikaans", code: "af" },
+	{ name: "Arabic (العربية)", code: "ar" },
+	{ name: "Armenian (Հայերեն)", code: "hy" },
+	{ name: "Azerbaijani (Azərbaycan)", code: "az" },
+	{ name: "Belarusian (Беларуская)", code: "be" },
+	{ name: "Bosnian (Bosanski)", code: "bs" },
+	{ name: "Bulgarian (Български)", code: "bg" },
+	{ name: "Catalan (Català)", code: "ca" },
+	{ name: "Croatian (Hrvatski)", code: "hr" },
+	{ name: "Czech (Čeština)", code: "cs" },
+	{ name: "Danish (Dansk)", code: "da" },
+	{ name: "Dutch (Nederlands)", code: "nl" },
+	{ name: "Estonian (Eesti)", code: "et" },
+	{ name: "Finnish (Suomi)", code: "fi" },
+	{ name: "French (Français)", code: "fr" },
+	{ name: "Galician (Galego)", code: "gl" },
+	{ name: "German (Deutsch)", code: "de" },
+	{ name: "Greek (Ελληνικά)", code: "el" },
+	{ name: "Hebrew (עברית)", code: "he" },
+	{ name: "Hindi (हिन्दी)", code: "hi" },
+	{ name: "Hungarian (Magyar)", code: "hu" },
+	{ name: "Icelandic (Íslenska)", code: "is" },
+	{ name: "Indonesian (Bahasa Indonesia)", code: "id" },
+	{ name: "Italian (Italiano)", code: "it" },
+	{ name: "Kannada (ಕನ್ನಡ)", code: "kn" },
+	{ name: "Kazakh (Қазақша)", code: "kk" },
+	{ name: "Korean (한국어)", code: "ko" },
+	{ name: "Latvian (Latviešu)", code: "lv" },
+	{ name: "Lithuanian (Lietuvių)", code: "lt" },
+	{ name: "Macedonian (Македонски)", code: "mk" },
+	{ name: "Malay (Bahasa Melayu)", code: "ms" },
+	{ name: "Marathi (मराठी)", code: "mr" },
+	{ name: "Maori (Te Reo Māori)", code: "mi" },
+	{ name: "Nepali (नेपाली)", code: "ne" },
+	{ name: "Norwegian (Norsk)", code: "no" },
+	{ name: "Persian (فارسی)", code: "fa" },
+	{ name: "Polish (Polski)", code: "pl" },
+	{ name: "Portuguese (Português)", code: "pt" },
+	{ name: "Romanian (Română)", code: "ro" },
+	{ name: "Russian (Русский)", code: "ru" },
+	{ name: "Serbian (Српски)", code: "sr" },
+	{ name: "Slovak (Slovenčina)", code: "sk" },
+	{ name: "Slovenian (Slovenščina)", code: "sl" },
+	{ name: "Swahili (Kiswahili)", code: "sw" },
+	{ name: "Swedish (Svenska)", code: "sv" },
+	{ name: "Tagalog", code: "tl" },
+	{ name: "Tamil (தமிழ்)", code: "ta" },
+	{ name: "Thai (ไทย)", code: "th" },
+	{ name: "Turkish (Türkçe)", code: "tr" },
+	{ name: "Ukrainian (Українська)", code: "uk" },
+	{ name: "Urdu (اردو)", code: "ur" },
+	{ name: "Vietnamese (Tiếng Việt)", code: "vi" },
+	{ name: "Welsh (Cymraeg)", code: "cy" },
+]

src/shared/ExtensionMessage.ts

@@ -4,6 +4,7 @@ import { AutoApprovalSettings } from "./AutoApprovalSettings"
 import { ApiConfiguration } from "./api"
 import { BrowserSettings } from "./BrowserSettings"
 import { ClineRulesToggles } from "./cline-rules"
+import { DictationSettings } from "./DictationSettings"
 import { FocusChainSettings } from "./FocusChainSettings"
 import { HistoryItem } from "./HistoryItem"
 import { McpDisplayMode } from "./McpDisplayMode"
@@ -35,6 +36,7 @@ export interface ExtensionState {
 	apiConfiguration?: ApiConfiguration
 	autoApprovalSettings: AutoApprovalSettings
 	browserSettings: BrowserSettings
+	dictationSettings: DictationSettings
 	remoteBrowserHost?: string
 	preferredLanguage?: string
 	openaiReasoningEffort?: OpenaiReasoningEffort

src/shared/api.ts

@@ -32,6 +32,7 @@ export type ApiProvider =
 	| "groq"
 	| "huggingface"
 	| "huawei-cloud-maas"
+	| "dify"
 	| "baseten"
 	| "vercel-ai-gateway"
 	| "zai"
@@ -105,6 +106,8 @@ export interface ApiHandlerOptions {
 	sapAiCoreTokenUrl?: string
 	sapAiCoreBaseUrl?: string
 	huaweiCloudMaasApiKey?: string
+	difyApiKey?: string
+	difyBaseUrl?: string
 	zaiApiKey?: string
 	zaiApiLine?: string
 	onRetryAttempt?: (attempt: number, maxRetries: number, delay: number, error: any) => void
@@ -3157,6 +3160,8 @@ export type MoonshotModelId = keyof typeof moonshotModels
 export const moonshotDefaultModelId = "kimi-k2-0711-preview" satisfies MoonshotModelId
 
 // Huawei Cloud MaaS
+// Dify.ai - No model selection needed, models are configured in Dify workflows
+
 export type HuaweiCloudMaasModelId = keyof typeof huaweiCloudMaasModels
 export const huaweiCloudMaasDefaultModelId: HuaweiCloudMaasModelId = "DeepSeek-V3"
 export const huaweiCloudMaasModels = {

src/shared/audioProgramConstants.ts

@@ -0,0 +1,72 @@
+export const AUDIO_PROGRAM_CONFIG = {
+	darwin: {
+		command: "ffmpeg",
+		fallbackPaths: ["/usr/local/bin/ffmpeg", "/opt/homebrew/bin/ffmpeg"],
+		getArgs: (outputFile: string) => [
+			"-f",
+			"avfoundation",
+			"-i",
+			":default",
+			"-c:a",
+			"libopus",
+			"-b:a",
+			"32k",
+			"-application",
+			"voip",
+			"-ar",
+			"16000",
+			"-ac",
+			"1",
+			outputFile,
+		],
+		error: "It looks like your system is missing the 'FFmpeg' utility, which is required for voice recording. To install it, please ensure you are in **Act Mode** and then send this message. I will handle the installation for you.\n\n**Installation command:** `brew install ffmpeg`",
+	},
+	linux: {
+		command: "ffmpeg",
+		fallbackPaths: ["/usr/bin/ffmpeg", "/usr/local/bin/ffmpeg", "/snap/bin/ffmpeg"],
+		getArgs: (outputFile: string) => [
+			"-f",
+			"alsa",
+			"-i",
+			"default",
+			"-c:a",
+			"libopus",
+			"-b:a",
+			"32k",
+			"-application",
+			"voip",
+			"-ar",
+			"16000",
+			"-ac",
+			"1",
+			outputFile,
+		],
+		error: "It looks like your system is missing the 'FFmpeg' utility, which is required for voice recording. To install it, please ensure you are in **Act Mode** and then send this message. I will handle the installation for you.\n\n**Installation command:** `sudo apt-get install ffmpeg`",
+	},
+	win32: {
+		command: "ffmpeg",
+		fallbackPaths: [
+			"C:\\ffmpeg\\bin\\ffmpeg.exe",
+			"C:\\Program Files\\ffmpeg\\bin\\ffmpeg.exe",
+			"C:\\Program Files (x86)\\ffmpeg\\bin\\ffmpeg.exe",
+		],
+		getArgs: (outputFile: string) => [
+			"-f",
+			"wasapi",
+			"-i",
+			"audio=default",
+			"-c:a",
+			"libopus",
+			"-b:a",
+			"32k",
+			"-application",
+			"voip",
+			"-ar",
+			"16000",
+			"-ac",
+			"1",
+			outputFile,
+		],
+		error: "It looks like your system is missing the 'FFmpeg' utility, which is required for voice recording. To install it, please ensure you are in **Act Mode** and then send this message. I will handle the installation for you.\n\n**Installation command:** `winget install Gyan.FFmpeg`",
+	},
+}

src/shared/proto-conversions/models/api-configuration-conversion.ts

@@ -254,6 +254,8 @@ function convertApiProviderToProto(provider: string | undefined): ProtoApiProvid
 			return ProtoApiProvider.VERCEL_AI_GATEWAY
 		case "zai":
 			return ProtoApiProvider.ZAI
+		case "dify":
+			return ProtoApiProvider.DIFY
 		default:
 			return ProtoApiProvider.ANTHROPIC
 	}
@@ -330,6 +332,8 @@ function convertProtoToApiProvider(provider: ProtoApiProvider): ApiProvider {
 			return "vercel-ai-gateway"
 		case ProtoApiProvider.ZAI:
 			return "zai"
+		case ProtoApiProvider.DIFY:
+			return "dify"
 		default:
 			return "anthropic"
 	}
@@ -407,6 +411,8 @@ export function convertApiConfigurationToProto(config: ApiConfiguration): ProtoA
 		huaweiCloudMaasApiKey: config.huaweiCloudMaasApiKey,
 		zaiApiLine: config.zaiApiLine,
 		zaiApiKey: config.zaiApiKey,
+		difyApiKey: config.difyApiKey,
+		difyBaseUrl: config.difyBaseUrl,
 
 		// Plan mode configurations
 		planModeApiProvider: config.planModeApiProvider ? convertApiProviderToProto(config.planModeApiProvider) : undefined,
@@ -549,6 +555,8 @@ export function convertProtoToApiConfiguration(protoConfig: ProtoApiConfiguratio
 		huaweiCloudMaasApiKey: protoConfig.huaweiCloudMaasApiKey,
 		zaiApiLine: protoConfig.zaiApiLine,
 		zaiApiKey: protoConfig.zaiApiKey,
+		difyApiKey: protoConfig.difyApiKey,
+		difyBaseUrl: protoConfig.difyBaseUrl,
 
 		// Plan mode configurations
 		planModeApiProvider:

src/shared/proto-conversions/state/settings-conversion.ts

@@ -1,6 +1,10 @@
 import { ApiConfiguration, ApiProvider, BedrockModelId } from "@shared/api"
 import { ApiConfiguration as ProtoApiConfiguration } from "@shared/proto/cline/state"
 
+/**
+ * Domain -> Proto conversions
+ */
+
 /**
  * Converts domain ApiConfiguration objects to proto ApiConfiguration objects
  */
@@ -70,6 +74,8 @@ export function convertApiConfigurationToProtoApiConfiguration(config: ApiConfig
 		sapAiCoreTokenUrl: config.sapAiCoreTokenUrl,
 		sapAiCoreBaseUrl: config.sapAiCoreBaseUrl,
 		vercelAiGatewayApiKey: config.vercelAiGatewayApiKey,
+		difyBaseUrl: config.difyBaseUrl,
+		difyApiKey: config.difyApiKey,
 
 		// Plan mode configurations
 		planModeApiProvider: config.planModeApiProvider,
@@ -140,6 +146,10 @@ export function convertApiConfigurationToProtoApiConfiguration(config: ApiConfig
 	})
 }
 
+/**
+ * Proto -> Domain conversions
+ */
+
 /**
  * Converts proto ApiConfiguration objects to domain ApiConfiguration objects
  */
@@ -208,6 +218,8 @@ export function convertProtoApiConfigurationToApiConfiguration(protoConfig: Prot
 		sapAiCoreTokenUrl: protoConfig.sapAiCoreTokenUrl,
 		sapAiCoreBaseUrl: protoConfig.sapAiCoreBaseUrl,
 		vercelAiGatewayApiKey: protoConfig.vercelAiGatewayApiKey,
+		difyApiKey: protoConfig.difyApiKey,
+		difyBaseUrl: protoConfig.difyBaseUrl,
 
 		// Plan mode configurations
 		planModeApiProvider: protoConfig.planModeApiProvider as ApiProvider,

webview-ui/package.json

@@ -17,7 +17,7 @@
 	"dependencies": {
 		"@floating-ui/react": "^0.27.4",
 		"@fontsource/azeret-mono": "^5.2.9",
-		"@heroui/react": "^2.8.0-beta.2",
+		"@heroui/react": "^2.8.2",
 		"@vscode/webview-ui-toolkit": "^1.4.0",
 		"debounce": "^2.1.1",
 		"dompurify": "^3.2.4",

webview-ui/src/components/chat/ChatTextArea.tsx

@@ -45,6 +45,7 @@ import {
 import { validateApiConfiguration, validateModelId } from "@/utils/validate"
 import ClineRulesToggleModal from "../cline-rules/ClineRulesToggleModal"
 import ServersToggleModal from "./ServersToggleModal"
+import VoiceRecorder from "./VoiceRecorder"
 
 const { MAX_IMAGES_AND_FILES_PER_MESSAGE } = CHAT_CONSTANTS
 
@@ -279,8 +280,15 @@ const ChatTextArea = forwardRef<HTMLTextAreaElement, ChatTextAreaProps>(
 		},
 		ref,
 	) => {
-		const { mode, apiConfiguration, openRouterModels, platform, localWorkflowToggles, globalWorkflowToggles } =
-			useExtensionState()
+		const {
+			mode,
+			apiConfiguration,
+			openRouterModels,
+			platform,
+			localWorkflowToggles,
+			globalWorkflowToggles,
+			dictationSettings,
+		} = useExtensionState()
 		const [isTextAreaFocused, setIsTextAreaFocused] = useState(false)
 		const [isDraggingOver, setIsDraggingOver] = useState(false)
 		const [gitCommits, setGitCommits] = useState<GitCommit[]>([])
@@ -317,6 +325,7 @@ const ChatTextArea = forwardRef<HTMLTextAreaElement, ChatTextAreaProps>(
 		const unsupportedFileTimerRef = useRef<NodeJS.Timeout | null>(null)
 		const [showDimensionError, setShowDimensionError] = useState(false)
 		const dimensionErrorTimerRef = useRef<NodeJS.Timeout | null>(null)
+		const [isVoiceRecording, setIsVoiceRecording] = useState(false)
 
 		const [fileSearchResults, setFileSearchResults] = useState<SearchResult[]>([])
 		const [searchLoading, setSearchLoading] = useState(false)
@@ -1410,6 +1419,11 @@ const ChatTextArea = forwardRef<HTMLTextAreaElement, ChatTextAreaProps>(
 			)
 		}
 
+		const handleSetVoiceRecording = (isRecording: boolean) => {
+			setIsVoiceRecording(isRecording)
+			sendingDisabled = isRecording
+		}
+
 		return (
 			<div>
 				<div
@@ -1601,7 +1615,7 @@ const ChatTextArea = forwardRef<HTMLTextAreaElement, ChatTextAreaProps>(
 							// borderLeft: "9px solid transparent", // NOTE: react-textarea-autosize doesn't calculate correct height when using borderLeft/borderRight so we need to use horizontal padding instead
 							// Instead of using boxShadow, we use a div with a border to better replicate the behavior when the textarea is focused
 							// boxShadow: "0px 0px 0px 1px var(--vscode-input-border)",
-							padding: "9px 28px 9px 9px",
+							padding: `9px ${dictationSettings?.dictationEnabled ? "48" : "28"}px 9px 9px`,
 							cursor: "text",
 							flex: 1,
 							zIndex: 1,
@@ -1654,6 +1668,43 @@ const ChatTextArea = forwardRef<HTMLTextAreaElement, ChatTextAreaProps>(
 								flexDirection: "row",
 								alignItems: "center",
 							}}>
+							{dictationSettings?.dictationEnabled === true && (
+								<VoiceRecorder
+									disabled={sendingDisabled}
+									language={dictationSettings?.dictationLanguage || "en"}
+									onProcessingStateChange={(isProcessing, message) => {
+										if (isProcessing && message) {
+											// Show processing message in input
+											const processingText = inputValue + (inputValue ? " " : "") + `[${message}]`
+											setInputValue(processingText)
+										}
+										// When processing is done, the onTranscription callback will handle the final text
+									}}
+									onRecordingStateChange={handleSetVoiceRecording}
+									onTranscription={(text) => {
+										// Remove any processing text first
+										const processingPattern = /\s*\[Transcribing\.\.\.\]$/
+										const cleanedValue = inputValue.replace(processingPattern, "")
+
+										if (!text) {
+											setInputValue(cleanedValue)
+											return
+										}
+
+										// Append the transcribed text to the cleaned input
+										const newValue = cleanedValue + (cleanedValue ? " " : "") + text
+										setInputValue(newValue)
+										// Focus the textarea and move cursor to end
+										setTimeout(() => {
+											if (textAreaRef.current) {
+												textAreaRef.current.focus()
+												const length = newValue.length
+												textAreaRef.current.setSelectionRange(length, length)
+											}
+										}, 0)
+									}}
+								/>
+							)}
 							{/* <div
 								className={`input-icon-button ${shouldDisableImages ? "disabled" : ""} codicon codicon-device-camera`}
 								onClick={() => {
@@ -1666,16 +1717,18 @@ const ChatTextArea = forwardRef<HTMLTextAreaElement, ChatTextAreaProps>(
 									fontSize: 16.5,
 								}}
 							/> */}
-							<div
-								className={`input-icon-button ${sendingDisabled ? "disabled" : ""} codicon codicon-send`}
-								data-testid="send-button"
-								onClick={() => {
-									if (!sendingDisabled) {
-										setIsTextAreaFocused(false)
-										onSend()
-									}
-								}}
-								style={{ fontSize: 15 }}></div>
+							{!isVoiceRecording && (
+								<div
+									className={`input-icon-button ${sendingDisabled ? "disabled" : ""} codicon codicon-send`}
+									data-testid="send-button"
+									onClick={() => {
+										if (!sendingDisabled) {
+											setIsTextAreaFocused(false)
+											onSend()
+										}
+									}}
+									style={{ fontSize: 15 }}></div>
+							)}
 						</div>
 					</div>
 				</div>

webview-ui/src/components/chat/VoiceRecorder.tsx

@@ -0,0 +1,243 @@
+import { TranscribeAudioRequest } from "@shared/proto/cline/dictation"
+import { EmptyRequest } from "@shared/proto/index.cline"
+import React, { useCallback, useEffect, useRef, useState } from "react"
+import { DictationServiceClient } from "@/services/grpc-client"
+import { formatSeconds } from "@/utils/format"
+import HeroTooltip from "../common/HeroTooltip"
+
+interface VoiceRecorderProps {
+	onTranscription: (text: string) => void
+	onProcessingStateChange?: (isProcessing: boolean, message?: string) => void
+	onRecordingStateChange?: (isRecording: boolean) => void
+	disabled?: boolean
+	language?: string
+}
+
+const MAX_DURATION = 5 * 60 // 5 minutes in seconds
+
+const VoiceRecorder: React.FC<VoiceRecorderProps> = ({
+	onTranscription,
+	onProcessingStateChange,
+	onRecordingStateChange,
+	disabled = false,
+	language = "en",
+}) => {
+	const [isRecording, setIsRecording] = useState(false)
+	const [isProcessing, setIsProcessing] = useState(false)
+	const [recordingDuration, setRecordingDuration] = useState(0)
+	const [error, setError] = useState<string | null>(null)
+	const pollingIntervalRef = useRef<NodeJS.Timeout | null>(null)
+
+	// Notify parent when recording state changes
+	useEffect(() => {
+		onRecordingStateChange?.(isRecording)
+	}, [isRecording, onRecordingStateChange])
+
+	const startRecording = useCallback(async () => {
+		try {
+			setIsRecording(true)
+			setError(null) // Clear any previous errors
+			onProcessingStateChange?.(false) // Clear any previous processing state
+			setRecordingDuration(0) // Reset recording duration
+
+			// Call Extension Host to start recording
+			const response = await DictationServiceClient.startRecording(EmptyRequest.create({}))
+
+			if (!response.success) {
+				console.error("Failed to start recording:", response.error)
+				setIsRecording(false)
+				setError(response.error || "Failed to start recording")
+				return
+			}
+
+			console.log("Recording started successfully")
+		} catch (error) {
+			console.error("Error starting recording:", error)
+			setIsRecording(false)
+			const errorMessage = error instanceof Error ? error.message : "Failed to start recording"
+			setError(errorMessage)
+		}
+	}, [onProcessingStateChange])
+
+	const stopRecording = useCallback(async () => {
+		try {
+			setIsRecording(false)
+			setIsProcessing(true)
+			onProcessingStateChange?.(true, "Processing...")
+
+			// Call Extension Host to stop recording and get audio
+			const response = await DictationServiceClient.stopRecording(EmptyRequest.create({}))
+
+			if (!response.success) {
+				console.error("Failed to stop recording:", response.error)
+				setIsProcessing(false)
+				const errorMessage = response.error || "Failed to stop recording"
+				setError(errorMessage)
+				onTranscription("")
+				return
+			}
+
+			if (!response.audioBase64) {
+				console.error("No audio data received")
+				setIsProcessing(false)
+				const errorMessage = "No audio data received"
+				setError(errorMessage)
+				onTranscription("")
+				return
+			}
+
+			// Update processing state for transcription
+			onProcessingStateChange?.(true, "Transcribing...")
+
+			// Transcribe the audio using OpenAI Whisper
+			const transcriptionResponse = await DictationServiceClient.transcribeAudio(
+				TranscribeAudioRequest.create({
+					audioBase64: response.audioBase64,
+					language: language,
+				}),
+			)
+
+			if (transcriptionResponse.error) {
+				console.error("Transcription error:", transcriptionResponse.error)
+				setError(transcriptionResponse.error)
+				onTranscription("")
+				// Clear the error after a delay
+				setTimeout(() => {
+					setError(null)
+					onProcessingStateChange?.(false)
+				}, 5000)
+			} else if (transcriptionResponse.text) {
+				setError(null)
+				onTranscription(transcriptionResponse.text)
+				onProcessingStateChange?.(false)
+			}
+		} catch (error) {
+			console.error("Error stopping recording:", error)
+			const errorMessage = error instanceof Error ? error.message : "An error occurred"
+			setError(errorMessage)
+			onTranscription("")
+		} finally {
+			setIsProcessing(false)
+		}
+	}, [onTranscription, onProcessingStateChange])
+
+	// Poll recording status while recording to update duration
+	useEffect(() => {
+		const pollRecordingStatus = async () => {
+			try {
+				const statusResponse = await DictationServiceClient.getRecordingStatus(EmptyRequest.create({}))
+				if (statusResponse.isRecording) {
+					setRecordingDuration(Math.floor(statusResponse.durationSeconds))
+
+					// Auto-stop if max duration reached
+					if (statusResponse.durationSeconds >= MAX_DURATION) {
+						stopRecording()
+					}
+				}
+			} catch (error) {
+				console.error("Error polling recording status:", error)
+			}
+		}
+
+		if (isRecording && !isProcessing) {
+			pollingIntervalRef.current = setInterval(pollRecordingStatus, 1000)
+		} else {
+			// Clear polling when not recording
+			if (pollingIntervalRef.current) {
+				clearInterval(pollingIntervalRef.current)
+				pollingIntervalRef.current = null
+			}
+		}
+
+		// Cleanup on unmount
+		return () => {
+			if (pollingIntervalRef.current) {
+				clearInterval(pollingIntervalRef.current)
+				pollingIntervalRef.current = null
+			}
+		}
+	}, [isRecording, isProcessing, stopRecording])
+
+	const cancelRecording = useCallback(async () => {
+		try {
+			setIsRecording(false)
+			setError(null)
+			onProcessingStateChange?.(false)
+			onTranscription("")
+
+			// Call Extension Host to cancel recording
+			const response = await DictationServiceClient.cancelRecording(EmptyRequest.create({}))
+
+			if (!response.success) {
+				console.error("Failed to cancel recording:", response.error)
+				setError(response.error || "Failed to cancel recording")
+				return
+			}
+
+			console.log("Recording canceled successfully")
+		} catch (error) {
+			console.error("Error canceling recording:", error)
+			const errorMessage = error instanceof Error ? error.message : "Failed to cancel recording"
+			setError(errorMessage)
+		}
+	}, [onProcessingStateChange, onTranscription])
+
+	const handleStartClick = useCallback(() => {
+		if (disabled || isProcessing) return
+		if (error) return setError(null)
+		startRecording()
+	}, [startRecording, disabled, isProcessing, error])
+
+	const handleCancelClick = useCallback(() => {
+		if (disabled || isProcessing) return
+		cancelRecording()
+	}, [cancelRecording, disabled, isProcessing])
+
+	const handleStopClick = useCallback(() => {
+		if (disabled || isProcessing) return
+		stopRecording()
+	}, [stopRecording, disabled, isProcessing])
+
+	// When not recording, show single mic button
+	if (!isRecording) {
+		const iconClass = isProcessing ? "codicon-loading" : error ? "codicon-error" : "codicon-mic"
+		const iconColor = error ? "var(--vscode-errorForeground)" : ""
+		const iconAnimation = isProcessing ? "animate-spin" : ""
+		const iconAdjustment = isProcessing ? "mt-0" : error ? "mt-1" : "mt-0.5"
+		const tooltipContent = isProcessing ? "Transcribing..." : error ? `Error: ${error}` : null
+
+		return (
+			<HeroTooltip content={tooltipContent} placement="top">
+				<div
+					className={`input-icon-button mr-1.5 text-base ${iconAdjustment} ${iconAnimation} ${disabled || isProcessing ? "disabled" : ""}`}
+					onClick={handleStartClick}
+					style={{ color: iconColor }}>
+					<span className={`codicon ${iconClass}`} />
+				</div>
+			</HeroTooltip>
+		)
+	}
+
+	return (
+		<div className={`flex items-center ${isRecording ? "mr-0.5" : "mr-1.5"}`}>
+			<HeroTooltip
+				content={`Stop Recording (${formatSeconds(recordingDuration)}/${formatSeconds(MAX_DURATION)})`}
+				placement="top">
+				<div
+					className={`input-icon-button text-base mr-1 mt-1 animate-pulse text-[var(--vscode-errorForeground)] ${disabled || isProcessing ? "disabled" : ""}`}
+					onClick={handleStopClick}>
+					<span className="codicon codicon-stop-circle" />
+				</div>
+			</HeroTooltip>
+			<HeroTooltip content="Cancel Recording" placement="top">
+				<div
+					className={`input-icon-button text-base mt-1 text-[var(--vscode-textForeground)] ${disabled || isProcessing ? "disabled" : ""}`}
+					onClick={handleCancelClick}>
+					<span className="codicon codicon-close" />
+				</div>
+			</HeroTooltip>
+		</div>
+	)
+}
+
+export default VoiceRecorder

webview-ui/src/components/chat/auto-approve-menu/AutoApproveBar.tsx

@@ -73,7 +73,7 @@ const AutoApproveBar = ({ style }: AutoApproveBarProps) => {
 
 	return (
 		<div
-			className="px-[10px] mx-[5px] select-none rounded-[10px_10px_0_0]"
+			className="px-[10px] mx-[15px] select-none rounded-[10px_10px_0_0]"
 			style={{
 				borderTop: `0.5px solid color-mix(in srgb, ${getAsVar(VSC_TITLEBAR_INACTIVE_FOREGROUND)} 20%, transparent)`,
 				overflowY: "auto",

webview-ui/src/components/chat/chat-view/components/layout/ActionButtons.tsx

@@ -85,7 +85,7 @@ export const ActionButtons: React.FC<ActionButtonsProps> = ({
 		}
 
 		return (
-			<div className="flex px-[15px] pt-[10px]">
+			<div className="flex px-[15px]">
 				<VSCodeButton
 					appearance="icon"
 					aria-label="Scroll to bottom"
@@ -107,7 +107,7 @@ export const ActionButtons: React.FC<ActionButtonsProps> = ({
 	const opacity = shouldShowButtons ? (enableButtons || isStreaming ? 1 : 0.5) : 0
 
 	return (
-		<div className={`flex px-[15px] ${shouldShowButtons ? "pt-[10px]" : "pt-0"}`} style={{ opacity }}>
+		<div className="flex px-[15px]" style={{ opacity }}>
 			{primaryButtonText && (
 				<VSCodeButton
 					appearance="primary"

webview-ui/src/components/settings/ApiOptions.tsx

@@ -16,6 +16,7 @@ import { CerebrasProvider } from "./providers/CerebrasProvider"
 import { ClaudeCodeProvider } from "./providers/ClaudeCodeProvider"
 import { ClineProvider } from "./providers/ClineProvider"
 import { DeepSeekProvider } from "./providers/DeepSeekProvider"
+import { DifyProvider } from "./providers/DifyProvider"
 import { DoubaoProvider } from "./providers/DoubaoProvider"
 import { FireworksProvider } from "./providers/FireworksProvider"
 import { GeminiProvider } from "./providers/GeminiProvider"
@@ -174,6 +175,7 @@ const ApiOptions = ({ showModelOptions, apiErrorMessage, modelIdErrorMessage, is
 					<VSCodeOption value="xai">xAI</VSCodeOption>
 					<VSCodeOption value="sambanova">SambaNova</VSCodeOption>
 					<VSCodeOption value="huawei-cloud-maas">Huawei Cloud MaaS</VSCodeOption>
+					<VSCodeOption value="dify">Dify.ai</VSCodeOption>
 				</VSCodeDropdown>
 			</DropdownContainer>
 
@@ -305,6 +307,10 @@ const ApiOptions = ({ showModelOptions, apiErrorMessage, modelIdErrorMessage, is
 				<HuaweiCloudMaasProvider currentMode={currentMode} isPopup={isPopup} showModelOptions={showModelOptions} />
 			)}
 
+			{apiConfiguration && selectedProvider === "dify" && (
+				<DifyProvider currentMode={currentMode} isPopup={isPopup} showModelOptions={showModelOptions} />
+			)}
+
 			{apiConfiguration && selectedProvider === "zai" && (
 				<ZAiProvider currentMode={currentMode} isPopup={isPopup} showModelOptions={showModelOptions} />
 			)}

webview-ui/src/components/settings/CollapsibleContent.tsx

@@ -0,0 +1,16 @@
+import styled from "styled-components"
+
+const CollapsibleContent = styled.div<{ isOpen: boolean }>`
+	overflow: hidden;
+	transition:
+		max-height 0.3s ease-in-out,
+		opacity 0.3s ease-in-out,
+		margin-top 0.3s ease-in-out,
+		visibility 0.3s ease-in-out;
+	max-height: ${({ isOpen }) => (isOpen ? "1000px" : "0")};
+	opacity: ${({ isOpen }) => (isOpen ? 1 : 0)};
+	margin-top: ${({ isOpen }) => (isOpen ? "15px" : "0")};
+	visibility: ${({ isOpen }) => (isOpen ? "visible" : "hidden")};
+`
+
+export default CollapsibleContent

webview-ui/src/components/settings/OpenRouterModelPicker.tsx

@@ -56,7 +56,7 @@ const featuredModels = [
 		label: "New",
 	},
 	{
-		id: "cline/sonic",
+		id: "x-ai/grok-code-fast-1",
 		description: "Advanced model with 262K context for complex coding",
 		label: "Free",
 	},

webview-ui/src/components/settings/providers/DifyProvider.tsx

@@ -0,0 +1,71 @@
+import { Mode } from "@shared/storage/types"
+import { VSCodeTextField } from "@vscode/webview-ui-toolkit/react"
+import { useExtensionState } from "@/context/ExtensionStateContext"
+import { ApiKeyField } from "../common/ApiKeyField"
+import { DebouncedTextField } from "../common/DebouncedTextField"
+import { ModelInfoView } from "../common/ModelInfoView"
+import { normalizeApiConfiguration } from "../utils/providerUtils"
+import { useApiConfigurationHandlers } from "../utils/useApiConfigurationHandlers"
+import { useDebouncedInput } from "../utils/useDebouncedInput"
+
+interface DifyProviderProps {
+	showModelOptions: boolean
+	isPopup?: boolean
+	currentMode: Mode
+}
+
+export const DifyProvider = ({ showModelOptions, isPopup, currentMode }: DifyProviderProps) => {
+	const { apiConfiguration } = useExtensionState()
+	const { handleFieldChange } = useApiConfigurationHandlers()
+
+	// Use debounced input for proper state management
+	const [baseUrlValue, setBaseUrlValue] = useDebouncedInput(apiConfiguration?.difyBaseUrl || "", (value) =>
+		handleFieldChange("difyBaseUrl", value),
+	)
+
+	const [apiKeyValue, setApiKeyValue] = useDebouncedInput(apiConfiguration?.difyApiKey || "", (value) =>
+		handleFieldChange("difyApiKey", value),
+	)
+
+	// Get the normalized configuration
+	const { selectedModelId, selectedModelInfo } = normalizeApiConfiguration(apiConfiguration, currentMode)
+
+	return (
+		<div>
+			<div style={{ display: "flex", flexDirection: "column", gap: 10 }}>
+				<DebouncedTextField
+					initialValue={apiConfiguration?.difyBaseUrl || ""}
+					onChange={(value) => {
+						handleFieldChange("difyBaseUrl", value)
+					}}
+					placeholder={"Enter base URL..."}
+					style={{ width: "100%", marginBottom: 10 }}
+					type="url">
+					<span style={{ fontWeight: 500 }}>Base URL</span>
+				</DebouncedTextField>
+
+				<ApiKeyField
+					initialValue={apiConfiguration?.difyApiKey || ""}
+					onChange={(value) => {
+						handleFieldChange("difyApiKey", value)
+					}}
+					providerName="Dify"
+				/>
+
+				<div style={{ fontSize: "12px", color: "var(--vscode-descriptionForeground)", marginTop: "5px" }}>
+					<p>
+						Dify is a platform that provides access to various AI models through a unified API. Configure your Dify
+						instance URL and API key to get started.
+					</p>
+					<p style={{ marginTop: "8px" }}>
+						<strong>Note:</strong> The model selection is handled within your Dify application configuration.
+					</p>
+				</div>
+			</div>
+
+			{showModelOptions && (
+				<ModelInfoView isPopup={isPopup} modelInfo={selectedModelInfo} selectedModelId={selectedModelId} />
+			)}
+		</div>
+	)
+}

webview-ui/src/components/settings/sections/BrowserSettingsSection.tsx

@@ -5,6 +5,7 @@ import styled from "styled-components"
 import { BROWSER_VIEWPORT_PRESETS } from "../../../../../src/shared/BrowserSettings"
 import { useExtensionState } from "../../../context/ExtensionStateContext"
 import { BrowserServiceClient } from "../../../services/grpc-client"
+import CollapsibleContent from "../CollapsibleContent"
 import { DebouncedTextField } from "../common/DebouncedTextField"
 import Section from "../Section"
 import { updateBrowserSetting } from "../utils/settingsHandlers"
@@ -45,19 +46,6 @@ const ConnectionStatusIndicator = ({
 	)
 }
 
-const CollapsibleContent = styled.div<{ isOpen: boolean }>`
-	overflow: hidden;
-	transition:
-		max-height 0.3s ease-in-out,
-		opacity 0.3s ease-in-out,
-		margin-top 0.3s ease-in-out,
-		visibility 0.3s ease-in-out;
-	max-height: ${({ isOpen }) => (isOpen ? "1000px" : "0")}; // Sufficiently large height
-	opacity: ${({ isOpen }) => (isOpen ? 1 : 0)};
-	margin-top: ${({ isOpen }) => (isOpen ? "15px" : "0")};
-	visibility: ${({ isOpen }) => (isOpen ? "visible" : "hidden")};
-`
-
 export const BrowserSettingsSection: React.FC<BrowserSettingsSectionProps> = ({ renderSectionHeader }) => {
 	const { browserSettings } = useExtensionState()
 	const [isCheckingConnection, setIsCheckingConnection] = useState(false)