Changeset version bump (#7473)

* v3.38.0 Release Notes

- Gemini 3 Pro Preview model
- AquaVoice Avalon model for voice-to-text dictation

- Automatic context truncation when AWS Bedrock token usage rate limits are exceeded
- SAP AI SDK JS packages upgraded to latest major version
- SAP provider OrchestrationClient now matches OrchestrationModuleConfig type and no longer uses invalid promptTemplating property
- Removed new_task tool from system prompts, updated slash command prompts, and added helper function for native tool calling validation

* Update CHANGELOG.md

---------

Co-authored-by: Arafatkatze <arafat.da.khan@gmail.com>

.changeset/fifty-readers-cross.md

@@ -0,0 +1,5 @@
+---
+"claude-dev": patch
+---
+
+Fix task timeline display height.

.changeset/fruity-wolves-fix.md

@@ -0,0 +1,5 @@
+---
+"claude-dev": patch
+---
+
+Ensure tool arguments are streamed during file operations when native tool calling is enabled.

.changeset/light-buckets-wait.md

@@ -0,0 +1,5 @@
+---
+"claude-dev": patch
+---
+
+Feat: add thought signature support for Gemini SDK

.changeset/solid-paths-battle.md

@@ -0,0 +1,5 @@
+---
+"claude-dev": patch
+---
+
+Skip MCP tool with invalid name (e.g. name too long) when native tool calling is enabled.

.vscodeignore

@@ -20,6 +20,9 @@ eslint-rules/**
 .husky/**
 .env
 
+# cli
+cli/**
+
 # Custom
 **/demo.gif
 .nvmrc

CHANGELOG.md

@@ -1,5 +1,53 @@
 # Changelog
 
+## [3.38.0]
+
+### Added
+- Gemini 3 Pro Preview model
+- AquaVoice Avalon model for voice-to-text dictation
+
+### Fixed
+- Automatic context truncation when AWS Bedrock token usage rate limits are exceeded
+- Removed new_task tool from system prompts, updated slash command prompts, and added helper function for native tool calling validation
+
+
+## [3.37.1]
+
+- Comprehensive changes to better support GPT 5.1 - System prompt, tools, deep-planning, focus chain, etc.
+- Add AGENTS.md support
+- feat(models): Add free minimax/mimax-m2 model to the model picker
+
+## [3.37.0]
+
+### Added
+
+- GPT-5.1 with model-specific prompting: tailored system prompts, tool usage, focus chain, and deep-planning optimizations
+- Nous Research provider with Hermes 4 model family and custom system prompts
+- Switched to Aqua Voice's Avalon model in speech to text transcription
+- Added Linux support for speech to text
+- Model-family breakouts for deep-planning prompting, laying groundwork for enhanced slash commands
+- Expanded HTTP proxy support throughout the codebase
+- Improved focus chain prompting for frontier models (Anthropic, OpenAI, Gemini, xAI)
+
+### Fixed
+
+- Duplicate tool results prevention through existence checking
+- XML entity escaping in model content processor
+- Commit message generation in command palette
+- OpenAI Compatible provider temperature parameter type conversion
+
+## Documentation
+
+- Added missing proto generation step in CONTRIBUTING.md
+- New `npm run dev` script for streamlined terminal workflow (fixes #7335)
+
+## [3.36.1]
+
+- fix: remove native tool calling support from Gemini and XAI provider due to invalid tool names issues
+- fix: disable native tool callings for grok code models
+- Add MCP tool usage to GLM
+- Removes reasoning_details content field from Anthropic providers
+
 ## [3.36.0]
 
 - Add: Hooks allow you to inject custom logic into Cline's workflow

CONTRIBUTING.md

@@ -46,7 +46,11 @@ We also welcome contributions to our [documentation](https://github.com/cline/cl
     ```bash
     npm run install:all
     ```
-4. Launch by pressing `F5` (or `Run`->`Start Debugging`) to open a new VSCode window with the extension loaded. (You may need to install the [esbuild problem matchers extension](https://marketplace.visualstudio.com/items?itemName=connor4312.esbuild-problem-matchers) if you run into issues building the project.)
+4. Generate Protocol Buffer files (required before first build):
+    ```bash
+    npm run protos
+    ```
+5. Launch by pressing `F5` (or `Run`->`Start Debugging`) to open a new VSCode window with the extension loaded. (You may need to install the [esbuild problem matchers extension](https://marketplace.visualstudio.com/items?itemName=connor4312.esbuild-problem-matchers) if you run into issues building the project.)
 
 
 
@@ -85,8 +89,10 @@ We also welcome contributions to our [documentation](https://github.com/cline/cl
 
 2. **Local Development**
     - Run `npm run install:all` to install dependencies
+    - Run `npm run protos` to generate Protocol Buffer files (required before first build)
     - Run `npm run test` to run tests locally
     - Run → Start Debugging or `>Debug: Select and Start Debugging` and wait for a new VS Code instance to open
+    - **Terminal Workflow**: Use `npm run dev` (generates protos + runs watch mode) or `npm run watch` (if protos already generated)
     - Before submitting PR, run `npm run format:fix` to format your code
 
 3. **Linux-specific Setup**

cli/pkg/cli/auth/byo_quick_setup.go

@@ -75,6 +75,12 @@ func QuickSetupFromFlags(ctx context.Context, provider, apiKey, modelID, baseURL
 		}
 	}
 
+	// Flush pending state changes to disk immediately
+	// This ensures all configuration changes are persisted before the instance terminates
+	if _, err := manager.GetClient().State.FlushPendingState(ctx, &cline.EmptyRequest{}); err != nil {
+		return fmt.Errorf("failed to flush pending state: %w", err)
+	}
+
 	// Success message
 	fmt.Printf("\n✓ Successfully configured %s provider\n", GetProviderDisplayName(providerEnum))
 	fmt.Printf("  Model: %s\n", finalModelID)
@@ -170,6 +176,7 @@ func validateQuickSetupProvider(providerID string) (cline.ApiProvider, error) {
 		cline.ApiProvider_XAI:           true,
 		cline.ApiProvider_CEREBRAS:      true,
 		cline.ApiProvider_OLLAMA:        true,
+		cline.ApiProvider_NOUSRESEARCH:  true,
 	}
 
 	if !supportedProviders[provider] {

cli/pkg/cli/auth/providers_byo.go

@@ -26,6 +26,7 @@ func GetBYOProviderList() []BYOProviderOption {
 		{Name: "Google Gemini", Provider: cline.ApiProvider_GEMINI},
 		{Name: "Ollama", Provider: cline.ApiProvider_OLLAMA},
 		{Name: "Cerebras", Provider: cline.ApiProvider_CEREBRAS},
+		{Name: "NousResearch", Provider: cline.ApiProvider_NOUSRESEARCH},
 		{Name: "Oracle Code Assist", Provider: cline.ApiProvider_OCA},
 	}
 }
@@ -100,6 +101,8 @@ func GetBYOProviderPlaceholder(provider cline.ApiProvider) string {
 		return "e.g., qwen3-coder:30b"
 	case cline.ApiProvider_CEREBRAS:
 		return "e.g., gpt-oss-120b"
+	case cline.ApiProvider_NOUSRESEARCH:
+		return "e.g., Hermes-4-405B"
 	case cline.ApiProvider_OCA:
 		return "e.g., oca/llama4"
 	default:

cli/pkg/cli/auth/providers_list.go

@@ -111,6 +111,7 @@ func (r *ProviderListResult) GetAllReadyProviders() []*ProviderDisplay {
 		cline.ApiProvider_GEMINI,
 		cline.ApiProvider_OLLAMA,
 		cline.ApiProvider_CEREBRAS,
+		cline.ApiProvider_NOUSRESEARCH,
 		cline.ApiProvider_OCA,
 		cline.ApiProvider_HICAP,
 	}
@@ -241,6 +242,8 @@ func mapProviderStringToEnum(providerStr string) (cline.ApiProvider, bool) {
 		return cline.ApiProvider_OCA, true
 	case "hicap":
 		return cline.ApiProvider_HICAP, true
+	case "nousResearch":
+		return cline.ApiProvider_NOUSRESEARCH, true
 	default:
 		return cline.ApiProvider_ANTHROPIC, false // Return 0 value with false
 	}
@@ -274,6 +277,8 @@ func GetProviderIDForEnum(provider cline.ApiProvider) string {
 		return "oca"
 	case cline.ApiProvider_HICAP:
 		return "hicap"
+	case cline.ApiProvider_NOUSRESEARCH:
+		return "nousResearch"
 	default:
 		return ""
 	}
@@ -353,6 +358,8 @@ func GetProviderDisplayName(provider cline.ApiProvider) string {
 		return "Oracle Code Assist"
 	case cline.ApiProvider_HICAP:
 		return "Hicap"
+	case cline.ApiProvider_NOUSRESEARCH:
+		return "NousResearch"
 	default:
 		return "Unknown"
 	}
@@ -475,6 +482,7 @@ func DetectAllConfiguredProviders(ctx context.Context, manager *task.Manager) ([
 		{cline.ApiProvider_OLLAMA, "ollamaBaseUrl"}, // Ollama uses baseUrl instead of API key
 		{cline.ApiProvider_CEREBRAS, "cerebrasApiKey"},
 		{cline.ApiProvider_HICAP, "hicapApiKey"},
+		{cline.ApiProvider_NOUSRESEARCH, "nousResearchApiKey"},
 	}
 
 	for _, providerCheck := range providersToCheck {

cli/pkg/cli/auth/update_api_configurations.go

@@ -163,6 +163,15 @@ func GetProviderFields(provider cline.ApiProvider) (ProviderFields, error) {
 			ActModeProviderSpecificModelIDField:  "actModeHicapModelId",
 		}, nil
 
+	case cline.ApiProvider_NOUSRESEARCH:
+		return ProviderFields{
+			APIKeyField:                          "nousResearchApiKey",
+			PlanModeModelIDField:                 "planModeApiModelId",
+			ActModeModelIDField:                  "actModeApiModelId",
+			PlanModeProviderSpecificModelIDField: "planModeNousResearchModelId",
+			ActModeProviderSpecificModelIDField:  "actModeNousResearchModelId",
+		}, nil
+
 	default:
 		return ProviderFields{}, fmt.Errorf("unsupported provider: %v", provider)
 	}
@@ -278,6 +287,8 @@ func setAPIKeyField(apiConfig *cline.ModelsApiConfiguration, fieldName string, v
 		apiConfig.OcaApiKey = value
 	case "hicapApiKey":
 		apiConfig.HicapApiKey = value
+	case "nousResearchApiKey":
+		apiConfig.NousResearchApiKey = value
 	}
 }
 
@@ -302,6 +313,9 @@ func setProviderSpecificModelID(apiConfig *cline.ModelsApiConfiguration, fieldNa
 	case "planModeHicapModelId":
 		apiConfig.PlanModeHicapModelId = value
 		apiConfig.ActModeHicapModelId = value
+	case "planModeNousResearchModelId":
+		apiConfig.PlanModeNousResearchModelId = value
+		apiConfig.ActModeNousResearchModelId = value
 	}
 }
 

cli/pkg/cli/auth/wizard_byo.go

@@ -517,7 +517,7 @@ func (pw *ProviderWizard) applyModelChange(provider cline.ApiProvider, modelID s
 		ModelInfo: modelInfo,
 	}
 
-	return UpdateProviderPartial(pw.ctx, pw.manager, provider, updates, false)
+	return UpdateProviderPartial(pw.ctx, pw.manager, provider, updates, true)
 }
 
 // SwitchToBYOProvider switches to a BYO provider that's already configured.

cli/pkg/cli/display/tool_renderer.go

@@ -106,6 +106,14 @@ func (tr *ToolRenderer) generateToolHeader(tool *types.ToolMessage, verbTense st
 		}
 		return fmt.Sprintf("### Cline %s `%s`", action, tool.Path)
 
+	case string(types.ToolTypeFileDeleted):
+		if verbTense == "wants to" {
+			action = "wants to delete"
+		} else {
+			action = "is deleting"
+		}
+		return fmt.Sprintf("### Cline %s `%s`", action, tool.Path)
+
 	case string(types.ToolTypeListFilesTopLevel):
 		if verbTense == "wants to" {
 			action = "wants to list files in"
@@ -199,7 +207,7 @@ func (tr *ToolRenderer) GenerateToolContentPreview(tool *types.ToolMessage) stri
 		previewMd := fmt.Sprintf("```\n%s\n```", preview)
 		return tr.renderMarkdown(previewMd)
 
-	case string(types.ToolTypeReadFile), string(types.ToolTypeWebFetch):
+	case string(types.ToolTypeReadFile), string(types.ToolTypeWebFetch), string(types.ToolTypeFileDeleted):
 		// No preview for read/fetch operations
 		return ""
 
@@ -226,7 +234,8 @@ func (tr *ToolRenderer) GenerateToolContentBody(tool *types.ToolMessage) string
 	toolParser := NewToolResultParser(tr.mdRenderer)
 
 	switch tool.Tool {
-	case string(types.ToolTypeReadFile):
+	case string(types.ToolTypeReadFile),
+		string(types.ToolTypeFileDeleted):
 		// readFile: show header only, no body
 		return ""
 

cli/pkg/cli/task/input_handler.go

@@ -256,6 +256,8 @@ func determineAutoApprovalAction(msg *types.ClineMessage) (string, error) {
 		case types.ToolTypeEditedExistingFile,
 			types.ToolTypeNewFileCreated:
 			return "edit_files", nil
+		case types.ToolTypeFileDeleted:
+			return "apply_patch", nil
 		default:
 			return "", fmt.Errorf("unsupported tool type: %s", toolMsg.Tool)
 		}

cli/pkg/cli/types/messages.go

@@ -107,6 +107,7 @@ const (
 	ToolTypeEditedExistingFile      ToolType = "editedExistingFile"
 	ToolTypeNewFileCreated          ToolType = "newFileCreated"
 	ToolTypeReadFile                ToolType = "readFile"
+	ToolTypeFileDeleted             ToolType = "fileDeleted"
 	ToolTypeListFilesTopLevel       ToolType = "listFilesTopLevel"
 	ToolTypeListFilesRecursive      ToolType = "listFilesRecursive"
 	ToolTypeListCodeDefinitionNames ToolType = "listCodeDefinitionNames"

cli/pkg/generated/providers.go

@@ -145,6 +145,7 @@ const (
 	XAI = "xai"
 	CEREBRAS = "cerebras"
 	OCA = "oca"
+	NOUSRESEARCH = "nousResearch"
 )
 
 // AllProviders returns a slice of enabled provider IDs for the CLI build.
@@ -161,6 +162,7 @@ var AllProviders = []string{
 	"xai",
 	"cerebras",
 	"oca",
+	"nousResearch",
 }
 
 // ConfigField represents a configuration field requirement
@@ -318,6 +320,15 @@ var rawConfigFields = `	[
 	    "fieldType": "password",
 	    "placeholder": "Enter your API key"
 	  },
+	  {
+	    "name": "nousResearchApiKey",
+	    "type": "string",
+	    "comment": "",
+	    "category": "nousResearch",
+	    "required": true,
+	    "fieldType": "password",
+	    "placeholder": "Enter your API key"
+	  },
 	  {
 	    "name": "ulid",
 	    "type": "string",
@@ -435,6 +446,15 @@ var rawConfigFields = `	[
 	    "fieldType": "url",
 	    "placeholder": "https://api.example.com"
 	  },
+	  {
+	    "name": "minimaxApiLine",
+	    "type": "string",
+	    "comment": "",
+	    "category": "general",
+	    "required": false,
+	    "fieldType": "string",
+	    "placeholder": ""
+	  },
 	  {
 	    "name": "ocaMode",
 	    "type": "string",
@@ -775,6 +795,24 @@ var rawModelDefinitions = `	{
 	      "supportsImages": false,
 	      "supportsPromptCache": false,
 	      "description": "A compact 20B open-weight Mixture-of-Experts language model designed for strong reasoning and tool use, ideal for edge devices and local inference."
+	    },
+	    "qwen.qwen3-coder-30b-a3b-v1:0": {
+	      "maxTokens": 8192,
+	      "contextWindow": 262144,
+	      "inputPrice": 0,
+	      "outputPrice": 0,
+	      "supportsImages": false,
+	      "supportsPromptCache": false,
+	      "description": "Qwen3 Coder 30B MoE model with 3.3B activated parameters, optimized for code generation and analysis with 256K context window."
+	    },
+	    "qwen.qwen3-coder-480b-a35b-v1:0": {
+	      "maxTokens": 8192,
+	      "contextWindow": 262144,
+	      "inputPrice": 0,
+	      "outputPrice": 1,
+	      "supportsImages": false,
+	      "supportsPromptCache": false,
+	      "description": "Qwen3 Coder 480B flagship MoE model with 35B activated parameters, designed for complex coding tasks with advanced reasoning capabilities and 256K context window."
 	    }
 	  },
 	  "gemini": {
@@ -1263,6 +1301,26 @@ var rawModelDefinitions = `	{
 	      "supportsPromptCache": false,
 	      "description": "SOTA performance with ~1500 tokens/s"
 	    }
+	  },
+	  "nousResearch": {
+	    "Hermes-4-405B": {
+	      "maxTokens": 8192,
+	      "contextWindow": 128000,
+	      "inputPrice": 0,
+	      "outputPrice": 0,
+	      "supportsImages": false,
+	      "supportsPromptCache": false,
+	      "description": "This is the largest model in the Hermes 4 family, and it is the fullest expression of our design, focused on advanced reasoning and creative depth rather than optimizing inference speed or cost."
+	    },
+	    "Hermes-4-70B": {
+	      "maxTokens": 8192,
+	      "contextWindow": 128000,
+	      "inputPrice": 0,
+	      "outputPrice": 0,
+	      "supportsImages": false,
+	      "supportsPromptCache": false,
+	      "description": "This incarnation of Hermes 4 balances scale and size. It handles complex reasoning tasks, while staying fast and cost effective. A versatile choice for many use cases."
+	    }
 	  }
 	}`
 
@@ -1432,6 +1490,18 @@ func GetProviderDefinitions() (map[string]ProviderDefinition, error) {
 		HasDynamicModels: false,
 		SetupInstructions: `Configure Oca API credentials`,
 	}
+
+	// NousResearch
+	definitions["nousResearch"] = ProviderDefinition{
+		ID:              "nousResearch",
+		Name:            "NousResearch",
+		RequiredFields:  getFieldsByProvider("nousResearch", configFields, true),
+		OptionalFields:  getFieldsByProvider("nousResearch", configFields, false),
+		Models:          modelDefinitions["nousResearch"],
+		DefaultModelID:  "Hermes-4-405B",
+		HasDynamicModels: false,
+		SetupInstructions: `Configure NousResearch API credentials`,
+	}
 	
 	return definitions, nil
 }
@@ -1459,6 +1529,7 @@ func GetProviderDisplayName(providerID string) string {
 		"xai": "X AI (Grok)",
 		"cerebras": "Cerebras",
 		"oca": "Oca",
+		"nousResearch": "NousResearch",
 	}
 	
 	if name, exists := displayNames[providerID]; exists {

docs/features/cline-rules.mdx

@@ -81,6 +81,20 @@ your-project/
 
 Cline's system prompt, on the other hand, is not user-editable ([here's where you can find it](https://github.com/cline/cline/blob/main/src/core/prompts/system.ts)). For a broader look at prompt engineering best practices, check out [this resource](https://docs.anthropic.com/en/docs/build-with-claude/prompt-engineering/overview).
 
+### AGENTS.md Standard Support
+
+Cline also supports the [AGENTS.md](https://agents.md/) standard as a fallback 
+(in addition to Cline Rules) by automatically detecting `AGENTS.md` files in 
+your workspace root. This allows you to use the same rules file across different AI 
+coding tools.
+
+```
+your-project/
+├── AGENTS.md
+├── src/
+└── ...
+```
+
 ### Tips for Writing Effective Cline Rules
 
 -   Be Clear and Concise: Use simple language and avoid ambiguity.

docs/features/dictation.mdx

@@ -3,7 +3,7 @@ title: "Dictation"
 description: "Communicate with Cline using your voice for faster, more natural AI collaboration"
 ---
 
-Dictation transforms how you work with AI. Instead of typing out complex thoughts, you speak naturally and share your complete intent. This isn't just about speed - though voice is faster - it's about unlocking the kind of fluid collaboration that typing can't match.
+Dictation transforms how you work with AI. Instead of typing out complex thoughts, you speak naturally and share your complete intent. This isn't just about speed - though voice is faster - it's about enabling fluid collaboration that typing can't match.
 
 ## Why Voice Changes Everything
 
@@ -35,11 +35,14 @@ Dictation works with any AI model you've configured. The transcription happens t
 
 ## System Requirements
 
+<Note>
+Dictation is currently not available on Windows. Support for Windows is planned for a future release.
+</Note>
+
 Dictation 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.
 

docs/features/hooks.mdx

@@ -76,7 +76,7 @@ echo "$input" | jq -r '.timestamp | type'
 
 This example script demonstrates the key mechanics of hook input/output: reading the JSON payload from stdin with `input=$(cat)`, and using `jq` to inspect the data structure and field types that your hook receives. This helps you understand what data is available before building more complex hook logic.
 
-#### Make it executable
+**Make it executable**
 
 ```bash
 chmod +x .clinerules/hooks/TaskStart
@@ -92,6 +92,32 @@ Start a task in Cline and verify your hook executes.
 Start with a simple hook that just logs information before building complex validation logic. This helps you understand the data structure and timing.
 </Tip>
 
+
+## What You Can Build
+
+Once you understand the basics, hooks open up creative possibilities:
+
+<CardGroup cols={2}>
+  <Card title="Intelligent Code Review" icon="code-branch">
+    Run linters or custom validators before files get saved. Block commits that don't pass checks. Track code quality metrics over time.
+  </Card>
+  
+  <Card title="Security Enforcement" icon="shield-halved">
+    Prevent operations that violate security policies. Detect when sensitive data might be exposed. Audit all file access for compliance.
+  </Card>
+  
+  <Card title="Development Analytics" icon="chart-line">
+    Measure how long different operations take. Identify patterns in how the AI works. Generate productivity reports from hook data.
+  </Card>
+  
+  <Card title="Integration Hub" icon="plug">
+    Connect to issue trackers when certain keywords appear. Update project management tools. Sync with external APIs at the right moments.
+  </Card>
+</CardGroup>
+
+The key is combining hooks with external tools. A hook can be the glue between Cline's workflow and the rest of your development ecosystem.
+
+
 ## Hook Types
 
 Cline provides multiple hook types that let you tap into different stages of the AI workflow. They're organized into categories based on their trigger points and use cases.
@@ -341,22 +367,6 @@ Context injection affects future decisions, not current ones. When a hook runs:
 This means PreToolUse hooks are for blocking bad actions, while PostToolUse hooks are for learning from completed ones.
 
 
-## What You Can Build
-
-Once you understand the basics, hooks open up creative possibilities:
-
-- **Intelligent Code Review**:
-Run linters or custom validators before files get saved. Block commits that don't pass checks. Track code quality metrics over time.
-
-- **Security Enforcement**:
-Prevent operations that violate security policies. Detect when sensitive data might be exposed. Audit all file access for compliance.
-
-- **Development Analytics**: Measure how long different operations take. Identify patterns in how the AI works. Generate productivity reports from hook data.
-
-- **Integration Hub**: Connect to issue trackers when certain keywords appear. Update project management tools. Sync with external APIs at the right moments.
-
-The key is combining hooks with external tools. A hook can be the glue between Cline's workflow and the rest of your development ecosystem.
-
 ## Troubleshooting
 
 ### Hook Not Running
@@ -371,12 +381,30 @@ The key is combining hooks with external tools. A hook can be the glue between C
 - Consider moving complex logic to a background process
 
 ### Context Not Affecting Behavior
-- Remember: context affects FUTURE decisions, not the current tool
-- The current AI behavior is based on the previous "API Request..." block
-- Your `contextModification` gets injected into the NEXT "API Request..." block
-- Use PreToolUse for validation (blocking) if you need immediate effect
-- Ensure context modifications are clear and actionable
-- Check that context isn't being truncated (50KB limit)
+Remember that context modifications affect future AI decisions, not the current operation. The AI's current behavior is based on the previous "API Request..." block, and your `contextModification` gets injected into the next "API Request..." block. This means if you need immediate effect, you should use PreToolUse hooks for validation and return `cancel: true` in your hook's JSON response to block Cline from continuing.
+
+When adding context, ensure your modifications are clear and actionable so the AI can understand and apply them effectively. Also check that your context isn't being truncated due to the 50KB limit, as this could prevent important information from reaching the AI.
+
+### Handling Strings with Quotes in JSON Payloads
+When your hook needs to include strings containing unescaped quote characters (`"`) in JSON output, use jq's `--arg` flag for proper escaping:
+
+```bash
+#!/usr/bin/env bash
+
+# When $output contains unescaped quote characters (")...
+output='{"foo":"bar"}'
+
+# Use the --arg flag for automatic string escaping
+jq -n --arg ctx "$output" '{cancel: false, contextModification: $ctx}'
+
+# This will result in:
+# {
+#     "cancel": false,
+#     "contextModification": "{\"foo\":\"bar\"}"
+# }
+```
+
+The `--arg` flag automatically escapes special characters, preventing JSON parsing errors when your context modification includes complex strings or nested JSON structures.
 
 <Warning>
 Hooks run with the same permissions as VS Code. They can access all workspace files and environment variables. Review hooks from untrusted sources before enabling them.

docs/mcp/connecting-to-a-remote-server.mdx

@@ -29,11 +29,33 @@ The "Remote Servers" tab allows you to connect to any MCP server that's accessib
 2. Fill in the required information:
     - **Server Name**: Provide a unique, descriptive name for the server
     - **Server URL**: Enter the complete URL endpoint of the MCP server (e.g., `https://example.com/mcp-sse`)
+    - **Transport Type**: Select the connection protocol (Streamable HTTP is recommended for modern servers)
 3. Click "Add Server" to initiate the connection
 4. Cline will attempt to connect to the server and display the connection status
 
 > **Note**: When connecting to a remote server, ensure you trust the source, as MCP servers can execute code in your environment.
 
+#### Transport Types
+
+Cline supports two transport protocols for remote MCP servers:
+
+-   **Streamable HTTP (Recommended)**: The modern MCP transport protocol with better performance, reliability, and full OAuth 2.1 authentication support. Use this for most remote servers.
+-   **SSE (Legacy)**: Server-Sent Events transport. Use this only if the server specifically requires SSE or doesn't support Streamable HTTP.
+
+#### OAuth Authentication
+
+Some MCP servers (like Vercel's MCP) require OAuth authentication to access your data securely. When connecting to an OAuth-enabled server:
+
+1. Add the server as usual with its URL
+2. If the server requires authentication, you'll see an error message asking to authenticate.
+3. Click the **"Authenticate"** button that appears
+4. Your browser will open to the server's authorization page
+5. Sign in and grant permission
+6. You'll be redirected back to Cline automatically
+7. The server will connect and show a green status dot
+
+Once authenticated, your credentials are securely stored and the server will reconnect automatically when you reload Cline. You won't need to authenticate again unless you delete the server or your credentials expire.
+
 ### Remote Server Discovery
 
 If you're looking for MCP servers to connect to, several third-party marketplaces provide directories of available servers with various capabilities.
@@ -90,9 +112,20 @@ Toggle the switch next to each server to enable or disable it:
 If a server fails to connect:
 
 1. An error message will be displayed with details about the failure
-2. Check that the server URL is correct and the server is running
-3. Use the "Restart Server" button to attempt reconnection
-4. If problems persist, you can delete the server and try adding it again
+2. **For OAuth errors**: Click the "Authenticate" button to complete the authorization flow
+3. Check that the server URL is correct and the server is running
+4. Try selecting a different transport type (Streamable HTTP vs SSE)
+5. Use the "Restart Server" button to attempt reconnection
+6. If problems persist, you can delete the server and try adding it again
+
+#### OAuth-Specific Issues
+
+If you're having trouble authenticating with an OAuth-enabled server:
+
+-   **"Authentication required" persists**: Make sure you completed the authorization flow in your browser and didn't cancel it
+-   **Browser doesn't open**: Check your system's default browser settings and ensure external URLs can be opened
+-   **Redirect errors**: Verify you're using the latest version of Cline - older versions may not support OAuth
+-   **Reset authentication**: Delete the server and re-add it to start fresh with a new OAuth flow
 
 ### Advanced Configuration
 
@@ -105,10 +138,11 @@ For advanced users, Cline stores MCP server configurations in a JSON file that c
 {
 	"mcpServers": {
 		"exampleServer": {
-			"url": "https://example.com/mcp-sse",
+			"url": "https://example.com/mcp-server",
+			"type": "streamableHttp",
 			"disabled": false,
 			"autoApprove": ["tool1", "tool2"],
-			"timeout": 30
+			"timeout": 60
 		}
 	}
 }
@@ -117,9 +151,10 @@ For advanced users, Cline stores MCP server configurations in a JSON file that c
 Key configuration options:
 
 -   **url**: The endpoint URL (for remote servers)
+-   **type**: Transport protocol - `"streamableHttp"` (recommended) or `"sse"` (legacy)
 -   **disabled**: Whether the server is currently enabled (true/false)
 -   **autoApprove**: List of tool names that don't require confirmation
--   **timeout**: Maximum time in seconds to wait for server responses
+-   **timeout**: Maximum time in seconds to wait for server responses (default: 60)
 
 For additional MCP settings, click the "Advanced MCP Settings" link to access VSCode settings.
 
@@ -130,3 +165,20 @@ Once connected, Cline can use the tools and resources provided by the MCP server
 1. A tool approval prompt will appear (unless auto-approved)
 2. Review the tool details and parameters before approving
 3. The tool will execute and return results to Cline
+
+### Example: Connecting to Vercel MCP
+
+[Vercel MCP](https://vercel.com/docs/mcp/vercel-mcp) is an OAuth-enabled server that provides tools for managing your Vercel projects and deployments:
+
+1. Click "Remote Servers" tab
+2. Enter:
+    - **Server Name**: `vercel`
+    - **Server URL**: `https://mcp.vercel.com`
+    - **Transport Type**: Streamable HTTP (pre-selected)
+3. Click "Add Server"
+4. You'll see "Authentication required" - click the **"Authenticate"** button
+5. Sign in to Vercel in your browser and authorize Cline
+6. Return to Cline - the server will automatically connect
+7. Vercel's tools (deploy, logs, projects) are now available to Cline!
+
+Your Vercel authentication persists across sessions, so you won't need to re-authenticate each time you use Cline.

docs/package-lock.json

@@ -5146,28 +5146,6 @@
 				"node": ">=6.0"
 			}
 		},
-		"node_modules/gray-matter/node_modules/argparse": {
-			"version": "1.0.10",
-			"resolved": "https://registry.npmjs.org/argparse/-/argparse-1.0.10.tgz",
-			"integrity": "sha512-o5Roy6tNG4SL/FOkCAN6RzjiakZS25RLYFrcMttJqbdd8BWrnA+fGz57iN5Pb06pvBGvl5gQ0B48dJlslXvoTg==",
-			"license": "MIT",
-			"dependencies": {
-				"sprintf-js": "~1.0.2"
-			}
-		},
-		"node_modules/gray-matter/node_modules/js-yaml": {
-			"version": "3.14.1",
-			"resolved": "https://registry.npmjs.org/js-yaml/-/js-yaml-3.14.1.tgz",
-			"integrity": "sha512-okMH7OXXJ7YrN9Ok3/SXrnu4iX9yOk+25nqX4imS2npuvTYDmo/QEZoqwZkYaIDk3jVvBOTOIEgEhaLOynBS9g==",
-			"license": "MIT",
-			"dependencies": {
-				"argparse": "^1.0.7",
-				"esprima": "^4.0.0"
-			},
-			"bin": {
-				"js-yaml": "bin/js-yaml.js"
-			}
-		},
 		"node_modules/has-bigints": {
 			"version": "1.1.0",
 			"resolved": "https://registry.npmjs.org/has-bigints/-/has-bigints-1.1.0.tgz",
@@ -6490,9 +6468,9 @@
 			"license": "MIT"
 		},
 		"node_modules/js-yaml": {
-			"version": "4.1.0",
-			"resolved": "https://registry.npmjs.org/js-yaml/-/js-yaml-4.1.0.tgz",
-			"integrity": "sha512-wpxZs9NoxZaJESJGIZTyDEaYpl0FKSA+FB9aJiyemKhMwkxQg63h4T1KJgUGHpTqPDNRcmmYLugrRjJlBtWvRA==",
+			"version": "4.1.1",
+			"resolved": "https://registry.npmjs.org/js-yaml/-/js-yaml-4.1.1.tgz",
+			"integrity": "sha512-qQKT4zQxXl8lLwBtHMWwaTcGfFOZviOJet3Oy/xmGk2gZH677CJM9EvtfdSkgWcATZhj/55JZ0rmy3myCT5lsA==",
 			"license": "MIT",
 			"dependencies": {
 				"argparse": "^2.0.1"
@@ -10235,12 +10213,6 @@
 				"url": "https://github.com/sponsors/wooorm"
 			}
 		},
-		"node_modules/sprintf-js": {
-			"version": "1.0.3",
-			"resolved": "https://registry.npmjs.org/sprintf-js/-/sprintf-js-1.0.3.tgz",
-			"integrity": "sha512-D9cPgkvLlV3t3IzL0D0YLvGA9Ahk4PcvVwUbN0dSGr1aP0Nrt4AEnTUbuGvquEC0mA64Gqt1fzirlRs5ibXx8g==",
-			"license": "BSD-3-Clause"
-		},
 		"node_modules/stack-utils": {
 			"version": "2.0.6",
 			"resolved": "https://registry.npmjs.org/stack-utils/-/stack-utils-2.0.6.tgz",
@@ -10609,9 +10581,9 @@
 			}
 		},
 		"node_modules/tar-fs": {
-			"version": "3.1.0",
-			"resolved": "https://registry.npmjs.org/tar-fs/-/tar-fs-3.1.0.tgz",
-			"integrity": "sha512-5Mty5y/sOF1YWj1J6GiBodjlDc05CUR8PKXrsnFAiSG0xA+GHeWLovaZPYUDXkH/1iKRf2+M5+OrRgzC7O9b7w==",
+			"version": "3.1.1",
+			"resolved": "https://registry.npmjs.org/tar-fs/-/tar-fs-3.1.1.tgz",
+			"integrity": "sha512-LZA0oaPOc2fVo82Txf3gw+AkEd38szODlptMYejQUhndHMLQ9M059uXR+AfS7DNo0NpINvSqDsvyaCrBVkptWg==",
 			"license": "MIT",
 			"dependencies": {
 				"pump": "^3.0.0",

docs/package.json

@@ -14,5 +14,9 @@
 	"description": "",
 	"dependencies": {
 		"mintlify": "^4.2.23"
+	},
+	"overrides": {
+		"tar-fs": "^3.1.1",
+		"js-yaml": "^4.1.1"
 	}
 }

docs/provider-config/cerebras.mdx

@@ -18,8 +18,7 @@ Cerebras delivers the world's fastest AI inference through their revolutionary w
 
 Cline supports the following Cerebras models:
 
--   `qwen-3-coder-480b-free` (Free tier) - High-performance coding model at no cost
--   `qwen-3-coder-480b` - Flagship 480B parameter coding model
+-   `zai-glm-4.6` - Intelligent general purpose model with 1,500 tokens/s
 -   `qwen-3-235b-a22b-instruct-2507` - Advanced instruction-following model
 -   `qwen-3-235b-a22b-thinking-2507` - Reasoning model with step-by-step thinking
 -   `llama-3.3-70b` - Meta's Llama 3.3 model optimized for speed

evals/package.json

@@ -1,47 +1,48 @@
 {
-  "name": "cline-evals",
-  "version": "0.1.0",
-  "description": "Evaluation scripts and tools for Cline",
-  "main": "cli/dist/index.js",
-  "scripts": {
-    "build:cli": "cd cli && tsc",
-    "start:cli": "cd cli && node dist/index.js",
-    "dev:cli": "cd cli && ts-node src/index.ts",
-    "diff-eval": "./diff-edits/run_and_open_dashboard.sh",
-    "test": "echo \"Error: no test specified\" && exit 1"
-  },
-  "keywords": [
-    "cline",
-    "evaluation",
-    "benchmark",
-    "diff-edits"
-  ],
-  "author": "",
-  "license": "MIT",
-  "dependencies": {
-    "axios": "^1.12.0",
-    "better-sqlite3": "^12.4.1",
-    "chalk": "5.6.2",
-    "dotenv": "^16.5.0",
-    "commander": "^9.4.1",
-    "execa": "^5.1.1",
-    "node-fetch": "^2.7.0",
-    "ora": "^5.4.1",
-    "sqlite": "^4.1.2",
-    "tiktoken": "^1.0.21",
-    "uuid": "^9.0.0",
-    "yargs": "^17.6.2"
-  },
-  "devDependencies": {
-    "@types/better-sqlite3": "^7.6.3",
-    "@types/node": "^18.11.18",
-    "@types/node-fetch": "^2.6.12",
-    "@types/uuid": "^9.0.0",
-    "@types/yargs": "^17.0.19",
-    "ts-node": "^10.9.1",
-    "typescript": "^4.9.4"
-  },
-  "overrides": {
-    "tar-fs": "^3.1.1"
-  }
+	"name": "cline-evals",
+	"version": "0.1.0",
+	"description": "Evaluation scripts and tools for Cline",
+	"main": "cli/dist/index.js",
+	"scripts": {
+		"build:cli": "cd cli && tsc",
+		"start:cli": "cd cli && node dist/index.js",
+		"dev:cli": "cd cli && ts-node src/index.ts",
+		"diff-eval": "./diff-edits/run_and_open_dashboard.sh",
+		"test": "echo \"Error: no test specified\" && exit 1"
+	},
+	"keywords": [
+		"cline",
+		"evaluation",
+		"benchmark",
+		"diff-edits"
+	],
+	"author": "",
+	"license": "MIT",
+	"dependencies": {
+		"axios": "^1.12.0",
+		"better-sqlite3": "^12.4.1",
+		"chalk": "5.6.2",
+		"dotenv": "^16.5.0",
+		"commander": "^9.4.1",
+		"execa": "^5.1.1",
+		"node-fetch": "^2.7.0",
+		"ora": "^5.4.1",
+		"sqlite": "^4.1.2",
+		"tiktoken": "^1.0.21",
+		"uuid": "^9.0.0",
+		"yargs": "^17.6.2"
+	},
+	"devDependencies": {
+		"@types/better-sqlite3": "^7.6.3",
+		"@types/node": "^18.11.18",
+		"@types/node-fetch": "^2.6.12",
+		"@types/uuid": "^9.0.0",
+		"@types/yargs": "^17.0.19",
+		"ts-node": "^10.9.1",
+		"typescript": "^4.9.4"
+	},
+	"overrides": {
+		"tar-fs": "^3.1.1",
+		"js-yaml": "^4.1.1"
+	}
 }

package.json

@@ -2,7 +2,7 @@
 	"name": "claude-dev",
 	"displayName": "Cline",
 	"description": "Autonomous coding agent right in your IDE, capable of creating/editing files, running commands, using the browser, and more with your permission every step of the way.",
-	"version": "3.36.0",
+	"version": "3.38.0",
 	"icon": "assets/icons/icon.png",
 	"engines": {
 		"vscode": "^1.84.0"
@@ -149,6 +149,12 @@
 				"category": "Cline",
 				"when": "cline.isDevMode"
 			},
+			{
+				"command": "cline.dev.expireMcpOAuthTokens",
+				"title": "Expire MCP OAuth Tokens (for testing)",
+				"category": "Cline",
+				"when": "cline.isDevMode"
+			},
 			{
 				"command": "cline.addToChat",
 				"title": "Add to Cline",
@@ -278,11 +284,11 @@
 			"commandPalette": [
 				{
 					"command": "cline.generateGitCommitMessage",
-					"when": "config.git.enabled && scmProvider == git && !cline.isGeneratingCommit"
+					"when": "config.git.enabled && !cline.isGeneratingCommit"
 				},
 				{
 					"command": "cline.abortGitCommitMessage",
-					"when": "config.git.enabled && scmProvider == git && cline.isGeneratingCommit"
+					"when": "config.git.enabled && cline.isGeneratingCommit"
 				}
 			]
 		},
@@ -304,6 +310,7 @@
 		"dev:cli:watch": "node scripts/dev-cli-watch.mjs",
 		"postcompile-standalone": "node scripts/package-standalone.mjs",
 		"postcompile-standalone-npm": "node scripts/package-standalone.mjs --target=npm",
+		"dev": "npm run protos && npm run watch",
 		"watch": "npm-run-all -p watch:*",
 		"watch:esbuild": "node esbuild.mjs --watch",
 		"watch:tsc": "tsc --noEmit --watch --project tsconfig.json",
@@ -330,7 +337,7 @@
 		"pretest": "npm run compile && npm run compile-tests && npm run compile-standalone && npm run lint",
 		"test": "npm-run-all test:unit test:integration",
 		"test:integration": "vscode-test",
-		"test:unit": "cross-env TS_NODE_PROJECT=./tsconfig.unit-test.json mocha",
+		"test:unit": "cross-env TS_NODE_PROJECT=./tsconfig.unit-test.json mocha # Use `UPDATE_SNAPSHOTS=true npm run test:unit` to rebuild prompt snapshots",
 		"test:coverage": "vscode-test --coverage",
 		"test:sca-server": "npx tsx watch scripts/test-standalone-core-api-server.ts",
 		"test:tp-orchestrator": "npx tsx scripts/testing-platform-orchestrator.ts",
@@ -410,7 +417,7 @@
 		"@bufbuild/protobuf": "^2.2.5",
 		"@cerebras/cerebras_cloud_sdk": "^1.35.0",
 		"@google-cloud/vertexai": "^1.9.3",
-		"@google/genai": "^1.11.0",
+		"@google/genai": "^1.30.0",
 		"@grpc/grpc-js": "^1.9.15",
 		"@grpc/reflection": "^1.0.4",
 		"@mistralai/mistralai": "^1.5.0",
@@ -435,8 +442,8 @@
 		"@opentelemetry/sdk-trace-node": "^1.30.1",
 		"@opentelemetry/semantic-conventions": "^1.37.0",
 		"@playwright/test": "^1.55.1",
-		"@sap-ai-sdk/ai-api": "^1.17.0",
-		"@sap-ai-sdk/orchestration": "^1.17.0",
+		"@sap-ai-sdk/ai-api": "^2.1.0",
+		"@sap-ai-sdk/orchestration": "^2.1.0",
 		"@sentry/browser": "^9.12.0",
 		"@streamparser/json": "^0.0.22",
 		"@tailwindcss/vite": "^4.1.14",
@@ -454,7 +461,6 @@
 		"exceljs": "^4.4.0",
 		"execa": "^9.5.2",
 		"fast-deep-equal": "^3.1.3",
-		"firebase": "^11.2.0",
 		"fzf": "^0.5.2",
 		"get-folder-size": "^5.0.0",
 		"globby": "^14.0.2",
@@ -464,8 +470,8 @@
 		"image-size": "^2.0.2",
 		"isbinaryfile": "^5.0.2",
 		"jschardet": "^3.1.4",
-		"jwt-decode": "^4.0.0",
 		"mammoth": "^1.11.0",
+		"nanoid": "^5.1.6",
 		"nice-grpc": "^2.1.12",
 		"node-machine-id": "^1.1.12",
 		"ollama": "^0.5.13",
@@ -496,7 +502,10 @@
 		"zod": "^3.24.2"
 	},
 	"overrides": {
-		"tar-fs": ">=3.1.1"
+		"tar-fs": ">=3.1.1",
+		"tar": "^7.5.2",
+		"vite": "^7.1.11",
+		"js-yaml": "^4.1.1"
 	},
 	"c8": {
 		"reporter": [

proto/cline/file.proto

@@ -49,6 +49,9 @@ service FileService {
   // Toggle a Windsurf rule (enable or disable)
   rpc toggleWindsurfRule(ToggleWindsurfRuleRequest) returns (ClineRulesToggles);
 
+  // Toggle an Agents rule (enable or disable)
+  rpc toggleAgentsRule(ToggleAgentsRuleRequest) returns (ClineRulesToggles);
+
   // Refreshes all rule toggles (Cline, External, and Workflows)
   rpc refreshRules(EmptyRequest) returns (RefreshedRules);
 
@@ -74,8 +77,9 @@ message RefreshedRules {
   ClineRulesToggles local_cline_rules_toggles = 2;
   ClineRulesToggles local_cursor_rules_toggles = 3;
   ClineRulesToggles local_windsurf_rules_toggles = 4;
-  ClineRulesToggles local_workflow_toggles = 5;
-  ClineRulesToggles global_workflow_toggles = 6;
+  ClineRulesToggles local_agents_rules_toggles = 5;
+  ClineRulesToggles local_workflow_toggles = 6;
+  ClineRulesToggles global_workflow_toggles = 7;
 }
 
 // Request to toggle a Windsurf rule
@@ -85,6 +89,13 @@ message ToggleWindsurfRuleRequest {
   bool enabled = 3; // Whether to enable or disable the rule
 }
 
+// Request to toggle an Agents rule
+message ToggleAgentsRuleRequest {
+  Metadata metadata = 1;
+  string rule_path = 2; // Path to the rule file
+  bool enabled = 3; // Whether to enable or disable the rule
+}
+
 // Request to convert a list of URIs to relative paths
 message RelativePathsRequest {
   Metadata metadata = 1;
@@ -156,10 +167,17 @@ message RuleFile {
   bool already_exists = 3; // For createRuleFile, indicates if file already existed
 }
 
+// Enum for rule scope (local, global, or remote)
+enum RuleScope {
+  LOCAL = 0;
+  GLOBAL = 1;
+  REMOTE = 2;
+}
+
 // Request to toggle a Cline rule
 message ToggleClineRuleRequest {
   Metadata metadata = 1;
-  bool is_global = 2; // Whether this is a global rule or workspace rule
+  RuleScope scope = 2; // Scope of the rule (local, global, or remote)
   string rule_path = 3; // Path to the rule file
   bool enabled = 4; // Whether to enable or disable the rule
 }
@@ -173,6 +191,7 @@ message ClineRulesToggles {
 message ToggleClineRules {
   ClineRulesToggles global_cline_rules_toggles = 1;
   ClineRulesToggles local_cline_rules_toggles = 2;
+  ClineRulesToggles remote_rules_toggles = 3;
 }
 
 // Request to toggle a Cursor rule
@@ -187,5 +206,5 @@ message ToggleWorkflowRequest {
   Metadata metadata = 1;
   string workflow_path = 2;
   bool enabled = 3;
-  bool is_global = 4;
+  RuleScope scope = 4; // Scope of the workflow (local, global, or remote)
 }

proto/cline/mcp.proto

@@ -18,6 +18,7 @@ service McpService {
   rpc toggleToolAutoApprove(ToggleToolAutoApproveRequest) returns (McpServers);
   rpc refreshMcpMarketplace(EmptyRequest) returns (McpMarketplaceCatalog);
   rpc openMcpSettings(EmptyRequest) returns (Empty);
+  rpc authenticateMcpServer(StringRequest) returns (Empty);
 
   // Subscribe to MCP marketplace catalog updates
   rpc subscribeToMcpMarketplaceCatalog(EmptyRequest) returns (stream McpMarketplaceCatalog);
@@ -43,6 +44,7 @@ message AddRemoteMcpServerRequest {
   Metadata metadata = 1;
   string server_name = 2;
   string server_url = 3;
+  optional string transport_type = 4;
 }
 
 message ToggleToolAutoApproveRequest {
@@ -91,6 +93,8 @@ message McpServer {
   repeated McpResourceTemplate resource_templates = 7;
   optional bool disabled = 8;
   optional int32 timeout = 9;
+  optional bool oauth_required = 10;
+  optional string oauth_auth_status = 11;
 }
 
 message McpServers {

proto/cline/models.proto

@@ -96,6 +96,8 @@ message OpenRouterModelInfo {
   optional ThinkingConfig thinking_config = 10;
   optional bool supports_global_endpoint = 11;
   repeated ModelTier tiers = 12;
+  optional string name = 13;
+  optional double temperature = 14;
 }
 
 // Shared response message for model information
@@ -312,7 +314,7 @@ message ApiConfiguration {
 // Request for updating API configuration (new - uses separate options and secrets)
 message UpdateApiConfigurationRequestNew {
   Metadata metadata = 1;
-  ApiConfiguration api_configuration = 2;
+  ApiConfiguration updates = 2;
 
   // Required field mask specifying which fields to update.
   // Field paths use dot notation with camelCase field names:
@@ -320,7 +322,7 @@ message UpdateApiConfigurationRequestNew {
   // - "options.openAiHeaders" (for options fields)
   // - "secrets.apiKey" (for secrets fields)
   // - "secrets.openRouterApiKey" (for secrets fields)
-  google.protobuf.FieldMask update_mask = 3;
+  repeated string update_mask = 3;
 }
 
 // Request for partially updating API configuration using FieldMask
@@ -421,6 +423,7 @@ enum ApiProvider {
   MINIMAX = 36;
   HICAP = 37;
   AIHUBMIX = 38;
+  NOUSRESEARCH = 39;
 }
 
 // Model info for OpenAI-compatible models
@@ -545,6 +548,7 @@ message ModelsApiConfiguration {
   optional string aihubmix_api_key = 82;
   optional string aihubmix_base_url = 83;
   optional string aihubmix_app_code = 84;
+  optional string nous_research_api_key = 85;
 
   // Plan mode configurations
   optional ApiProvider plan_mode_api_provider = 100;
@@ -584,6 +588,7 @@ message ModelsApiConfiguration {
   optional OpenRouterModelInfo plan_mode_hicap_model_info = 134;
   optional string plan_mode_aihubmix_model_id = 135;
   optional OpenAiCompatibleModelInfo plan_mode_aihubmix_model_info = 136;
+  optional string plan_mode_nous_research_model_id = 137;
 
   // Act mode configurations
   optional ApiProvider act_mode_api_provider = 200;
@@ -623,4 +628,5 @@ message ModelsApiConfiguration {
   optional OpenRouterModelInfo act_mode_hicap_model_info = 234;
   optional string act_mode_aihubmix_model_id = 235;
   optional OpenAiCompatibleModelInfo act_mode_aihubmix_model_info = 236;
+  optional string act_mode_nous_research_model_id = 237;
 }

proto/cline/state.proto

@@ -23,6 +23,7 @@ service StateService {
   rpc updateSettingsCli(UpdateSettingsRequestCli) returns (Empty);
   rpc updateTaskSettings(UpdateTaskSettingsRequest) returns (Empty);
   rpc updateTelemetrySetting(TelemetrySettingRequest) returns (Empty);
+  rpc captureOnboardingProgress(OnboardingProgressRequest) returns (Empty);
   rpc setWelcomeViewCompleted(BooleanRequest) returns (Empty);
   rpc updateInfoBannerVersion(Int64Request) returns (Empty);
   rpc updateModelBannerVersion(Int64Request) returns (Empty);
@@ -30,6 +31,7 @@ service StateService {
   rpc installClineCli(EmptyRequest) returns (Empty);
   rpc checkCliInstallation(EmptyRequest) returns (Boolean);
   rpc getProcessInfo(EmptyRequest) returns (ProcessInfo);
+  rpc flushPendingState(EmptyRequest) returns (Empty);
 }
 
 message AutoApprovalActions {
@@ -88,6 +90,7 @@ message Secrets {
   optional string oca_api_key = 37;
   optional string oca_refresh_token = 38;
   optional string hicap_api_key = 39;
+  optional string mcp_oauth_secrets = 40;
 }
 
 message Settings {
@@ -359,6 +362,7 @@ message UpdateSettingsRequest {
   optional int32 subagent_terminal_output_line_limit = 30;
   optional string cline_env = 31;
   optional bool native_tool_call_enabled = 32;
+  optional OnboardingModelGroup onboarding_models = 33;
 }
 
 message UpdateTerminalConnectionTimeoutRequest {
@@ -379,3 +383,24 @@ message ProcessInfo {
   optional string version = 2;
   optional int64 uptime_ms = 3;
 }
+
+message OnboardingProgressRequest {
+  int32 step = 1;
+  optional string action = 2;
+  optional bool completed = 3;
+  optional string model_selected = 4;
+}
+
+message OnboardingModelGroup {
+  repeated OnboardingModel models = 1;
+}
+
+message OnboardingModel {
+  string id = 1;
+  string name = 2;
+  int32 score = 3;
+  int32 latency = 4;
+  string badge = 5;
+  string group = 6;
+  OpenRouterModelInfo info = 7;
+}

proto/cline/task.proto

@@ -70,6 +70,7 @@ message TaskResponse {
   int32 tokens_out = 8;
   int32 cache_writes = 9;
   int32 cache_reads = 10;
+  string model_id = 11;
 }
 
 // Request for getting task history with filtering
@@ -99,6 +100,7 @@ message TaskItem {
   int32 tokens_out = 8;
   int32 cache_writes = 9;
   int32 cache_reads = 10;
+  string model_id = 11;
 }
 
 // Request for ask response operation

proto/cline/ui.proto

@@ -32,6 +32,7 @@ enum ClineAsk {
   CONDENSE = 13;
   REPORT_BUG = 14;
   SUMMARIZE_TASK = 15;
+  ACT_MODE_RESPOND = 16;
 }
 
 // Enum for ClineSay types
@@ -77,6 +78,7 @@ enum ClineSayToolType {
   LIST_CODE_DEFINITION_NAMES = 5;
   SEARCH_FILES = 6;
   WEB_FETCH = 7;
+  FILE_DELETED = 8;
 }
 
 // Enum for browser actions
@@ -183,6 +185,11 @@ message ClineApiReqInfo {
   ApiReqRetryStatus retry_status = 9;
 }
 
+message ClineModelInfo {
+  string provider_id = 1;
+  string model_id = 2;
+}
+
 // Main ClineMessage type
 message ClineMessage {
   int64 ts = 1;
@@ -209,6 +216,7 @@ message ClineMessage {
   ClineAskQuestion ask_question = 20;
   ClineAskNewTask ask_new_task = 21;
   ClineApiReqInfo api_req_info = 22;
+  ClineModelInfo model_info = 23;
 }
 
 // UiService provides methods for managing UI interactions

scripts/cli-providers.mjs

@@ -95,6 +95,7 @@ const ENABLED_PROVIDERS = [
 	"ollama", // Ollama local models
 	"cerebras", // Cerebras models
 	"oca", // Oracle Code Assist
+	"nousResearch", // NousResearch provider
 ]
 
 /**

scripts/download-ripgrep.mjs

@@ -11,7 +11,7 @@ import fs from "fs"
 import https from "https"
 import path from "path"
 import { pipeline } from "stream/promises"
-import tar from "tar"
+import * as tar from "tar"
 import { promisify } from "util"
 import { createGunzip } from "zlib"
 

src/api/providers/dify.ts

@@ -1,5 +1,5 @@
-import { Anthropic } from "@anthropic-ai/sdk"
 import { ModelInfo } from "@shared/api"
+import { ClineStorageMessage } from "@/shared/messages/content"
 import { fetch } from "@/shared/net"
 import { ApiHandler } from "../../core/api/index"
 import { ApiStream } from "../../core/api/transform/stream"
@@ -33,7 +33,7 @@ export class DifyHandler implements ApiHandler {
 		}
 	}
 
-	async *createMessage(systemPrompt: string, messages: Anthropic.Messages.MessageParam[]): ApiStream {
+	async *createMessage(systemPrompt: string, messages: ClineStorageMessage[]): ApiStream {
 		console.log("[DIFY DEBUG] createMessage called with:", {
 			systemPromptLength: systemPrompt?.length || 0,
 			messagesCount: messages?.length || 0,
@@ -255,7 +255,7 @@ export class DifyHandler implements ApiHandler {
 		}
 	}
 
-	private convertMessagesToQuery(systemPrompt: string, messages: Anthropic.Messages.MessageParam[]): string {
+	private convertMessagesToQuery(systemPrompt: string, messages: ClineStorageMessage[]): 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()

src/config.ts

@@ -9,14 +9,6 @@ export interface EnvironmentConfig {
 	appBaseUrl: string
 	apiBaseUrl: string
 	mcpBaseUrl: string
-	firebase: {
-		apiKey: string
-		authDomain: string
-		projectId: string
-		storageBucket?: string
-		messagingSenderId?: string
-		appId?: string
-	}
 }
 
 class ClineEndpoint {
@@ -28,8 +20,8 @@ class ClineEndpoint {
 	private environment: Environment = Environment.production
 
 	private constructor() {
-		// Set environment at module load
-		const _env = process?.env?.CLINE_ENVIRONMENT
+		// Set environment at module load. Use override if provided.
+		const _env = process?.env?.CLINE_ENVIRONMENT_OVERRIDE || process?.env?.CLINE_ENVIRONMENT
 		if (_env && Object.values(Environment).includes(_env as Environment)) {
 			this.environment = _env as Environment
 			return
@@ -62,15 +54,7 @@ class ClineEndpoint {
 					environment: Environment.staging,
 					appBaseUrl: "https://staging-app.cline.bot",
 					apiBaseUrl: "https://core-api.staging.int.cline.bot",
-					mcpBaseUrl: "https://api.cline.bot/v1/mcp",
-					firebase: {
-						apiKey: "AIzaSyASSwkwX1kSO8vddjZkE5N19QU9cVQ0CIk",
-						authDomain: "cline-staging.firebaseapp.com",
-						projectId: "cline-staging",
-						storageBucket: "cline-staging.firebasestorage.app",
-						messagingSenderId: "853479478430",
-						appId: "1:853479478430:web:2de0dba1c63c3262d4578f",
-					},
+					mcpBaseUrl: "https://core-api.staging.int.cline.bot/v1/mcp",
 				}
 			case Environment.local:
 				return {
@@ -78,11 +62,6 @@ class ClineEndpoint {
 					appBaseUrl: "http://localhost:3000",
 					apiBaseUrl: "http://localhost:7777",
 					mcpBaseUrl: "https://api.cline.bot/v1/mcp",
-					firebase: {
-						apiKey: "AIzaSyD8wtkd1I-EICuAg6xgAQpRdwYTvwxZG2w",
-						authDomain: "cline-preview.firebaseapp.com",
-						projectId: "cline-preview",
-					},
 				}
 			default:
 				return {
@@ -90,14 +69,6 @@ class ClineEndpoint {
 					appBaseUrl: "https://app.cline.bot",
 					apiBaseUrl: "https://api.cline.bot",
 					mcpBaseUrl: "https://api.cline.bot/v1/mcp",
-					firebase: {
-						apiKey: "AIzaSyC5rx59Xt8UgwdU3PCfzUF7vCwmp9-K2vk",
-						authDomain: "cline-prod.firebaseapp.com",
-						projectId: "cline-prod",
-						storageBucket: "cline-prod.firebasestorage.app",
-						messagingSenderId: "941048379330",
-						appId: "1:941048379330:web:45058eedeefc5cdfcc485b",
-					},
 				}
 		}
 	}

src/core/api/index.ts

@@ -25,6 +25,7 @@ import { MinimaxHandler } from "./providers/minimax"
 import { MistralHandler } from "./providers/mistral"
 import { MoonshotHandler } from "./providers/moonshot"
 import { NebiusHandler } from "./providers/nebius"
+import { NousResearchHandler } from "./providers/nousresearch"
 import { OcaHandler } from "./providers/oca"
 import { OllamaHandler } from "./providers/ollama"
 import { OpenAiHandler } from "./providers/openai"
@@ -416,6 +417,12 @@ function createHandlerForProvider(
 				hicapApiKey: options.hicapApiKey,
 				hicapModelId: mode === "plan" ? options.planModeHicapModelId : options.actModeHicapModelId,
 			})
+		case "nousResearch":
+			return new NousResearchHandler({
+				onRetryAttempt: options.onRetryAttempt,
+				nousResearchApiKey: options.nousResearchApiKey,
+				apiModelId: mode === "plan" ? options.planModeNousResearchModelId : options.actModeNousResearchModelId,
+			})
 		default:
 			return new AnthropicHandler({
 				onRetryAttempt: options.onRetryAttempt,

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

@@ -1,8 +1,8 @@
-import { Anthropic } from "@anthropic-ai/sdk"
 import { afterEach, beforeEach, describe, it } from "mocha"
 import sinon from "sinon"
 import "should"
 import { ClaudeCodeHandler } from "@core/api/providers/claude-code"
+import { ClineStorageMessage } from "@/shared/messages/content"
 
 describe("ClaudeCodeHandler", () => {
 	let handler: ClaudeCodeHandler
@@ -71,7 +71,7 @@ describe("ClaudeCodeHandler", () => {
 			runClaudeCodeStub.returns(mockGenerator() as any)
 
 			const systemPrompt = "You are a helpful assistant."
-			const messages: Anthropic.Messages.MessageParam[] = [{ role: "user", content: "Hello" }]
+			const messages: ClineStorageMessage[] = [{ role: "user", content: "Hello" }]
 
 			const usageData: any[] = []
 
@@ -140,7 +140,7 @@ describe("ClaudeCodeHandler", () => {
 			runClaudeCodeStub.returns(mockGenerator() as any)
 
 			const systemPrompt = "You are a helpful assistant."
-			const messages: Anthropic.Messages.MessageParam[] = [{ role: "user", content: "Hello" }]
+			const messages: ClineStorageMessage[] = [{ role: "user", content: "Hello" }]
 
 			const usageData: any[] = []
 
@@ -199,7 +199,7 @@ describe("ClaudeCodeHandler", () => {
 			runClaudeCodeStub.returns(mockGenerator() as any)
 
 			const systemPrompt = "You are a helpful assistant."
-			const messages: Anthropic.Messages.MessageParam[] = [{ role: "user", content: "Hello" }]
+			const messages: ClineStorageMessage[] = [{ role: "user", content: "Hello" }]
 
 			const usageData: any[] = []
 

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

@@ -1,8 +1,8 @@
-import Anthropic from "@anthropic-ai/sdk"
 import { LiteLlmHandler, type LiteLlmModelInfoResponse } from "@core/api/providers/litellm"
 import { convertToOpenAiMessages } from "@core/api/transform/openai-format"
 import { expect } from "chai"
 import sinon from "sinon"
+import { ClineStorageMessage } from "@/shared/messages/content"
 import { mockFetchForTesting } from "@/shared/net"
 
 const fakeClient = {
@@ -109,7 +109,7 @@ describe("LiteLlmHandler", () => {
 
 			it("sends the system prompt and messages with the openai format", async () => {
 				const systemPrompt = "Test System Prompt"
-				const messages: Anthropic.Messages.MessageParam[] = [
+				const messages: ClineStorageMessage[] = [
 					{
 						role: "user",
 						content: "first message",
@@ -161,7 +161,7 @@ describe("LiteLlmHandler", () => {
 
 			it("inserts the cache control in the system prompt and the last two user messages", async () => {
 				const systemPrompt = "Test System Prompt"
-				const messages: Anthropic.Messages.MessageParam[] = [
+				const messages: ClineStorageMessage[] = [
 					{
 						role: "user",
 						content: "first message",

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

@@ -1,9 +1,9 @@
 import { afterEach, before, beforeEach, describe, it } from "mocha"
 import "should"
-import { Anthropic } from "@anthropic-ai/sdk"
 import { ApiHandlerOptions } from "@shared/api"
 import axios from "axios"
 import sinon from "sinon"
+import { ClineStorageMessage } from "@/shared/messages/content"
 import { OllamaHandler } from "../ollama"
 
 describe("OllamaHandler", () => {
@@ -59,7 +59,7 @@ describe("OllamaHandler", () => {
 			} as any)
 
 			const systemPrompt = "You are a helpful assistant."
-			const messages: Anthropic.Messages.MessageParam[] = [{ role: "user", content: "Hello" }]
+			const messages: ClineStorageMessage[] = [{ role: "user", content: "Hello" }]
 
 			const result = []
 			const usageInfo = []
@@ -114,7 +114,7 @@ describe("OllamaHandler", () => {
 			}
 
 			const systemPrompt = "You are a helpful assistant."
-			const messages: Anthropic.Messages.MessageParam[] = [{ role: "user", content: "Hello" }]
+			const messages: ClineStorageMessage[] = [{ role: "user", content: "Hello" }]
 
 			// Start the request and catch the error
 			let errorMessage = ""
@@ -158,7 +158,7 @@ describe("OllamaHandler", () => {
 			} as any)
 
 			const systemPrompt = "You are a helpful assistant."
-			const messages: Anthropic.Messages.MessageParam[] = [{ role: "user", content: "Hello" }]
+			const messages: ClineStorageMessage[] = [{ role: "user", content: "Hello" }]
 
 			const result = []
 
@@ -204,7 +204,7 @@ describe("OllamaHandler", () => {
 			}
 
 			const systemPrompt = "You are a helpful assistant."
-			const messages: Anthropic.Messages.MessageParam[] = [{ role: "user", content: "Hello" }]
+			const messages: ClineStorageMessage[] = [{ role: "user", content: "Hello" }]
 
 			const result = []
 

src/core/api/providers/aihubmix.ts

@@ -27,7 +27,7 @@ export class AIhubmixHandler implements ApiHandler {
 		const { baseURL, appCode, ...rest } = options
 		this.options = {
 			baseURL: baseURL ?? "https://aihubmix.com",
-			appCode: appCode ?? "KUWF9311", // 应用代码，享受折扣
+			appCode: appCode ?? "KUWF9311",
 			...rest,
 		}
 	}
@@ -81,7 +81,6 @@ export class AIhubmixHandler implements ApiHandler {
 				this.geminiClient = new GoogleGenAI({
 					apiKey: this.options.apiKey,
 					httpOptions: {
-						// AIhubmix Gemini 兼容网关，按 Google GenAI 路径规范
 						baseUrl: `${this.options.baseURL}/gemini`,
 						headers: {
 							// @ts-expect-error
@@ -97,9 +96,6 @@ export class AIhubmixHandler implements ApiHandler {
 		return this.geminiClient
 	}
 
-	/**
-	 * 根据模型名称路由到对应的客户端
-	 */
 	private routeModel(modelName: string): "anthropic" | "openai" | "gemini" | "openai-response" {
 		const id = modelName || ""
 		if (id.startsWith("claude")) {
@@ -114,9 +110,6 @@ export class AIhubmixHandler implements ApiHandler {
 		return "openai"
 	}
 
-	/**
-	 * 修复空工具时的 tool_choice 问题
-	 */
 	private fixToolChoice(requestBody: any): any {
 		if (requestBody.tools?.length === 0 && requestBody.tool_choice) {
 			delete requestBody.tool_choice
@@ -203,25 +196,21 @@ export class AIhubmixHandler implements ApiHandler {
 		const client = this.ensureOpenaiClient()
 		const modelId = this.options.modelId || "gpt-4o-mini"
 
-		// 将 Anthropic 风格消息转换为 Responses API 的 input 结构
 		const input = (messages || []).map((m: any) => {
 			const role = m.role || "user"
 			const contentArray = Array.isArray(m.content) ? m.content : [{ type: "text", text: m.content }]
 			const content = contentArray
 				.filter((c: any) => c != null)
 				.map((c: any) => {
-					// 图片
 					if (c.type === "image" || c.type === "input_image" || c.type === "image_url") {
 						return { type: "input_image", image_url: c.image_url || c.url || c.source?.url }
 					}
-					// 文本（用户 -> input_text，助手 -> output_text）
 					const text = c.text ?? (typeof c === "string" ? c : "")
 					return { type: role === "assistant" ? "output_text" : "input_text", text }
 				})
 			return { role, content }
 		})
 
-		// 使用 Responses 流式 API，以事件驱动产出
 		const stream = await (client as any).responses.stream({
 			model: modelId,
 			instructions: systemPrompt,
@@ -261,7 +250,6 @@ export class AIhubmixHandler implements ApiHandler {
 			stream: true,
 		}
 
-		// 修复空工具问题
 		const fixedRequestBody = this.fixToolChoice(requestBody)
 
 		const stream = await client.chat.completions.create(fixedRequestBody)

src/core/api/providers/anthropic.ts

@@ -1,11 +1,13 @@
 import { Anthropic } from "@anthropic-ai/sdk"
-import { Tool as AnthropicTool, MessageParam } from "@anthropic-ai/sdk/resources/index"
+import { Tool as AnthropicTool } from "@anthropic-ai/sdk/resources/index"
 import { Stream as AnthropicStream } from "@anthropic-ai/sdk/streaming"
 import { AnthropicModelId, anthropicDefaultModelId, anthropicModels, CLAUDE_SONNET_1M_SUFFIX, ModelInfo } from "@shared/api"
+import { ClineStorageMessage } from "@/shared/messages/content"
 import { fetch } from "@/shared/net"
 import { ClineTool } from "@/shared/tools"
 import { ApiHandler, CommonApiHandlerOptions } from "../index"
 import { withRetry } from "../retry"
+import { sanitizeAnthropicMessages } from "../transform/anthropic-format"
 import { ApiStream } from "../transform/stream"
 
 interface AnthropicHandlerOptions extends CommonApiHandlerOptions {
@@ -42,7 +44,7 @@ export class AnthropicHandler implements ApiHandler {
 	}
 
 	@withRetry()
-	async *createMessage(systemPrompt: string, messages: Anthropic.Messages.MessageParam[], tools?: ClineTool[]): ApiStream {
+	async *createMessage(systemPrompt: string, messages: ClineStorageMessage[], tools?: ClineTool[]): ApiStream {
 		const client = this.ensureClient()
 
 		const model = this.getModel()
@@ -63,6 +65,7 @@ export class AnthropicHandler implements ApiHandler {
 		switch (modelId) {
 			// 'latest' alias does not support cache_control
 			case "claude-haiku-4-5-20251001":
+			case "claude-sonnet-4-5-20250929:1m":
 			case "claude-sonnet-4-5-20250929":
 			case "claude-sonnet-4-20250514":
 			case "claude-3-7-sonnet-20250219":
@@ -84,35 +87,7 @@ export class AnthropicHandler implements ApiHandler {
 				const lastUserMsgIndex = userMsgIndices[userMsgIndices.length - 1] ?? -1
 				const secondLastMsgUserIndex = userMsgIndices[userMsgIndices.length - 2] ?? -1
 
-				const anthropicMessages: Array<MessageParam> = messages.map((message, index) => {
-					if (index === lastUserMsgIndex || index === secondLastMsgUserIndex) {
-						return {
-							...message,
-							content:
-								typeof message.content === "string"
-									? [
-											{
-												type: "text",
-												text: message.content,
-												cache_control: {
-													type: "ephemeral",
-												},
-											},
-										]
-									: message.content.map((content, contentIndex) =>
-											contentIndex === message.content.length - 1
-												? {
-														...content,
-														cache_control: {
-															type: "ephemeral",
-														},
-													}
-												: content,
-										),
-						}
-					}
-					return message
-				})
+				const anthropicMessages = sanitizeAnthropicMessages(messages, lastUserMsgIndex, secondLastMsgUserIndex)
 
 				stream = await client.messages.create(
 					{
@@ -161,7 +136,7 @@ export class AnthropicHandler implements ApiHandler {
 					max_tokens: model.info.maxTokens || 8192,
 					temperature: 0,
 					system: [{ text: systemPrompt, type: "text" }],
-					messages,
+					messages: sanitizeAnthropicMessages(messages),
 					// tools,
 					// tool_choice: { type: "auto" },
 					stream: true,
@@ -170,20 +145,21 @@ export class AnthropicHandler implements ApiHandler {
 			}
 		}
 
-		let thinkingDeltaAccumulator = ""
 		const lastStartedToolCall = { id: "", name: "", arguments: "" }
 
 		for await (const chunk of stream) {
 			switch (chunk?.type) {
 				case "message_start":
-					// tells us cache reads/writes/input/output
-					const usage = chunk.message.usage
-					yield {
-						type: "usage",
-						inputTokens: usage.input_tokens || 0,
-						outputTokens: usage.output_tokens || 0,
-						cacheWriteTokens: usage.cache_creation_input_tokens || undefined,
-						cacheReadTokens: usage.cache_read_input_tokens || undefined,
+					{
+						// tells us cache reads/writes/input/output
+						const usage = chunk.message.usage
+						yield {
+							type: "usage",
+							inputTokens: usage.input_tokens || 0,
+							outputTokens: usage.output_tokens || 0,
+							cacheWriteTokens: usage.cache_creation_input_tokens || undefined,
+							cacheReadTokens: usage.cache_read_input_tokens || undefined,
+						}
 					}
 					break
 				case "message_delta":
@@ -204,15 +180,7 @@ export class AnthropicHandler implements ApiHandler {
 							yield {
 								type: "reasoning",
 								reasoning: chunk.content_block.thinking || "",
-							}
-							const thinking = chunk.content_block.thinking
-							const signature = chunk.content_block.signature
-							if (thinking && signature) {
-								yield {
-									type: "ant_thinking",
-									thinking,
-									signature,
-								}
+								signature: chunk.content_block.signature,
 							}
 							break
 						case "redacted_thinking":
@@ -220,10 +188,7 @@ export class AnthropicHandler implements ApiHandler {
 							yield {
 								type: "reasoning",
 								reasoning: "[Redacted thinking block]",
-							}
-							yield {
-								type: "ant_redacted_thinking",
-								data: chunk.content_block.data,
+								redacted_data: chunk.content_block.data,
 							}
 							break
 						case "tool_use":
@@ -252,20 +217,19 @@ export class AnthropicHandler implements ApiHandler {
 				case "content_block_delta":
 					switch (chunk.delta.type) {
 						case "thinking_delta":
-							// 'reasoning' type just displays in the UI, but ant_thinking will be used to send the thinking traces back to the API
+							// 'reasoning' type just displays in the UI, but reasoning with signature will be used to send the thinking traces back to the API
 							yield {
 								type: "reasoning",
 								reasoning: chunk.delta.thinking,
 							}
-							thinkingDeltaAccumulator += chunk.delta.thinking
 							break
 						case "signature_delta":
 							// It's used when sending the thinking block back to the API
 							// API expects this in completed form, not as array of deltas
-							if (thinkingDeltaAccumulator && chunk.delta.signature) {
+							if (chunk.delta.signature) {
 								yield {
-									type: "ant_thinking",
-									thinking: thinkingDeltaAccumulator,
+									type: "reasoning",
+									reasoning: "", // reasoning text is already sent via thinking_delta
 									signature: chunk.delta.signature,
 								}
 							}

src/core/api/providers/asksage.ts

@@ -1,5 +1,5 @@
-import { Anthropic } from "@anthropic-ai/sdk"
 import { AskSageModelId, askSageDefaultModelId, askSageDefaultURL, askSageModels, ModelInfo } from "@shared/api"
+import { ClineStorageMessage } from "@/shared/messages/content"
 import { fetch } from "@/shared/net"
 import { ApiHandler, CommonApiHandlerOptions } from ".."
 import { withRetry } from "../retry"
@@ -47,7 +47,7 @@ export class AskSageHandler implements ApiHandler {
 	}
 
 	@withRetry()
-	async *createMessage(systemPrompt: string, messages: Anthropic.Messages.MessageParam[]): ApiStream {
+	async *createMessage(systemPrompt: string, messages: ClineStorageMessage[]): ApiStream {
 		try {
 			const model = this.getModel()
 

src/core/api/providers/baseten.ts

@@ -1,7 +1,7 @@
-import { Anthropic } from "@anthropic-ai/sdk"
 import { BasetenModelId, basetenDefaultModelId, basetenModels, ModelInfo } from "@shared/api"
 import { calculateApiCostOpenAI } from "@utils/cost"
 import OpenAI from "openai"
+import { ClineStorageMessage } from "@/shared/messages/content"
 import { fetch } from "@/shared/net"
 import { ApiHandler, CommonApiHandlerOptions } from "../"
 import { withRetry } from "../retry"
@@ -98,7 +98,7 @@ export class BasetenHandler implements ApiHandler {
 	}
 
 	@withRetry()
-	async *createMessage(systemPrompt: string, messages: Anthropic.Messages.MessageParam[]): ApiStream {
+	async *createMessage(systemPrompt: string, messages: ClineStorageMessage[]): ApiStream {
 		const client = this.ensureClient()
 		const model = this.getModel()
 		const maxTokens = this.getOptimalMaxTokens(model)

src/core/api/providers/bedrock.ts

@@ -1,4 +1,3 @@
-import { Anthropic } from "@anthropic-ai/sdk"
 // Import proper AWS SDK types
 import type { ContentBlock, Message } from "@aws-sdk/client-bedrock-runtime"
 import {
@@ -12,6 +11,7 @@ import { fromNodeProviderChain } from "@aws-sdk/credential-providers"
 import { BedrockModelId, bedrockDefaultModelId, bedrockModels, CLAUDE_SONNET_1M_SUFFIX, ModelInfo } from "@shared/api"
 import { calculateApiCostOpenAI, calculateApiCostQwen } from "@utils/cost"
 import { ExtensionRegistryInfo } from "@/registry"
+import { ClineStorageMessage } from "@/shared/messages/content"
 import { ApiHandler, CommonApiHandlerOptions } from "../"
 import { withRetry } from "../retry"
 import { convertToR1Format } from "../transform/r1-format"
@@ -121,7 +121,7 @@ export class AwsBedrockHandler implements ApiHandler {
 	}
 
 	@withRetry({ maxRetries: 4 })
-	async *createMessage(systemPrompt: string, messages: Anthropic.Messages.MessageParam[]): ApiStream {
+	async *createMessage(systemPrompt: string, messages: ClineStorageMessage[]): ApiStream {
 		// cross region inference requires prefixing the model id with the region
 		const rawModelId = await this.getModelId()
 
@@ -342,7 +342,7 @@ export class AwsBedrockHandler implements ApiHandler {
 	 */
 	private async *createDeepseekMessage(
 		systemPrompt: string,
-		messages: Anthropic.Messages.MessageParam[],
+		messages: ClineStorageMessage[],
 		modelId: string,
 		model: { id: string; info: ModelInfo },
 	): ApiStream {
@@ -480,7 +480,7 @@ export class AwsBedrockHandler implements ApiHandler {
 	 * First uses convertToR1Format to merge consecutive messages with the same role,
 	 * then converts to the string format that DeepSeek R1 expects
 	 */
-	private formatDeepseekR1Prompt(systemPrompt: string, messages: Anthropic.Messages.MessageParam[]): string {
+	private formatDeepseekR1Prompt(systemPrompt: string, messages: ClineStorageMessage[]): string {
 		// First use convertToR1Format to merge consecutive messages with the same role
 		const r1Messages = convertToR1Format([{ role: "user", content: systemPrompt }, ...messages])
 
@@ -513,7 +513,7 @@ export class AwsBedrockHandler implements ApiHandler {
 	 * Estimates token count based on text length (approximate)
 	 * Note: This is a rough estimation, as the actual token count depends on the tokenizer
 	 */
-	private estimateInputTokens(systemPrompt: string, messages: Anthropic.Messages.MessageParam[]): number {
+	private estimateInputTokens(systemPrompt: string, messages: ClineStorageMessage[]): number {
 		// For Deepseek R1, we estimate the token count of the formatted prompt
 		// The formatted prompt includes special tokens and consistent formatting
 		const formattedPrompt = this.formatDeepseekR1Prompt(systemPrompt, messages)
@@ -680,11 +680,7 @@ export class AwsBedrockHandler implements ApiHandler {
 				}
 			}
 		} catch (error) {
-			console.error("Error processing Converse API response:", error)
-			yield {
-				type: "text",
-				text: `[ERROR] Failed to process response: ${error instanceof Error ? error.message : String(error)}`,
-			}
+			throw error
 		}
 	}
 
@@ -703,9 +699,20 @@ export class AwsBedrockHandler implements ApiHandler {
 				text: `[ERROR] Model stream error: ${chunk.modelStreamErrorException.message}`,
 			}
 		} else if (chunk.validationException) {
+			// Check if this is a context window error - if so, throw it
+			// so the retry mechanism can handle truncation
+			const message = chunk.validationException.message || ""
+			const isContextError = /input.*too long|context.*exceed|maximum.*token|input length.*max.*tokens/i.test(message)
+
+			if (isContextError) {
+				// Throw as exception so context management can handle it
+				throw chunk.validationException
+			}
+
+			// Otherwise yield as error text
 			yield {
 				type: "text",
-				text: `[ERROR] Validation error: ${chunk.validationException.message}`,
+				text: `[ERROR] Validation error: ${message}`,
 			}
 		} else if (chunk.throttlingException) {
 			yield {
@@ -779,7 +786,7 @@ export class AwsBedrockHandler implements ApiHandler {
 	 */
 	private async *createAnthropicMessage(
 		systemPrompt: string,
-		messages: Anthropic.Messages.MessageParam[],
+		messages: ClineStorageMessage[],
 		modelId: string,
 		model: { id: string; info: ModelInfo },
 		enable1mContextWindow: boolean,
@@ -835,7 +842,7 @@ export class AwsBedrockHandler implements ApiHandler {
 	 * Formats messages for models using the Converse API specification
 	 * Used by both Anthropic and Nova models to avoid code duplication
 	 */
-	private formatMessagesForConverseAPI(messages: Anthropic.Messages.MessageParam[]): Message[] {
+	private formatMessagesForConverseAPI(messages: ClineStorageMessage[]): Message[] {
 		return messages.map((message) => {
 			// Determine role (user or assistant)
 			const role = message.role === "user" ? ConversationRole.USER : ConversationRole.ASSISTANT
@@ -968,7 +975,7 @@ export class AwsBedrockHandler implements ApiHandler {
 	 */
 	private async *createNovaMessage(
 		systemPrompt: string,
-		messages: Anthropic.Messages.MessageParam[],
+		messages: ClineStorageMessage[],
 		modelId: string,
 		model: { id: string; info: ModelInfo },
 	): ApiStream {
@@ -1008,7 +1015,7 @@ export class AwsBedrockHandler implements ApiHandler {
 	 */
 	private async *createOpenAIMessage(
 		systemPrompt: string,
-		messages: Anthropic.Messages.MessageParam[],
+		messages: ClineStorageMessage[],
 		modelId: string,
 		model: { id: string; info: ModelInfo },
 	): ApiStream {
@@ -1143,7 +1150,7 @@ export class AwsBedrockHandler implements ApiHandler {
 	 */
 	private async *createQwenMessage(
 		systemPrompt: string,
-		messages: Anthropic.Messages.MessageParam[],
+		messages: ClineStorageMessage[],
 		modelId: string,
 		model: { id: string; info: ModelInfo },
 	): ApiStream {

src/core/api/providers/cerebras.ts

@@ -1,6 +1,6 @@
-import { Anthropic } from "@anthropic-ai/sdk"
 import Cerebras from "@cerebras/cerebras_cloud_sdk"
 import { CerebrasModelId, cerebrasDefaultModelId, cerebrasModels, ModelInfo } from "@shared/api"
+import { ClineStorageMessage } from "@/shared/messages/content"
 import { fetch } from "@/shared/net"
 import { ApiHandler, CommonApiHandlerOptions } from "../index"
 import { withRetry } from "../retry"
@@ -46,7 +46,7 @@ export class CerebrasHandler implements ApiHandler {
 		baseDelay: 5000, // Start with 5 second delay
 		maxDelay: 60000, // Allow up to 60 second delays to respect rate limits
 	})
-	async *createMessage(systemPrompt: string, messages: Anthropic.Messages.MessageParam[]): ApiStream {
+	async *createMessage(systemPrompt: string, messages: ClineStorageMessage[]): ApiStream {
 		const client = this.ensureClient()
 
 		// Convert Anthropic messages to Cerebras format

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

@@ -1,7 +1,7 @@
-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 { ClineStorageMessage } from "@/shared/messages/content"
 import { type ApiHandler, CommonApiHandlerOptions } from ".."
 import { withRetry } from "../retry"
 import { type ApiStream, ApiStreamUsageChunk } from "../transform/stream"
@@ -24,7 +24,7 @@ export class ClaudeCodeHandler implements ApiHandler {
 		baseDelay: 2000,
 		maxDelay: 15000,
 	})
-	async *createMessage(systemPrompt: string, messages: Anthropic.Messages.MessageParam[]): ApiStream {
+	async *createMessage(systemPrompt: string, messages: ClineStorageMessage[]): ApiStream {
 		// Filter out image blocks since Claude Code doesn't support them
 		const filteredMessages = filterMessagesForClaudeCode(messages)
 
@@ -113,7 +113,18 @@ export class ClaudeCodeHandler implements ApiHandler {
 							}
 							break
 						case "tool_use":
-							console.error(`tool_use is not supported yet. Received: ${JSON.stringify(content)}`)
+							// Yield tool_use blocks to the streaming pipeline for proper tool execution
+							yield {
+								type: "tool_calls",
+								tool_call: {
+									call_id: content.id,
+									function: {
+										id: content.id,
+										name: content.name,
+										arguments: content.input,
+									},
+								},
+							}
 							break
 					}
 				}

src/core/api/providers/cline.ts

@@ -1,4 +1,3 @@
-import { Anthropic } from "@anthropic-ai/sdk"
 import { ModelInfo, openRouterDefaultModelId, openRouterDefaultModelInfo } from "@shared/api"
 import { shouldSkipReasoningForModel } from "@utils/model-utils"
 import axios from "axios"
@@ -9,6 +8,7 @@ import { ClineAccountService } from "@/services/account/ClineAccountService"
 import { AuthService } from "@/services/auth/AuthService"
 import { buildClineExtraHeaders } from "@/services/EnvUtils"
 import { CLINE_ACCOUNT_AUTH_ERROR_MESSAGE } from "@/shared/ClineAccount"
+import { ClineStorageMessage } from "@/shared/messages/content"
 import { fetch, getAxiosSettings } from "@/shared/net"
 import { ApiHandler, CommonApiHandlerOptions } from "../"
 import { withRetry } from "../retry"
@@ -96,7 +96,7 @@ export class ClineHandler implements ApiHandler {
 	}
 
 	@withRetry()
-	async *createMessage(systemPrompt: string, messages: Anthropic.Messages.MessageParam[], tools?: OpenAITool[]): ApiStream {
+	async *createMessage(systemPrompt: string, messages: ClineStorageMessage[], tools?: OpenAITool[]): ApiStream {
 		try {
 			const client = await this.ensureClient()
 
@@ -154,10 +154,12 @@ export class ClineHandler implements ApiHandler {
 						type: "text",
 						text: delta.content,
 					}
+					continue
 				}
 
 				if (delta?.tool_calls) {
 					yield* toolCallProcessor.processToolCallDeltas(delta.tool_calls)
+					continue
 				}
 
 				// Reasoning tokens are returned separately from the content
@@ -165,9 +167,9 @@ export class ClineHandler implements ApiHandler {
 				if ("reasoning" in delta && delta.reasoning && !shouldSkipReasoningForModel(this.options.openRouterModelId)) {
 					yield {
 						type: "reasoning",
-						// @ts-ignore-next-line
-						reasoning: delta.reasoning,
+						reasoning: typeof delta.reasoning === "string" ? delta.reasoning : JSON.stringify(delta.reasoning),
 					}
+					continue
 				}
 
 				/* 
@@ -185,16 +187,18 @@ export class ClineHandler implements ApiHandler {
 					!shouldSkipReasoningForModel(this.options.openRouterModelId)
 				) {
 					yield {
-						type: "reasoning_details",
-						reasoning_details: delta.reasoning_details,
+						type: "reasoning",
+						reasoning: "",
+						details: delta.reasoning_details,
 					}
+					continue
 				}
 
 				if (!didOutputUsage && chunk.usage) {
 					// @ts-ignore-next-line
 					let totalCost = (chunk.usage.cost || 0) + (chunk.usage.cost_details?.upstream_inference_cost || 0)
 
-					if (this.getModel().id === "x-ai/grok-code-fast-1") {
+					if (this.getModel().id === "x-ai/grok-code-fast-1" || this.getModel().id === "minimax/minimax-m2") {
 						totalCost = 0
 					}
 

src/core/api/providers/deepseek.ts

@@ -1,8 +1,8 @@
-import { Anthropic } from "@anthropic-ai/sdk"
 import { DeepSeekModelId, deepSeekDefaultModelId, deepSeekModels, ModelInfo } from "@shared/api"
 import { calculateApiCostOpenAI } from "@utils/cost"
 import OpenAI from "openai"
 import type { ChatCompletionTool as OpenAITool } from "openai/resources/chat/completions"
+import { ClineStorageMessage } from "@/shared/messages/content"
 import { fetch } from "@/shared/net"
 import { ApiHandler, CommonApiHandlerOptions } from "../"
 import { withRetry } from "../retry"
@@ -75,7 +75,7 @@ export class DeepSeekHandler implements ApiHandler {
 	}
 
 	@withRetry()
-	async *createMessage(systemPrompt: string, messages: Anthropic.Messages.MessageParam[], tools?: OpenAITool[]): ApiStream {
+	async *createMessage(systemPrompt: string, messages: ClineStorageMessage[], tools?: OpenAITool[]): ApiStream {
 		const client = this.ensureClient()
 		const model = this.getModel()
 

src/core/api/providers/dify.ts

@@ -1,4 +1,4 @@
-import { Anthropic } from "@anthropic-ai/sdk"
+import { ClineStorageMessage } from "@/shared/messages/content"
 import { fetch } from "@/shared/net"
 import { ModelInfo } from "../../../shared/api"
 import { ApiHandler } from "../index"
@@ -97,7 +97,7 @@ export class DifyHandler implements ApiHandler {
 		}
 	}
 
-	async *createMessage(systemPrompt: string, messages: Anthropic.Messages.MessageParam[]): ApiStream {
+	async *createMessage(systemPrompt: string, messages: ClineStorageMessage[]): ApiStream {
 		console.log("[DIFY DEBUG] createMessage called with:", {
 			systemPromptLength: systemPrompt?.length || 0,
 			messagesCount: messages?.length || 0,
@@ -384,7 +384,7 @@ export class DifyHandler implements ApiHandler {
 		}
 	}
 
-	private convertMessagesToQuery(systemPrompt: string, messages: Anthropic.Messages.MessageParam[]): string {
+	private convertMessagesToQuery(systemPrompt: string, messages: ClineStorageMessage[]): 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()

src/core/api/providers/doubao.ts

@@ -1,6 +1,6 @@
-import { Anthropic } from "@anthropic-ai/sdk"
 import { DoubaoModelId, doubaoDefaultModelId, doubaoModels, ModelInfo } from "@shared/api"
 import OpenAI from "openai"
+import { ClineStorageMessage } from "@/shared/messages/content"
 import { fetch } from "@/shared/net"
 import { ApiHandler, CommonApiHandlerOptions } from ".."
 import { withRetry } from "../retry"
@@ -50,7 +50,7 @@ export class DoubaoHandler implements ApiHandler {
 	}
 
 	@withRetry()
-	async *createMessage(systemPrompt: string, messages: Anthropic.Messages.MessageParam[]): ApiStream {
+	async *createMessage(systemPrompt: string, messages: ClineStorageMessage[]): ApiStream {
 		const client = this.ensureClient()
 		const model = this.getModel()
 		const openAiMessages: OpenAI.Chat.ChatCompletionMessageParam[] = [

src/core/api/providers/fireworks.ts

@@ -1,6 +1,6 @@
-import { Anthropic } from "@anthropic-ai/sdk"
 import { FireworksModelId, fireworksDefaultModelId, fireworksModels, ModelInfo } from "@shared/api"
 import OpenAI from "openai"
+import { ClineStorageMessage } from "@/shared/messages/content"
 import { fetch } from "@/shared/net"
 import { ApiHandler, CommonApiHandlerOptions } from ".."
 import { withRetry } from "../retry"
@@ -41,7 +41,7 @@ export class FireworksHandler implements ApiHandler {
 	}
 
 	@withRetry()
-	async *createMessage(systemPrompt: string, messages: Anthropic.Messages.MessageParam[]): ApiStream {
+	async *createMessage(systemPrompt: string, messages: ClineStorageMessage[]): ApiStream {
 		const client = this.ensureClient()
 		const modelId = this.options.fireworksModelId ?? ""
 

src/core/api/providers/gemini.ts

@@ -1,4 +1,3 @@
-import type { Anthropic } from "@anthropic-ai/sdk"
 // Restore GenerateContentConfig import and add GenerateContentResponseUsageMetadata
 import {
 	ApiError,
@@ -7,10 +6,11 @@ import {
 	type GenerateContentResponseUsageMetadata,
 	GoogleGenAI,
 	FunctionDeclaration as GoogleTool,
-	Part,
+	ThinkingLevel,
 } from "@google/genai"
 import { GeminiModelId, geminiDefaultModelId, geminiModels, ModelInfo } from "@shared/api"
 import { telemetryService } from "@/services/telemetry"
+import { ClineStorageMessage } from "@/shared/messages/content"
 import { ApiHandler, CommonApiHandlerOptions } from "../"
 import { RetriableError, withRetry } from "../retry"
 import { convertAnthropicMessageToGemini } from "../transform/gemini-format"
@@ -110,34 +110,45 @@ export class GeminiHandler implements ApiHandler {
 		baseDelay: 2000,
 		maxDelay: 15000,
 	})
-	async *createMessage(systemPrompt: string, messages: Anthropic.Messages.MessageParam[], tools?: GoogleTool[]): ApiStream {
+	async *createMessage(systemPrompt: string, messages: ClineStorageMessage[], tools?: GoogleTool[]): ApiStream {
 		const client = this.ensureClient()
 		const { id: modelId, info } = this.getModel()
 		const contents = messages.map(convertAnthropicMessageToGemini)
 
 		// Configure thinking budget if supported
-		const thinkingBudget = this.options.thinkingBudgetTokens ?? 0
-		const _maxBudget = info.thinkingConfig?.maxBudget ?? 0
+		const _thinkingBudget = this.options.thinkingBudgetTokens ?? 0
+		const maxBudget = info.thinkingConfig?.maxBudget ?? 24576
+		const thinkingBudget = Math.min(_thinkingBudget, maxBudget)
+		const thinkLevel = info.thinkingConfig?.thinkingLevel
+		// When ThinkingLevel is defineded, thinking budget cannot be zero
+		// and only level is used to control thinking behavior.
+		const thinkingLevel = thinkLevel ? (thinkLevel === "low" ? ThinkingLevel.LOW : ThinkingLevel.HIGH) : undefined
 
 		// Set up base generation config
 		const requestConfig: GenerateContentConfig = {
 			// Add base URL if configured
 			httpOptions: this.options.geminiBaseUrl ? { baseUrl: this.options.geminiBaseUrl } : undefined,
-			...{ systemInstruction: systemPrompt },
+			systemInstruction: systemPrompt,
 			// Set temperature (default to 0)
-			temperature: 0,
+			// Gemini 3.0 recommends 1.0
+			temperature: info.temperature ?? 1,
 		}
 
 		// Add thinking config if the model supports it
-		if (thinkingBudget > 0) {
-			requestConfig.thinkingConfig = {
-				thinkingBudget: thinkingBudget,
-				includeThoughts: true,
-			}
+		requestConfig.thinkingConfig = {
+			// Turn off thinking:
+			// thinkingBudget: 0
+			// Turn on dynamic thinking:
+			// thinkingBudget: -1
+			// Turn on fixed thinking budget:
+			thinkingBudget: thinkingLevel ? undefined : thinkingBudget,
+			thinkingLevel,
+			includeThoughts: thinkingBudget > 0,
 		}
 
 		// Generate content using the configured parameters
 		const sdkCallStartTime = Date.now()
+		let responseId: string | undefined
 		let sdkFirstChunkTime: number | undefined
 		let ttftSdkMs: number | undefined
 		let apiSuccess = false
@@ -148,7 +159,8 @@ export class GeminiHandler implements ApiHandler {
 		let thoughtsTokenCount = 0 // Initialize thought token counts
 		let lastUsageMetadata: GenerateContentResponseUsageMetadata | undefined
 
-		if (tools?.length) {
+		const isNativeToolCallsEnabled = tools?.length
+		if (isNativeToolCallsEnabled) {
 			requestConfig.tools = [{ functionDeclarations: tools }]
 			requestConfig.toolConfig = {
 				// Force the model to call 'any' function.
@@ -176,56 +188,45 @@ export class GeminiHandler implements ApiHandler {
 				}
 
 				// Handle thinking content from Gemini's response
-				const candidateForThoughts = chunk?.candidates?.[0]
-				const partsForThoughts = candidateForThoughts?.content?.parts
-				let thoughts = "" // Initialize as empty string
-
-				if (partsForThoughts) {
-					// This ensures partsForThoughts is a Part[] array
-					for (const part of partsForThoughts) {
-						const { thought, text } = part as Part
-						if (thought && text) {
-							// Ensure part.text exists
-							// Handle the thought part
-							thoughts += text + "\n" // Append thought and a newline
+				const parts = chunk?.candidates?.[0]?.content?.parts || []
+				for (const part of parts) {
+					if (part.thought && part.text) {
+						yield {
+							type: "reasoning",
+							id: chunk.responseId,
+							reasoning: part.text || "",
+							signature: part.thoughtSignature,
+						}
+					} else if (part.text) {
+						yield {
+							type: "text",
+							text: part.text,
+							id: chunk.responseId,
+							signature: part.thoughtSignature,
 						}
 					}
-				}
-
-				if (thoughts.trim() !== "") {
-					yield {
-						type: "reasoning",
-						reasoning: thoughts.trim(),
-					}
-					thoughts = "" // Reset thoughts after yielding
-				}
-
-				if (chunk.text) {
-					yield {
-						type: "text",
-						text: chunk.text,
-					}
-				}
-
-				if (tools && chunk.functionCalls && chunk.functionCalls?.length > 0) {
-					for (const functionCall of chunk.functionCalls) {
-						if (functionCall.args) {
-							console.log("[GeminiHandler] tool call received:", functionCall)
+					if (part.functionCall) {
+						const functionCall = part.functionCall
+						const args = Object.entries(functionCall.args || {}).filter(([_key, val]) => !!val)
+						if (functionCall.args && args.length > 0) {
 							yield {
 								type: "tool_calls",
+								id: chunk.responseId,
 								tool_call: {
 									function: {
-										id: functionCall.id || functionCall.name,
+										id: chunk.responseId,
 										name: functionCall.name,
 										arguments: JSON.stringify(functionCall.args),
 									},
 								},
+								signature: part.thoughtSignature,
 							}
 						}
 					}
 				}
 
 				if (chunk.usageMetadata) {
+					responseId = chunk.responseId
 					lastUsageMetadata = chunk.usageMetadata
 					promptTokens = lastUsageMetadata.promptTokenCount ?? promptTokens
 					outputTokens = lastUsageMetadata.candidatesTokenCount ?? outputTokens
@@ -251,6 +252,7 @@ export class GeminiHandler implements ApiHandler {
 					cacheReadTokens,
 					cacheWriteTokens: 0,
 					totalCost,
+					id: responseId,
 				}
 			}
 		} catch (error) {

src/core/api/providers/groq.ts

@@ -1,8 +1,8 @@
-import { Anthropic } from "@anthropic-ai/sdk"
 import { GroqModelId, groqDefaultModelId, groqModels, ModelInfo } from "@shared/api"
 import { calculateApiCostOpenAI } from "@utils/cost"
 import OpenAI from "openai"
 import type { ChatCompletionTool as OpenAITool } from "openai/resources/chat/completions"
+import { ClineStorageMessage } from "@/shared/messages/content"
 import { fetch } from "@/shared/net"
 import { ApiHandler, CommonApiHandlerOptions } from "../"
 import { withRetry } from "../retry"
@@ -192,7 +192,7 @@ export class GroqHandler implements ApiHandler {
 	}
 
 	@withRetry()
-	async *createMessage(systemPrompt: string, messages: Anthropic.Messages.MessageParam[], tools?: OpenAITool[]): ApiStream {
+	async *createMessage(systemPrompt: string, messages: ClineStorageMessage[], tools?: OpenAITool[]): ApiStream {
 		const client = this.ensureClient()
 		const model = this.getModel()
 		const modelFamily = this.detectModelFamily(model.id)

src/core/api/providers/hicap.ts

@@ -1,7 +1,7 @@
-import { Anthropic } from "@anthropic-ai/sdk"
 import { hicapModelInfoSaneDefaults, ModelInfo } from "@shared/api"
 import OpenAI from "openai"
 import type { ChatCompletionReasoningEffort } from "openai/resources/chat/completions"
+import { ClineStorageMessage } from "@/shared/messages/content"
 import { ApiHandler, CommonApiHandlerOptions } from "../index"
 import { withRetry } from "../retry"
 import { convertToOpenAiMessages } from "../transform/openai-format"
@@ -44,7 +44,7 @@ export class HicapHandler implements ApiHandler {
 	}
 
 	@withRetry()
-	async *createMessage(systemPrompt: string, messages: Anthropic.Messages.MessageParam[]): ApiStream {
+	async *createMessage(systemPrompt: string, messages: ClineStorageMessage[]): ApiStream {
 		const client = this.ensureClient()
 		const modelId = this.options.hicapModelId ?? ""
 

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

@@ -1,7 +1,7 @@
-import { Anthropic } from "@anthropic-ai/sdk"
 import { HuaweiCloudMaasModelId, huaweiCloudMaasDefaultModelId, huaweiCloudMaasModels, ModelInfo } from "@shared/api"
 import OpenAI from "openai"
 import type { ChatCompletionTool as OpenAITool } from "openai/resources/chat/completions"
+import { ClineStorageMessage } from "@/shared/messages/content"
 import { fetch } from "@/shared/net"
 import { ApiHandler, CommonApiHandlerOptions } from ".."
 import { withRetry } from "../retry"
@@ -62,7 +62,7 @@ export class HuaweiCloudMaaSHandler implements ApiHandler {
 	}
 
 	@withRetry()
-	async *createMessage(systemPrompt: string, messages: Anthropic.Messages.MessageParam[], tools?: OpenAITool[]): ApiStream {
+	async *createMessage(systemPrompt: string, messages: ClineStorageMessage[], tools?: OpenAITool[]): ApiStream {
 		const client = this.ensureClient()
 		const model = this.getModel()
 		const openAiMessages: OpenAI.Chat.ChatCompletionMessageParam[] = [

src/core/api/providers/huggingface.ts

@@ -1,8 +1,8 @@
-import { Anthropic } from "@anthropic-ai/sdk"
 import { HuggingFaceModelId, huggingFaceDefaultModelId, huggingFaceModels, ModelInfo } from "@shared/api"
 import { calculateApiCostOpenAI } from "@utils/cost"
 import OpenAI from "openai"
 import type { ChatCompletionTool as OpenAITool } from "openai/resources/chat/completions"
+import { ClineStorageMessage } from "@/shared/messages/content"
 import { fetch } from "@/shared/net"
 import { ApiHandler, CommonApiHandlerOptions } from "../"
 import { withRetry } from "../retry"
@@ -69,7 +69,7 @@ export class HuggingFaceHandler implements ApiHandler {
 	}
 
 	@withRetry()
-	async *createMessage(systemPrompt: string, messages: Anthropic.Messages.MessageParam[], tools?: OpenAITool[]): ApiStream {
+	async *createMessage(systemPrompt: string, messages: ClineStorageMessage[], tools?: OpenAITool[]): ApiStream {
 		try {
 			const client = this.ensureClient()
 			const model = this.getModel()

src/core/api/providers/litellm.ts

@@ -1,6 +1,7 @@
 import { Anthropic } from "@anthropic-ai/sdk"
 import { LiteLLMModelInfo, liteLlmDefaultModelId, liteLlmModelInfoSaneDefaults } from "@shared/api"
 import OpenAI from "openai"
+import { ClineStorageMessage } from "@/shared/messages/content"
 import { fetch } from "@/shared/net"
 import { isAnthropicModelId } from "@/utils/model-utils"
 import { ApiHandler, CommonApiHandlerOptions } from ".."
@@ -183,7 +184,7 @@ export class LiteLlmHandler implements ApiHandler {
 	}
 
 	@withRetry()
-	async *createMessage(systemPrompt: string, messages: Anthropic.Messages.MessageParam[]): ApiStream {
+	async *createMessage(systemPrompt: string, messages: ClineStorageMessage[]): ApiStream {
 		const client = this.ensureClient()
 		const formattedMessages = convertToOpenAiMessages(messages)
 		const systemMessage: OpenAI.Chat.ChatCompletionSystemMessageParam | Anthropic.Messages.TextBlockParam = {

src/core/api/providers/lmstudio.ts

@@ -1,7 +1,7 @@
-import type { Anthropic } from "@anthropic-ai/sdk"
 import { type ModelInfo, openAiModelInfoSaneDefaults } from "@shared/api"
 import OpenAI from "openai"
 import type { ChatCompletionTool as OpenAITool } from "openai/resources/chat/completions"
+import { ClineStorageMessage } from "@/shared/messages/content"
 import { fetch } from "@/shared/net"
 import type { ApiHandler, CommonApiHandlerOptions } from "../"
 import { withRetry } from "../retry"
@@ -40,7 +40,7 @@ export class LmStudioHandler implements ApiHandler {
 	}
 
 	@withRetry({ retryAllErrors: true })
-	async *createMessage(systemPrompt: string, messages: Anthropic.Messages.MessageParam[], tools?: OpenAITool[]): ApiStream {
+	async *createMessage(systemPrompt: string, messages: ClineStorageMessage[], tools?: OpenAITool[]): ApiStream {
 		const client = this.ensureClient()
 		const openAiMessages: OpenAI.Chat.ChatCompletionMessageParam[] = [
 			{ role: "system", content: systemPrompt },

src/core/api/providers/minimax.ts

@@ -2,6 +2,7 @@ import { Anthropic } from "@anthropic-ai/sdk"
 import { Tool as AnthropicTool } from "@anthropic-ai/sdk/resources/index"
 import { Stream as AnthropicStream } from "@anthropic-ai/sdk/streaming"
 import { MinimaxModelId, ModelInfo, minimaxDefaultModelId, minimaxModels } from "@/shared/api"
+import { ClineStorageMessage } from "@/shared/messages/content"
 import { fetch } from "@/shared/net"
 import { ClineTool } from "@/shared/tools"
 import { ApiHandler, CommonApiHandlerOptions } from "../index"
@@ -45,7 +46,7 @@ export class MinimaxHandler implements ApiHandler {
 	}
 
 	@withRetry()
-	async *createMessage(systemPrompt: string, messages: Anthropic.Messages.MessageParam[], tools?: ClineTool[]): ApiStream {
+	async *createMessage(systemPrompt: string, messages: ClineStorageMessage[], tools?: ClineTool[]): ApiStream {
 		const client = this.ensureClient()
 		const model = this.getModel()
 
@@ -66,12 +67,11 @@ export class MinimaxHandler implements ApiHandler {
 			tool_choice: nativeToolsOn ? { type: "any" } : undefined,
 		})
 
-		let thinkingDeltaAccumulator = ""
 		const lastStartedToolCall = { id: "", name: "", arguments: "" }
 
 		for await (const chunk of stream) {
 			switch (chunk?.type) {
-				case "message_start":
+				case "message_start": {
 					// tells us cache reads/writes/input/output
 					const usage = chunk.message.usage
 					yield {
@@ -82,6 +82,7 @@ export class MinimaxHandler implements ApiHandler {
 						cacheReadTokens: usage.cache_read_input_tokens || undefined,
 					}
 					break
+				}
 				case "message_delta":
 					// tells us stop_reason, stop_sequence, and output tokens along the way and at the end of the message
 					yield {
@@ -100,13 +101,11 @@ export class MinimaxHandler implements ApiHandler {
 								type: "reasoning",
 								reasoning: chunk.content_block.thinking || "",
 							}
-							const thinking = chunk.content_block.thinking
-							const signature = chunk.content_block.signature
-							if (thinking && signature) {
+							if (chunk.content_block.thinking && chunk.content_block.signature) {
 								yield {
-									type: "ant_thinking",
-									thinking,
-									signature,
+									type: "reasoning",
+									reasoning: chunk.content_block.thinking,
+									signature: chunk.content_block.signature,
 								}
 							}
 							break
@@ -115,10 +114,7 @@ export class MinimaxHandler implements ApiHandler {
 							yield {
 								type: "reasoning",
 								reasoning: "[Redacted thinking block]",
-							}
-							yield {
-								type: "ant_redacted_thinking",
-								data: chunk.content_block.data,
+								redacted_data: chunk.content_block.data,
 							}
 							break
 						case "tool_use":
@@ -147,20 +143,19 @@ export class MinimaxHandler implements ApiHandler {
 				case "content_block_delta":
 					switch (chunk.delta.type) {
 						case "thinking_delta":
-							// 'reasoning' type just displays in the UI, but ant_thinking will be used to send the thinking traces back to the API
+							// 'reasoning' type just displays in the UI, but reasoning with signature will be used to send the thinking traces back to the API
 							yield {
 								type: "reasoning",
 								reasoning: chunk.delta.thinking,
 							}
-							thinkingDeltaAccumulator += chunk.delta.thinking
 							break
 						case "signature_delta":
 							// It's used when sending the thinking block back to the API
 							// API expects this in completed form, not as array of deltas
-							if (thinkingDeltaAccumulator && chunk.delta.signature) {
+							if (chunk.delta.signature) {
 								yield {
-									type: "ant_thinking",
-									thinking: thinkingDeltaAccumulator,
+									type: "reasoning",
+									reasoning: "",
 									signature: chunk.delta.signature,
 								}
 							}

src/core/api/providers/mistral.ts

@@ -1,9 +1,9 @@
-import { Anthropic } from "@anthropic-ai/sdk"
 import { Mistral } from "@mistralai/mistralai"
 import { HTTPClient } from "@mistralai/mistralai/lib/http"
 import { Tool as MistralTool } from "@mistralai/mistralai/models/components/tool"
 import { MistralModelId, ModelInfo, mistralDefaultModelId, mistralModels } from "@shared/api"
 import type { ChatCompletionTool as OpenAITool } from "openai/resources/chat/completions"
+import { ClineStorageMessage } from "@/shared/messages/content"
 import { fetch } from "@/shared/net"
 import { ApiHandler, CommonApiHandlerOptions } from "../"
 import { withRetry } from "../retry"
@@ -48,7 +48,7 @@ export class MistralHandler implements ApiHandler {
 	}
 
 	@withRetry()
-	async *createMessage(systemPrompt: string, messages: Anthropic.Messages.MessageParam[], tools?: OpenAITool[]): ApiStream {
+	async *createMessage(systemPrompt: string, messages: ClineStorageMessage[], tools?: OpenAITool[]): ApiStream {
 		const client = this.ensureClient()
 		const stream = await client.chat
 			.stream({

src/core/api/providers/moonshot.ts

@@ -1,7 +1,7 @@
-import { Anthropic } from "@anthropic-ai/sdk"
 import OpenAI from "openai"
 import type { ChatCompletionTool as OpenAITool } from "openai/resources/chat/completions"
 import { ModelInfo, MoonshotModelId, moonshotDefaultModelId, moonshotModels } from "@/shared/api"
+import { ClineStorageMessage } from "@/shared/messages/content"
 import { fetch } from "@/shared/net"
 import { ApiHandler, CommonApiHandlerOptions } from "../index"
 import { withRetry } from "../retry"
@@ -40,7 +40,7 @@ export class MoonshotHandler implements ApiHandler {
 	}
 
 	@withRetry()
-	async *createMessage(systemPrompt: string, messages: Anthropic.Messages.MessageParam[], tools?: OpenAITool[]): ApiStream {
+	async *createMessage(systemPrompt: string, messages: ClineStorageMessage[], tools?: OpenAITool[]): ApiStream {
 		const client = this.ensureClient()
 		const model = this.getModel()
 

src/core/api/providers/nebius.ts

@@ -1,7 +1,7 @@
-import { Anthropic } from "@anthropic-ai/sdk"
 import { type ModelInfo, type NebiusModelId, nebiusDefaultModelId, nebiusModels } from "@shared/api"
 import OpenAI from "openai"
 import type { ChatCompletionTool as OpenAITool } from "openai/resources/chat/completions"
+import { ClineStorageMessage } from "@/shared/messages/content"
 import { fetch } from "@/shared/net"
 import { ApiHandler, CommonApiHandlerOptions } from "../index"
 import { withRetry } from "../retry"
@@ -39,7 +39,7 @@ export class NebiusHandler implements ApiHandler {
 	}
 
 	@withRetry()
-	async *createMessage(systemPrompt: string, messages: Anthropic.Messages.MessageParam[], tools?: OpenAITool[]): ApiStream {
+	async *createMessage(systemPrompt: string, messages: ClineStorageMessage[], tools?: OpenAITool[]): ApiStream {
 		const client = this.ensureClient()
 		const model = this.getModel()
 

src/core/api/providers/nousresearch.ts

@@ -0,0 +1,92 @@
+import { ModelInfo, NousResearchModelId, nousResearchDefaultModelId, nousResearchModels } from "@shared/api"
+import OpenAI from "openai"
+import { ClineStorageMessage } from "@/shared/messages/content"
+import { ApiHandler, CommonApiHandlerOptions } from "../index"
+import { withRetry } from "../retry"
+import { convertToOpenAiMessages } from "../transform/openai-format"
+import { ApiStream } from "../transform/stream"
+
+interface NousResearchHandlerOptions extends CommonApiHandlerOptions {
+	nousResearchApiKey?: string
+	apiModelId?: string
+}
+
+export class NousResearchHandler implements ApiHandler {
+	private options: NousResearchHandlerOptions
+	private client: OpenAI | undefined
+
+	constructor(options: NousResearchHandlerOptions) {
+		this.options = options
+	}
+
+	private ensureClient(): OpenAI {
+		if (!this.client) {
+			if (!this.options.nousResearchApiKey) {
+				throw new Error("NousResearch API key is required")
+			}
+			try {
+				this.client = new OpenAI({
+					baseURL: "https://inference-api.nousResearch.com/v1",
+					apiKey: this.options.nousResearchApiKey,
+				})
+			} catch (error: any) {
+				throw new Error(`Error creating NousResearch client: ${error.message}`)
+			}
+		}
+		return this.client
+	}
+
+	@withRetry()
+	async *createMessage(systemPrompt: string, messages: ClineStorageMessage[]): ApiStream {
+		const client = this.ensureClient()
+		const model = this.getModel()
+
+		const openAiMessages: OpenAI.Chat.ChatCompletionMessageParam[] = [
+			{ role: "system", content: systemPrompt },
+			...convertToOpenAiMessages(messages),
+		]
+
+		const stream = await client.chat.completions.create({
+			model: model.id,
+			messages: openAiMessages,
+			temperature: 0,
+			stream: true,
+			stream_options: { include_usage: true },
+		})
+
+		for await (const chunk of stream) {
+			const delta = chunk.choices[0]?.delta
+			if (delta?.content) {
+				yield {
+					type: "text",
+					text: delta.content,
+				}
+			}
+
+			if (delta && "reasoning_content" in delta && delta.reasoning_content) {
+				yield {
+					type: "reasoning",
+					reasoning: (delta.reasoning_content as string | undefined) || "",
+				}
+			}
+
+			if (chunk.usage) {
+				yield {
+					type: "usage",
+					inputTokens: chunk.usage.prompt_tokens || 0,
+					outputTokens: chunk.usage.completion_tokens || 0,
+				}
+			}
+		}
+	}
+
+	getModel(): { id: NousResearchModelId; info: ModelInfo } {
+		const modelId = this.options.apiModelId
+
+		if (modelId && modelId in nousResearchModels) {
+			const id = modelId as NousResearchModelId
+			return { id, info: nousResearchModels[id] }
+		}
+		return { id: nousResearchDefaultModelId, info: nousResearchModels[nousResearchDefaultModelId] }
+	}
+}

src/core/api/providers/oca.ts

@@ -1,4 +1,3 @@
-import { Anthropic } from "@anthropic-ai/sdk"
 import { LiteLLMModelInfo, liteLlmDefaultModelId, liteLlmModelInfoSaneDefaults } from "@shared/api"
 import OpenAI, { APIError, OpenAIError } from "openai"
 import type { FinalRequestOptions, Headers as OpenAIHeaders } from "openai/core"
@@ -11,6 +10,7 @@ import {
 } from "@/services/auth/oca/utils/constants"
 import { createOcaHeaders } from "@/services/auth/oca/utils/utils"
 import { Logger } from "@/services/logging/Logger"
+import { ClineStorageMessage } from "@/shared/messages/content"
 import { fetch } from "@/shared/net"
 import { ApiHandler, type CommonApiHandlerOptions } from ".."
 import { withRetry } from "../retry"
@@ -139,7 +139,7 @@ export class OcaHandler implements ApiHandler {
 	}
 
 	@withRetry()
-	async *createMessage(systemPrompt: string, messages: Anthropic.Messages.MessageParam[], tools?: OpenAITool[]): ApiStream {
+	async *createMessage(systemPrompt: string, messages: ClineStorageMessage[], tools?: OpenAITool[]): ApiStream {
 		const client = this.ensureClient()
 		const formattedMessages = convertToOpenAiMessages(messages)
 		const systemMessage: OpenAI.Chat.ChatCompletionSystemMessageParam = {

src/core/api/providers/ollama.ts

@@ -1,6 +1,6 @@
-import type { Anthropic } from "@anthropic-ai/sdk"
 import { type ModelInfo, openAiModelInfoSaneDefaults } from "@shared/api"
 import { type Config, type Message, Ollama } from "ollama"
+import { ClineStorageMessage } from "@/shared/messages/content"
 import type { ApiHandler, CommonApiHandlerOptions } from "../"
 import { withRetry } from "../retry"
 import { convertToOllamaMessages } from "../transform/ollama-format"
@@ -48,7 +48,7 @@ export class OllamaHandler implements ApiHandler {
 	}
 
 	@withRetry({ retryAllErrors: true })
-	async *createMessage(systemPrompt: string, messages: Anthropic.Messages.MessageParam[]): ApiStream {
+	async *createMessage(systemPrompt: string, messages: ClineStorageMessage[]): ApiStream {
 		const client = this.ensureClient()
 		const ollamaMessages: Message[] = [{ role: "system", content: systemPrompt }, ...convertToOllamaMessages(messages)]
 

src/core/api/providers/openai-native.ts

@@ -1,8 +1,8 @@
-import { Anthropic } from "@anthropic-ai/sdk"
 import { ModelInfo, OpenAiNativeModelId, openAiNativeDefaultModelId, openAiNativeModels } from "@shared/api"
 import { calculateApiCostOpenAI } from "@utils/cost"
 import OpenAI from "openai"
 import type { ChatCompletionReasoningEffort, ChatCompletionTool } from "openai/resources/chat/completions"
+import { ClineStorageMessage } from "@/shared/messages/content"
 import { fetch } from "@/shared/net"
 import { ApiHandler, CommonApiHandlerOptions } from "../"
 import { withRetry } from "../retry"
@@ -59,11 +59,7 @@ export class OpenAiNativeHandler implements ApiHandler {
 	}
 
 	@withRetry()
-	async *createMessage(
-		systemPrompt: string,
-		messages: Anthropic.Messages.MessageParam[],
-		tools?: ChatCompletionTool[],
-	): ApiStream {
+	async *createMessage(systemPrompt: string, messages: ClineStorageMessage[], tools?: ChatCompletionTool[]): ApiStream {
 		const client = this.ensureClient()
 		const model = this.getModel()
 		const toolCallProcessor = new ToolCallProcessor()
@@ -115,6 +111,9 @@ export class OpenAiNativeHandler implements ApiHandler {
 			case "gpt-5-2025-08-07":
 			case "gpt-5-mini-2025-08-07":
 			case "gpt-5-nano-2025-08-07":
+			case "gpt-5.1-2025-11-13":
+			case "gpt-5.1-chat-latest":
+			case "gpt-5.1": {
 				const stream = await client.chat.completions.create({
 					model: model.id,
 					temperature: 1,
@@ -148,6 +147,7 @@ export class OpenAiNativeHandler implements ApiHandler {
 					}
 				}
 				break
+			}
 			default: {
 				const stream = await client.chat.completions.create({
 					model: model.id,

src/core/api/providers/openai.ts

@@ -1,7 +1,7 @@
-import { Anthropic } from "@anthropic-ai/sdk"
 import { azureOpenAiDefaultApiVersion, ModelInfo, OpenAiCompatibleModelInfo, openAiModelInfoSaneDefaults } from "@shared/api"
 import OpenAI, { AzureOpenAI } from "openai"
 import type { ChatCompletionReasoningEffort, ChatCompletionTool } from "openai/resources/chat/completions"
+import { ClineStorageMessage } from "@/shared/messages/content"
 import { fetch } from "@/shared/net"
 import { ApiHandler, CommonApiHandlerOptions } from "../index"
 import { withRetry } from "../retry"
@@ -65,11 +65,7 @@ export class OpenAiHandler implements ApiHandler {
 	}
 
 	@withRetry()
-	async *createMessage(
-		systemPrompt: string,
-		messages: Anthropic.Messages.MessageParam[],
-		tools?: ChatCompletionTool[],
-	): ApiStream {
+	async *createMessage(systemPrompt: string, messages: ClineStorageMessage[], tools?: ChatCompletionTool[]): ApiStream {
 		const client = this.ensureClient()
 		const modelId = this.options.openAiModelId ?? ""
 		const isDeepseekReasoner = modelId.includes("deepseek-reasoner")
@@ -81,7 +77,13 @@ export class OpenAiHandler implements ApiHandler {
 			{ role: "system", content: systemPrompt },
 			...convertToOpenAiMessages(messages),
 		]
-		let temperature: number | undefined = this.options.openAiModelInfo?.temperature ?? openAiModelInfoSaneDefaults.temperature
+		let temperature: number | undefined
+		if (this.options.openAiModelInfo?.temperature !== undefined) {
+			const tempValue = Number(this.options.openAiModelInfo.temperature)
+			temperature = tempValue === 0 ? undefined : tempValue
+		} else {
+			temperature = openAiModelInfoSaneDefaults.temperature
+		}
 		let reasoningEffort: ChatCompletionReasoningEffort | undefined
 		let maxTokens: number | undefined
 

src/core/api/providers/openrouter.ts

@@ -1,10 +1,10 @@
 import { setTimeout as setTimeoutPromise } from "node:timers/promises"
-import { Anthropic } from "@anthropic-ai/sdk"
 import { ModelInfo, openRouterDefaultModelId, openRouterDefaultModelInfo } from "@shared/api"
 import { shouldSkipReasoningForModel } from "@utils/model-utils"
 import axios from "axios"
 import OpenAI from "openai"
 import type { ChatCompletionTool as OpenAITool } from "openai/resources/chat/completions"
+import { ClineStorageMessage } from "@/shared/messages/content"
 import { fetch, getAxiosSettings } from "@/shared/net"
 import { ApiHandler, CommonApiHandlerOptions } from "../"
 import { withRetry } from "../retry"
@@ -54,7 +54,7 @@ export class OpenRouterHandler implements ApiHandler {
 	}
 
 	@withRetry()
-	async *createMessage(systemPrompt: string, messages: Anthropic.Messages.MessageParam[], tools?: OpenAITool[]): ApiStream {
+	async *createMessage(systemPrompt: string, messages: ClineStorageMessage[], tools?: OpenAITool[]): ApiStream {
 		const client = this.ensureClient()
 		this.lastGenerationId = undefined
 
@@ -127,8 +127,7 @@ export class OpenRouterHandler implements ApiHandler {
 			if ("reasoning" in delta && delta.reasoning && !shouldSkipReasoningForModel(this.options.openRouterModelId)) {
 				yield {
 					type: "reasoning",
-					// @ts-ignore-next-line
-					reasoning: delta.reasoning,
+					reasoning: typeof delta.reasoning === "string" ? delta.reasoning : JSON.stringify(delta.reasoning),
 				}
 			}
 
@@ -142,8 +141,9 @@ export class OpenRouterHandler implements ApiHandler {
 				!shouldSkipReasoningForModel(this.options.openRouterModelId)
 			) {
 				yield {
-					type: "reasoning_details",
-					reasoning_details: delta.reasoning_details,
+					type: "reasoning",
+					reasoning: "",
+					details: delta.reasoning_details,
 				}
 			}
 
@@ -214,11 +214,9 @@ export class OpenRouterHandler implements ApiHandler {
 	}
 
 	getModel(): { id: string; info: ModelInfo } {
-		const modelId = this.options.openRouterModelId
-		const modelInfo = this.options.openRouterModelInfo
-		if (modelId && modelInfo) {
-			return { id: modelId, info: modelInfo }
+		return {
+			id: this.options.openRouterModelId || openRouterDefaultModelId,
+			info: this.options.openRouterModelInfo || openRouterDefaultModelInfo,
 		}
-		return { id: openRouterDefaultModelId, info: openRouterDefaultModelInfo }
 	}
 }

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

@@ -1,10 +1,10 @@
 import { promises as fs } from "node:fs"
-import { Anthropic } from "@anthropic-ai/sdk"
 import { ModelInfo, QwenCodeModelId, qwenCodeDefaultModelId, qwenCodeModels } from "@shared/api"
 import OpenAI from "openai"
 import type { ChatCompletionTool as OpenAITool } from "openai/resources/chat/completions"
 import * as os from "os"
 import * as path from "path"
+import { ClineStorageMessage } from "@/shared/messages/content"
 import { fetch } from "@/shared/net"
 import { ApiHandler, CommonApiHandlerOptions } from "../"
 import { withRetry } from "../retry"
@@ -177,7 +177,7 @@ export class QwenCodeHandler implements ApiHandler {
 	}
 
 	@withRetry()
-	async *createMessage(systemPrompt: string, messages: Anthropic.Messages.MessageParam[], tools?: OpenAITool[]): ApiStream {
+	async *createMessage(systemPrompt: string, messages: ClineStorageMessage[], tools?: OpenAITool[]): ApiStream {
 		await this.ensureAuthenticated()
 		const client = this.ensureClient()
 		const model = this.getModel()

src/core/api/providers/qwen.ts

@@ -1,4 +1,3 @@
-import { Anthropic } from "@anthropic-ai/sdk"
 import {
 	InternationalQwenModelId,
 	internationalQwenDefaultModelId,
@@ -11,6 +10,7 @@ import {
 } from "@shared/api"
 import OpenAI from "openai"
 import type { ChatCompletionTool as OpenAITool } from "openai/resources/chat/completions"
+import { ClineStorageMessage } from "@/shared/messages/content"
 import { fetch } from "@/shared/net"
 import { ApiHandler, CommonApiHandlerOptions } from "../"
 import { withRetry } from "../retry"
@@ -81,7 +81,7 @@ export class QwenHandler implements ApiHandler {
 	}
 
 	@withRetry()
-	async *createMessage(systemPrompt: string, messages: Anthropic.Messages.MessageParam[], tools?: OpenAITool[]): ApiStream {
+	async *createMessage(systemPrompt: string, messages: ClineStorageMessage[], tools?: OpenAITool[]): ApiStream {
 		const client = this.ensureClient()
 		const model = this.getModel()
 		const isDeepseekReasoner = model.id.includes("deepseek-r1")

src/core/api/providers/requesty.ts

@@ -1,9 +1,9 @@
-import { Anthropic } from "@anthropic-ai/sdk"
 import { ModelInfo, requestyDefaultModelId, requestyDefaultModelInfo } from "@shared/api"
 import { calculateApiCostOpenAI } from "@utils/cost"
 import OpenAI from "openai"
+import { toRequestyServiceStringUrl } from "@/shared/clients/requesty"
+import { ClineStorageMessage } from "@/shared/messages/content"
 import { fetch } from "@/shared/net"
-import { toRequestyServiceStringUrl } from "@/shared/providers/requesty"
 import { ApiHandler, CommonApiHandlerOptions } from "../index"
 import { withRetry } from "../retry"
 import { convertToOpenAiMessages } from "../transform/openai-format"
@@ -59,7 +59,7 @@ export class RequestyHandler implements ApiHandler {
 	}
 
 	@withRetry()
-	async *createMessage(systemPrompt: string, messages: Anthropic.Messages.MessageParam[]): ApiStream {
+	async *createMessage(systemPrompt: string, messages: ClineStorageMessage[]): ApiStream {
 		const client = this.ensureClient()
 		const model = this.getModel()
 

src/core/api/providers/sambanova.ts

@@ -1,7 +1,7 @@
-import { Anthropic } from "@anthropic-ai/sdk"
 import { ModelInfo, SambanovaModelId, sambanovaDefaultModelId, sambanovaModels } from "@shared/api"
 import OpenAI from "openai"
 import type { ChatCompletionTool as OpenAITool } from "openai/resources/chat/completions"
+import { ClineStorageMessage } from "@/shared/messages/content"
 import { fetch } from "@/shared/net"
 import { ApiHandler, CommonApiHandlerOptions } from "../index"
 import { withRetry } from "../retry"
@@ -42,7 +42,7 @@ export class SambanovaHandler implements ApiHandler {
 	}
 
 	@withRetry()
-	async *createMessage(systemPrompt: string, messages: Anthropic.Messages.MessageParam[], tools?: OpenAITool[]): ApiStream {
+	async *createMessage(systemPrompt: string, messages: ClineStorageMessage[], tools?: OpenAITool[]): ApiStream {
 		const client = this.ensureClient()
 		const model = this.getModel()
 

src/core/api/providers/sapaicore.ts

@@ -4,10 +4,11 @@ import {
 	ConversationRole as BedrockConversationRole,
 	type Message as BedrockMessage,
 } from "@aws-sdk/client-bedrock-runtime"
-import { ChatMessages, LlmModuleConfig, OrchestrationClient, TemplatingModuleConfig } from "@sap-ai-sdk/orchestration"
+import { ChatMessage, OrchestrationClient, OrchestrationModuleConfig } from "@sap-ai-sdk/orchestration"
 import { ModelInfo, SapAiCoreModelId, sapAiCoreDefaultModelId, sapAiCoreModels } from "@shared/api"
 import axios from "axios"
 import OpenAI from "openai"
+import { ClineStorageMessage } from "@/shared/messages/content"
 import { getAxiosSettings } from "@/shared/net"
 import { ApiHandler, CommonApiHandlerOptions } from "../"
 import { withRetry } from "../retry"
@@ -115,7 +116,7 @@ namespace Bedrock {
 	 * Formats messages for models using the Converse API specification
 	 * Used by both Anthropic and Nova models to avoid code duplication
 	 */
-	export function formatMessagesForConverseAPI(messages: Anthropic.Messages.MessageParam[]): BedrockMessage[] {
+	export function formatMessagesForConverseAPI(messages: ClineStorageMessage[]): BedrockMessage[] {
 		return messages.map((message) => {
 			// Determine role (user or assistant)
 			const role = message.role === "user" ? BedrockConversationRole.USER : BedrockConversationRole.ASSISTANT
@@ -315,7 +316,7 @@ namespace Gemini {
 	 */
 	export function prepareRequestPayload(
 		systemPrompt: string,
-		messages: Anthropic.Messages.MessageParam[],
+		messages: ClineStorageMessage[],
 		model: { id: SapAiCoreModelId; info: ModelInfo },
 		thinkingBudgetTokens?: number,
 	): any {
@@ -458,7 +459,7 @@ export class SapAiCoreHandler implements ApiHandler {
 	}
 
 	@withRetry()
-	async *createMessage(systemPrompt: string, messages: Anthropic.Messages.MessageParam[]): ApiStream {
+	async *createMessage(systemPrompt: string, messages: ClineStorageMessage[]): ApiStream {
 		if (this.options.sapAiCoreUseOrchestrationMode) {
 			yield* this.createMessageWithOrchestration(systemPrompt, messages)
 		} else {
@@ -490,29 +491,31 @@ export class SapAiCoreHandler implements ApiHandler {
 		this.isAiCoreEnvSetup = true
 	}
 
-	private async *createMessageWithOrchestration(systemPrompt: string, messages: Anthropic.Messages.MessageParam[]): ApiStream {
+	private async *createMessageWithOrchestration(systemPrompt: string, messages: ClineStorageMessage[]): ApiStream {
 		try {
 			// Ensure AI Core environment variable is set up (only runs once)
 			this.ensureAiCoreEnvSetup()
 			const model = this.getModel()
 
-			// Define the LLM to be used by the Orchestration pipeline
-			const llm: LlmModuleConfig = {
-				model_name: model.id,
+			const orchestrationConfig: OrchestrationModuleConfig = {
+				promptTemplating: {
+					model: {
+						name: model.id,
+					},
+					prompt: {
+						template: [
+							{
+								role: "system",
+								content: systemPrompt,
+							},
+						],
+					},
+				},
 			}
 
-			const templating: TemplatingModuleConfig = {
-				template: [
-					{
-						role: "system",
-						content: systemPrompt,
-					},
-				],
-			}
-			const orchestrationClient = new OrchestrationClient(
-				{ llm, templating },
-				{ resourceGroup: this.options.sapAiResourceGroup || "default" },
-			)
+			const orchestrationClient = new OrchestrationClient(orchestrationConfig, {
+				resourceGroup: this.options.sapAiResourceGroup || "default",
+			})
 
 			const sapMessages = this.convertMessageParamToSAPMessages(messages)
 
@@ -538,7 +541,7 @@ export class SapAiCoreHandler implements ApiHandler {
 		}
 	}
 
-	private async *createMessageWithDeployments(systemPrompt: string, messages: Anthropic.Messages.MessageParam[]): ApiStream {
+	private async *createMessageWithDeployments(systemPrompt: string, messages: ClineStorageMessage[]): ApiStream {
 		const token = await this.getToken()
 		const headers = {
 			Authorization: `Bearer ${token}`,
@@ -1040,8 +1043,8 @@ export class SapAiCoreHandler implements ApiHandler {
 		}
 		return { id: sapAiCoreDefaultModelId, info: sapAiCoreModels[sapAiCoreDefaultModelId] }
 	}
-	private convertMessageParamToSAPMessages(messages: Anthropic.Messages.MessageParam[]): ChatMessages {
+	private convertMessageParamToSAPMessages(messages: ClineStorageMessage[]): ChatMessage[] {
 		// Use the existing OpenAI converter since the logic is identical
-		return convertToOpenAiMessages(messages) as ChatMessages
+		return convertToOpenAiMessages(messages) as ChatMessage[]
 	}
 }

src/core/api/providers/together.ts

@@ -1,7 +1,7 @@
-import { Anthropic } from "@anthropic-ai/sdk"
 import { ModelInfo, openAiModelInfoSaneDefaults } from "@shared/api"
 import OpenAI from "openai"
 import type { ChatCompletionTool as OpenAITool } from "openai/resources/chat/completions"
+import { ClineStorageMessage } from "@/shared/messages/content"
 import { fetch } from "@/shared/net"
 import { ApiHandler, CommonApiHandlerOptions } from "../index"
 import { withRetry } from "../retry"
@@ -42,7 +42,7 @@ export class TogetherHandler implements ApiHandler {
 	}
 
 	@withRetry()
-	async *createMessage(systemPrompt: string, messages: Anthropic.Messages.MessageParam[], tools?: OpenAITool[]): ApiStream {
+	async *createMessage(systemPrompt: string, messages: ClineStorageMessage[], tools?: OpenAITool[]): ApiStream {
 		const client = this.ensureClient()
 		const modelId = this.options.togetherModelId ?? ""
 		const isDeepseekReasoner = modelId.includes("deepseek-reasoner")

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

@@ -1,7 +1,7 @@
-import { Anthropic } from "@anthropic-ai/sdk"
 import { ModelInfo, openRouterDefaultModelId, openRouterDefaultModelInfo } from "@shared/api"
 import OpenAI from "openai"
 import type { ChatCompletionTool as OpenAITool } from "openai/resources/chat/completions"
+import { ClineStorageMessage } from "@/shared/messages/content"
 import { fetch } from "@/shared/net"
 import { ApiHandler, CommonApiHandlerOptions } from "../index"
 import { withRetry } from "../retry"
@@ -47,7 +47,7 @@ export class VercelAIGatewayHandler implements ApiHandler {
 	}
 
 	@withRetry()
-	async *createMessage(systemPrompt: string, messages: Anthropic.Messages.MessageParam[], tools?: OpenAITool[]): ApiStream {
+	async *createMessage(systemPrompt: string, messages: ClineStorageMessage[], tools?: OpenAITool[]): ApiStream {
 		const client = this.ensureClient()
 		const modelId = this.getModel().id
 		const modelInfo = this.getModel().info
@@ -95,28 +95,23 @@ export class VercelAIGatewayHandler implements ApiHandler {
 					delta.reasoning_details.length // exists and non-0
 				) {
 					yield {
-						type: "reasoning_details",
-						reasoning_details: delta.reasoning_details,
+						type: "reasoning",
+						reasoning: "",
+						details: delta.reasoning_details,
 					}
 				}
 
 				if (!didOutputUsage && chunk.usage) {
-					const inputTokens = chunk.usage.prompt_tokens || 0
-					const outputTokens =
-						(chunk.usage.completion_tokens || 0) + (chunk.usage.completion_tokens_details?.reasoning_tokens || 0)
-
-					const cacheReadTokens = chunk.usage.prompt_tokens_details?.cached_tokens || 0
 					// @ts-ignore - Vercel AI Gateway extends OpenAI types
-					const cacheWriteTokens = chunk.usage.cache_creation_input_tokens || 0
+					const totalCost = (chunk.usage.cost || 0) + (chunk.usage.cost_details?.upstream_inference_cost || 0)
 
 					yield {
 						type: "usage",
-						inputTokens: inputTokens,
-						outputTokens: outputTokens,
-						cacheWriteTokens: cacheWriteTokens,
-						cacheReadTokens: cacheReadTokens,
-						// @ts-expect-error - Vercel AI Gateway extends OpenAI types
-						totalCost: chunk.usage.cost || 0,
+						cacheWriteTokens: 0,
+						cacheReadTokens: chunk.usage.prompt_tokens_details?.cached_tokens || 0,
+						inputTokens: (chunk.usage.prompt_tokens || 0) - (chunk.usage.prompt_tokens_details?.cached_tokens || 0),
+						outputTokens: chunk.usage.completion_tokens || 0,
+						totalCost,
 					}
 					didOutputUsage = true
 				}

src/core/api/providers/vertex.ts

@@ -1,11 +1,12 @@
-import { Anthropic } from "@anthropic-ai/sdk"
 import { Tool as AnthropicTool } from "@anthropic-ai/sdk/resources/index"
 import { AnthropicVertex } from "@anthropic-ai/vertex-sdk"
 import { FunctionDeclaration as GoogleTool } from "@google/genai"
 import { ModelInfo, VertexModelId, vertexDefaultModelId, vertexModels } from "@shared/api"
+import { ClineStorageMessage } from "@/shared/messages/content"
 import { ClineTool } from "@/shared/tools"
 import { ApiHandler, CommonApiHandlerOptions } from "../"
 import { withRetry } from "../retry"
+import { sanitizeAnthropicMessages } from "../transform/anthropic-format"
 import { ApiStream } from "../transform/stream"
 import { GeminiHandler } from "./gemini"
 
@@ -66,7 +67,7 @@ export class VertexHandler implements ApiHandler {
 	}
 
 	@withRetry()
-	async *createMessage(systemPrompt: string, messages: Anthropic.Messages.MessageParam[], tools?: ClineTool[]): ApiStream {
+	async *createMessage(systemPrompt: string, messages: ClineStorageMessage[], tools?: ClineTool[]): ApiStream {
 		const model = this.getModel()
 		const modelId = model.id
 
@@ -122,46 +123,7 @@ export class VertexHandler implements ApiHandler {
 								cache_control: { type: "ephemeral" },
 							},
 						],
-						messages: messages.map((message, index) => {
-							if (index === lastUserMsgIndex || index === secondLastMsgUserIndex) {
-								return {
-									...message,
-									content:
-										typeof message.content === "string"
-											? [
-													{
-														type: "text",
-														text: message.content,
-														cache_control: {
-															type: "ephemeral",
-														},
-													},
-												]
-											: message.content.map((content, contentIndex) =>
-													contentIndex === message.content.length - 1
-														? {
-																...content,
-																cache_control: {
-																	type: "ephemeral",
-																},
-															}
-														: content,
-												),
-								}
-							}
-							return {
-								...message,
-								content:
-									typeof message.content === "string"
-										? [
-												{
-													type: "text",
-													text: message.content,
-												},
-											]
-										: message.content,
-							}
-						}),
+						messages: sanitizeAnthropicMessages(messages, lastUserMsgIndex, secondLastMsgUserIndex),
 						stream: true,
 						tools: tools?.length ? (tools as AnthropicTool[]) : undefined,
 						// tool_choice options:
@@ -187,18 +149,7 @@ export class VertexHandler implements ApiHandler {
 							type: "text",
 						},
 					],
-					messages: messages.map((message) => ({
-						...message,
-						content:
-							typeof message.content === "string"
-								? [
-										{
-											type: "text",
-											text: message.content,
-										},
-									]
-								: message.content,
-					})),
+					messages: sanitizeAnthropicMessages(messages),
 					stream: true,
 					tools: tools?.length ? (tools as AnthropicTool[]) : undefined,
 					// tool_choice options:
@@ -215,7 +166,7 @@ export class VertexHandler implements ApiHandler {
 
 		for await (const chunk of stream) {
 			switch (chunk?.type) {
-				case "message_start":
+				case "message_start": {
 					const usage = chunk.message.usage
 					yield {
 						type: "usage",
@@ -225,6 +176,7 @@ export class VertexHandler implements ApiHandler {
 						cacheReadTokens: usage.cache_read_input_tokens || undefined,
 					}
 					break
+				}
 				case "message_delta":
 					yield {
 						type: "usage",

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

@@ -1,8 +1,8 @@
-import { Anthropic } from "@anthropic-ai/sdk"
 import { ModelInfo, openAiModelInfoSaneDefaults } from "@shared/api"
 import { SELECTOR_SEPARATOR, stringifyVsCodeLmModelSelector } from "@shared/vsCodeSelectorUtils"
 import { calculateApiCostAnthropic } from "@utils/cost"
 import * as vscode from "vscode"
+import { ClineStorageMessage } from "@/shared/messages/content"
 import { ApiHandler, CommonApiHandlerOptions, SingleCompletionHandler } from "../"
 import { withRetry } from "../retry"
 import { ApiStream } from "../transform/stream"
@@ -366,7 +366,7 @@ export class VsCodeLmHandler implements ApiHandler, SingleCompletionHandler {
 	}
 
 	@withRetry()
-	async *createMessage(systemPrompt: string, messages: Anthropic.Messages.MessageParam[]): ApiStream {
+	async *createMessage(systemPrompt: string, messages: ClineStorageMessage[]): ApiStream {
 		// Ensure clean state before starting a new request
 		this.ensureCleanState()
 		const client: vscode.LanguageModelChat = await this.getClient()

src/core/api/providers/xai.ts

@@ -1,9 +1,9 @@
-import { Anthropic } from "@anthropic-ai/sdk"
 import { ModelInfo, XAIModelId, xaiDefaultModelId, xaiModels } from "@shared/api"
 import { shouldSkipReasoningForModel } from "@utils/model-utils"
 import OpenAI from "openai"
 import type { ChatCompletionTool as OpenAITool } from "openai/resources/chat/completions"
 import { ChatCompletionReasoningEffort } from "openai/resources/chat/completions"
+import { ClineStorageMessage } from "@/shared/messages/content"
 import { fetch } from "@/shared/net"
 import { ApiHandler, CommonApiHandlerOptions } from "../"
 import { withRetry } from "../retry"
@@ -44,7 +44,7 @@ export class XAIHandler implements ApiHandler {
 	}
 
 	@withRetry()
-	async *createMessage(systemPrompt: string, messages: Anthropic.Messages.MessageParam[], tools?: OpenAITool[]): ApiStream {
+	async *createMessage(systemPrompt: string, messages: ClineStorageMessage[], tools?: OpenAITool[]): ApiStream {
 		const client = this.ensureClient()
 		const modelId = this.getModel().id
 		// ensure reasoning effort is either "low" or "high" for grok-3-mini

src/core/api/providers/zai.ts

@@ -1,4 +1,3 @@
-import { Anthropic } from "@anthropic-ai/sdk"
 import {
 	internationalZAiDefaultModelId,
 	internationalZAiModelId,
@@ -10,6 +9,7 @@ import {
 } from "@shared/api"
 import OpenAI from "openai"
 import type { ChatCompletionTool as OpenAITool } from "openai/resources/chat/completions"
+import { ClineStorageMessage } from "@/shared/messages/content"
 import { fetch } from "@/shared/net"
 import { version as extensionVersion } from "../../../../package.json"
 import { ApiHandler, CommonApiHandlerOptions } from ".."
@@ -76,7 +76,7 @@ export class ZAiHandler implements ApiHandler {
 	}
 
 	@withRetry()
-	async *createMessage(systemPrompt: string, messages: Anthropic.Messages.MessageParam[], tools?: OpenAITool[]): ApiStream {
+	async *createMessage(systemPrompt: string, messages: ClineStorageMessage[], tools?: OpenAITool[]): ApiStream {
 		const client = this.ensureClient()
 		const model = this.getModel()
 		const openAiMessages: OpenAI.Chat.ChatCompletionMessageParam[] = [

src/core/api/transform/anthropic-format.ts

@@ -1,19 +1,83 @@
-import { Tool as AnthropicTool } from "@anthropic-ai/sdk/resources/index"
-import { ChatCompletionTool as OpenAITool } from "openai/resources/chat/completions"
+import Anthropic from "@anthropic-ai/sdk"
+import { ClineContent, ClineStorageMessage } from "@/shared/messages/content"
 
 /**
- * Converts an OpenAI ChatCompletionTool into an Anthropic Tool definition
+ * Sanitize Anthropic messages by removing reasoning details and adding ephemeral cache control
+ * to the last two user messages to prevent them from being stored in Anthropic's cache.
  */
-export function openAIToolToAnthropic(openAITool: OpenAITool): AnthropicTool {
-	const func = openAITool.function
+export function sanitizeAnthropicMessages(
+	messages: Array<ClineStorageMessage>,
+	lastUserMsgIndex?: number,
+	secondLastMsgUserIndex?: number,
+): Array<Anthropic.Messages.MessageParam> {
+	return messages.map((_message, index) => {
+		const message = removeUnknownParams(_message)
+		const addCacheControl = lastUserMsgIndex !== undefined && secondLastMsgUserIndex !== undefined
+		// Construct message
+		if (addCacheControl && (index === lastUserMsgIndex || index === secondLastMsgUserIndex)) {
+			return {
+				...message,
+				content:
+					typeof message.content === "string"
+						? [
+								{
+									type: "text",
+									text: message.content,
+									cache_control: {
+										type: "ephemeral",
+									},
+								},
+							]
+						: message.content.map((content, contentIndex) =>
+								contentIndex === message.content.length - 1
+									? {
+											...content,
+											cache_control: {
+												type: "ephemeral",
+											},
+										}
+									: content,
+							),
+			}
+		}
 
+		return {
+			...message,
+			content:
+				typeof message.content === "string"
+					? [
+							{
+								type: "text",
+								text: message.content,
+							},
+						]
+					: message.content,
+		}
+	})
+}
+
+/**
+ * Remove reasoning details and other known params that are not Anthropic specific.
+ */
+function removeUnknownParams(param: ClineStorageMessage): Anthropic.Messages.MessageParam {
+	// Construct new content array with known Anthropic content blocks only.
 	return {
-		name: func.name,
-		description: func.description || "",
-		input_schema: {
-			type: "object",
-			properties: func.parameters?.properties || {},
-			required: func.parameters?.required || [],
-		},
+		role: param.role === "user" ? "user" : "assistant",
+		content: Array.isArray(param.content) ? param.content.map(sanitizeAnthropicContentBlock) : param.content, // String content remains unchanged
 	}
 }
+
+/**
+ * Clean a content block by removing Cline-specific fields and returning only provider-compatible fields
+ */
+function sanitizeAnthropicContentBlock(block: ClineContent): Anthropic.ContentBlock {
+	// Fast path: if no reasoning_details property exists, return as-is
+	// Including reasoning_details in non-openrouter/cline providers may cause API errors
+	if ("reasoning_details" in block || "call_id" in block || "summary" in block || "signature" in block) {
+		// biome-ignore lint/correctness/noUnusedVariables: intentional destructuring to remove properties
+		const { reasoning_details, call_id, summary, signature, ...cleanBlock } = block as any
+		return cleanBlock as Anthropic.ContentBlock
+	}
+
+	return block as Anthropic.ContentBlock
+}

src/core/api/transform/gemini-format.ts

@@ -1,44 +1,50 @@
 import { Anthropic } from "@anthropic-ai/sdk"
 import { Content, GenerateContentResponse, Part } from "@google/genai"
+import { ClineStorageMessage } from "@/shared/messages/content"
 
-export function convertAnthropicContentToGemini(content: string | Anthropic.ContentBlockParam[]): Part[] {
+export function convertAnthropicContentToGemini(content: string | ClineStorageMessage["content"]): Part[] {
 	if (typeof content === "string") {
 		return [{ text: content }]
 	}
-	return content.flatMap((block): Part => {
-		switch (block.type) {
-			case "text":
-				return { text: block.text }
-			case "image":
-				if (block.source.type !== "base64") {
-					throw new Error("Unsupported image source type")
-				}
-				return {
-					inlineData: {
-						data: block.source.data,
-						mimeType: block.source.media_type,
-					},
-				}
-			case "tool_use":
-				return {
-					functionCall: {
-						name: block.name,
-						args: block.input as Record<string, unknown>,
-					},
-				}
-			case "tool_result":
-				return {
-					functionResponse: {
-						name: block.tool_use_id,
-						response: {
-							result: block.content,
+	return content
+		.flatMap((block): Part | undefined => {
+			switch (block.type) {
+				case "text":
+					return { text: block.text, thoughtSignature: block.signature }
+				case "image":
+					if (block.source.type !== "base64") {
+						throw new Error("Unsupported image source type")
+					}
+					return {
+						inlineData: {
+							data: block.source.data,
+							mimeType: block.source.media_type,
 						},
-					},
-				}
-			default:
-				throw new Error(`Unsupported content block type: ${block.type}`)
-		}
-	})
+					}
+				case "tool_use":
+					return {
+						functionCall: {
+							name: block.name,
+							args: block.input as Record<string, unknown>,
+						},
+						thoughtSignature: block.signature,
+					}
+				case "tool_result":
+					return {
+						functionResponse: {
+							name: block.tool_use_id,
+							response: {
+								result: block.content,
+							},
+						},
+					}
+				case "thinking":
+					return { text: block.thinking, thought: true, thoughtSignature: block.signature }
+				default:
+					return undefined
+			}
+		})
+		.filter((part): part is Part => part !== undefined) // Filter out unsupported blocks
 }
 
 export function convertAnthropicMessageToGemini(message: Anthropic.Messages.MessageParam): Content {

src/core/api/transform/ollama-format.ts

@@ -1,7 +1,13 @@
-import { Anthropic } from "@anthropic-ai/sdk"
 import { Message } from "ollama"
+import {
+	ClineAssistantToolUseBlock,
+	ClineImageContentBlock,
+	ClineStorageMessage,
+	ClineTextContentBlock,
+	ClineUserToolResultContentBlock,
+} from "@/shared/messages/content"
 
-export function convertToOllamaMessages(anthropicMessages: Anthropic.Messages.MessageParam[]): Message[] {
+export function convertToOllamaMessages(anthropicMessages: Omit<ClineStorageMessage, "modelInfo">[]): Message[] {
 	const ollamaMessages: Message[] = []
 
 	for (const anthropicMessage of anthropicMessages) {
@@ -13,8 +19,8 @@ export function convertToOllamaMessages(anthropicMessages: Anthropic.Messages.Me
 		} else {
 			if (anthropicMessage.role === "user") {
 				const { nonToolMessages, toolMessages } = anthropicMessage.content.reduce<{
-					nonToolMessages: (Anthropic.TextBlockParam | Anthropic.ImageBlockParam)[]
-					toolMessages: Anthropic.ToolResultBlockParam[]
+					nonToolMessages: (ClineTextContentBlock | ClineImageContentBlock)[]
+					toolMessages: ClineUserToolResultContentBlock[]
 				}>(
 					(acc, part) => {
 						if (part.type === "tool_result") {
@@ -70,8 +76,8 @@ export function convertToOllamaMessages(anthropicMessages: Anthropic.Messages.Me
 				}
 			} else if (anthropicMessage.role === "assistant") {
 				const { nonToolMessages, toolMessages } = anthropicMessage.content.reduce<{
-					nonToolMessages: (Anthropic.TextBlockParam | Anthropic.ImageBlockParam)[]
-					toolMessages: Anthropic.ToolUseBlockParam[]
+					nonToolMessages: (ClineTextContentBlock | ClineImageContentBlock)[]
+					toolMessages: ClineAssistantToolUseBlock[]
 				}>(
 					(acc, part) => {
 						if (part.type === "tool_use") {

src/core/api/transform/openai-format.ts

@@ -1,8 +1,17 @@
 import { Anthropic } from "@anthropic-ai/sdk"
 import OpenAI from "openai"
+import {
+	ClineAssistantRedactedThinkingBlock,
+	ClineAssistantThinkingBlock,
+	ClineAssistantToolUseBlock,
+	ClineImageContentBlock,
+	ClineStorageMessage,
+	ClineTextContentBlock,
+	ClineUserToolResultContentBlock,
+} from "@/shared/messages/content"
 
 export function convertToOpenAiMessages(
-	anthropicMessages: Anthropic.Messages.MessageParam[],
+	anthropicMessages: Omit<ClineStorageMessage, "modelInfo">[],
 ): OpenAI.Chat.ChatCompletionMessageParam[] {
 	const openAiMessages: OpenAI.Chat.ChatCompletionMessageParam[] = []
 
@@ -23,8 +32,8 @@ export function convertToOpenAiMessages(
          */
 			if (anthropicMessage.role === "user") {
 				const { nonToolMessages, toolMessages } = anthropicMessage.content.reduce<{
-					nonToolMessages: (Anthropic.TextBlockParam | Anthropic.ImageBlockParam)[]
-					toolMessages: Anthropic.ToolResultBlockParam[]
+					nonToolMessages: (ClineTextContentBlock | ClineImageContentBlock)[]
+					toolMessages: ClineUserToolResultContentBlock[]
 				}>(
 					(acc, part) => {
 						if (part.type === "tool_result") {
@@ -38,7 +47,7 @@ export function convertToOpenAiMessages(
 				)
 
 				// Process tool result messages FIRST since they must follow the tool use messages
-				const toolResultImages: Anthropic.Messages.ImageBlockParam[] = []
+				const toolResultImages: ClineImageContentBlock[] = []
 				toolMessages.forEach((toolMessage) => {
 					// The Anthropic SDK allows tool results to be a string or an array of text and image blocks, enabling rich and structured content. In contrast, the OpenAI SDK only supports tool results as a single string, so we map the Anthropic tool result parts into one concatenated string to maintain compatibility.
 					let content: string
@@ -102,8 +111,13 @@ export function convertToOpenAiMessages(
 				}
 			} else if (anthropicMessage.role === "assistant") {
 				const { nonToolMessages, toolMessages } = anthropicMessage.content.reduce<{
-					nonToolMessages: (Anthropic.TextBlockParam | Anthropic.ImageBlockParam)[]
-					toolMessages: Anthropic.ToolUseBlockParam[]
+					nonToolMessages: (
+						| ClineTextContentBlock
+						| ClineImageContentBlock
+						| ClineAssistantThinkingBlock
+						| ClineAssistantRedactedThinkingBlock
+					)[]
+					toolMessages: ClineAssistantToolUseBlock[]
 				}>(
 					(acc, part) => {
 						if (part.type === "tool_use") {
@@ -119,6 +133,7 @@ export function convertToOpenAiMessages(
 				// Process non-tool messages
 				let content: string | undefined
 				const reasoningDetails: any[] = []
+				const thinkingBlock = []
 				if (nonToolMessages.length > 0) {
 					nonToolMessages.forEach((part) => {
 						// @ts-ignore-next-line
@@ -134,13 +149,16 @@ export function convertToOpenAiMessages(
 							// @ts-ignore-next-line
 							// delete part.reasoning_details
 						}
+						if (part.type === "thinking" && part.thinking) {
+							thinkingBlock.push(part)
+						}
 					})
 					content = nonToolMessages
 						.map((part) => {
-							if (part.type === "image") {
-								return "" // impossible as the assistant cannot send images
+							if (part.type === "text" && part.text) {
+								return part.text
 							}
-							return part.text
+							return ""
 						})
 						.join("\n")
 				}

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

@@ -138,6 +138,10 @@ export async function createOpenRouterStream(
 		topP = 0.95
 		openAiMessages = convertToR1Format([{ role: "user", content: systemPrompt }, ...messages])
 	}
+	if (model.id.startsWith("google/gemini-3.0") || model.id === "google/gemini-3.0") {
+		// Recommended value from google
+		temperature = 1.0
+	}
 
 	let reasoning: { max_tokens: number } | undefined
 	switch (model.id) {

src/core/api/transform/stream.ts

@@ -1,37 +1,20 @@
-export type ApiStream = AsyncGenerator<ApiStreamChunk>
-export type ApiStreamChunk =
-	| ApiStreamTextChunk
-	| ApiStreamReasoningChunk
-	| ApiStreamReasoningDetailsChunk
-	| ApiStreamAnthropicThinkingChunk
-	| ApiStreamAnthropicRedactedThinkingChunk
-	| ApiStreamUsageChunk
-	| ApiStreamToolCallsChunk
+export type ApiStream = AsyncGenerator<ApiStreamChunk> & { id?: string }
+export type ApiStreamChunk = ApiStreamTextChunk | ApiStreamThinkingChunk | ApiStreamUsageChunk | ApiStreamToolCallsChunk
 
 export interface ApiStreamTextChunk {
 	type: "text"
+	/**
+	 * Text content generated by the model
+	 */
 	text: string
-}
-
-export interface ApiStreamReasoningChunk {
-	type: "reasoning"
-	reasoning: string
-}
-
-export interface ApiStreamReasoningDetailsChunk {
-	type: "reasoning_details"
-	reasoning_details: any // openrouter has various properties that we can pass back unmodified in api requests to preserve reasoning traces
-}
-
-export interface ApiStreamAnthropicThinkingChunk {
-	type: "ant_thinking"
-	thinking: string
-	signature: string
-}
-
-export interface ApiStreamAnthropicRedactedThinkingChunk {
-	type: "ant_redacted_thinking"
-	data: string
+	/**
+	 * The response ID associated with this chunk
+	 */
+	id?: string
+	/**
+	 * The thought signature associated with this chunk used by Gemini
+	 */
+	signature?: string
 }
 
 export interface ApiStreamUsageChunk {
@@ -42,19 +25,74 @@ export interface ApiStreamUsageChunk {
 	cacheReadTokens?: number
 	thoughtsTokenCount?: number // openrouter
 	totalCost?: number // openrouter
+	/**
+	 * The response ID associated with this response
+	 */
+	id?: string
 }
 
 export interface ApiStreamToolCallsChunk {
 	type: "tool_calls"
+	/**
+	 * The tool call information
+	 */
 	tool_call: ApiStreamToolCall
+	/**
+	 * The response ID associated with this chunk
+	 */
+	id?: string
+	/**
+	 * The thought signature associated with this chunk used by Gemini
+	 */
+	signature?: string
 }
 
 export interface ApiStreamToolCall {
-	call_id?: string // The call / request ID associated with this tool call
+	/**
+	 * The call ID associated with this tool call
+	 */
+	call_id?: string
 	// Information about the tool being called
 	function: {
-		id?: string // The tool call ID
+		/**
+		 * The tool call ID
+		 */
+		id?: string
+		/**
+		 * Name of the tool
+		 */
 		name?: string
+		/**
+		 * The arguments passed to the tool execution
+		 */
 		arguments?: any
 	}
 }
+
+export interface ApiStreamThinkingChunk {
+	type: "reasoning"
+	/**
+	 * The reasoning text generated by the model.
+	 * Redacted reasoning block will have this field set to "[REDACTED]" or an empty string.
+	 */
+	reasoning: string
+	/**
+	 * openrouter has various properties that we can pass back unmodified in api requests to preserve reasoning traces
+	 * This is also where we store the summary details for OpenAI.
+	 */
+	details?: unknown
+	/**
+	 * It's used when sending the thinking block back to the  API.
+	 * API expects this in completed form, not as array of deltas.
+	 * Also used by Gemini for thought signature associated with this chunk
+	 */
+	signature?: string
+	/**
+	 * redacted data
+	 */
+	redacted_data?: string
+	/**
+	 * The response ID associated with this chunk
+	 */
+	id?: string
+}

src/core/api/transform/tool-call-processor.ts

@@ -79,7 +79,7 @@ export function getOpenAIToolParams(tools?: OpenAITool[]) {
 		? {
 				tools,
 				tool_choice: tools ? ("auto" as ChatCompletionToolChoiceOption) : undefined,
-				parallel_tool_calls: tools ? true : undefined,
+				parallel_tool_calls: tools ? false : undefined, // Set to false to force single tool calls
 			}
 		: {
 				tools: undefined,

src/core/api/transform/tool-use-handler.ts

@@ -1,8 +1,8 @@
-import { Anthropic } from "@anthropic-ai/sdk"
 import type { ToolUse } from "@core/assistant-message"
 import { JSONParser } from "@streamparser/json"
 import { McpHub } from "@/services/mcp/McpHub"
 import { CLINE_MCP_TOOL_IDENTIFIER } from "@/shared/mcp"
+import { ClineAssistantToolUseBlock } from "@/shared/messages/content"
 import { ClineDefaultTool } from "@/shared/tools"
 
 export interface PendingToolUse {
@@ -10,6 +10,7 @@ export interface PendingToolUse {
 	name: string
 	input: string
 	parsedInput?: unknown
+	signature?: string
 	jsonParser?: JSONParser
 	call_id?: string
 }
@@ -19,6 +20,7 @@ interface ToolUseDeltaBlock {
 	type?: string
 	name?: string
 	input?: string
+	signature?: string
 }
 
 const ESCAPE_MAP: Record<string, string> = {
@@ -32,7 +34,7 @@ const ESCAPE_MAP: Record<string, string> = {
 const ESCAPE_PATTERN = /\\[ntr"\\]/g
 
 /**
- * Handles streaming native tool use blocks and converts them to Anthropic.ToolUseBlockParam format
+ * Handles streaming native tool use blocks and converts them to ClineAssistantToolUseBlock format
  */
 export class ToolUseHandler {
 	private pendingToolUses = new Map<string, PendingToolUse>()
@@ -50,17 +52,22 @@ export class ToolUseHandler {
 		if (delta.name) {
 			pending.name = delta.name
 		}
+
+		if (delta.signature) {
+			pending.signature = delta.signature
+		}
+
 		if (delta.input) {
 			pending.input += delta.input
 			try {
 				pending.jsonParser?.write(delta.input)
 			} catch {
-				// Expected during streaming
+				// Expected during streaming - JSONParser may not have complete JSON yet
 			}
 		}
 	}
 
-	getFinalizedToolUse(id: string): Anthropic.ToolUseBlockParam | undefined {
+	getFinalizedToolUse(id: string): ClineAssistantToolUseBlock | undefined {
 		const pending = this.pendingToolUses.get(id)
 		if (!pending?.name) {
 			return undefined
@@ -82,11 +89,12 @@ export class ToolUseHandler {
 			id: pending.id,
 			name: pending.name,
 			input,
+			signature: pending.signature,
 		}
 	}
 
-	getAllFinalizedToolUses(): Anthropic.ToolUseBlockParam[] {
-		const results: Anthropic.ToolUseBlockParam[] = []
+	getAllFinalizedToolUses(): ClineAssistantToolUseBlock[] {
+		const results: ClineAssistantToolUseBlock[] = []
 		for (const id of this.pendingToolUses.keys()) {
 			const toolUse = this.getFinalizedToolUse(id)
 			if (toolUse) {
@@ -102,19 +110,24 @@ export class ToolUseHandler {
 
 	getPartialToolUsesAsContent(): ToolUse[] {
 		const results: ToolUse[] = []
+		const pendingToolUses = this.pendingToolUses.values()
 
-		for (const pending of this.pendingToolUses.values()) {
+		for (const pending of pendingToolUses) {
 			if (!pending.name) {
 				continue
 			}
 
+			// Try to get the most up-to-date parsed input
+			// Priority: parsedInput (from JSONParser) > fallback to manual parsing
 			let input: any = {}
 			if (pending.parsedInput != null) {
 				input = pending.parsedInput
 			} else if (pending.input) {
+				// Try full JSON parse first
 				try {
 					input = JSON.parse(pending.input)
 				} catch {
+					// Fall back to extracting partial fields from incomplete JSON
 					input = this.extractPartialJsonFields(pending.input)
 				}
 			}
@@ -131,10 +144,11 @@ export class ToolUseHandler {
 					},
 					partial: true,
 					isNativeToolCall: true,
+					signature: pending.signature,
 				})
 			} else {
 				const params: Record<string, string> = {}
-				if (typeof input === "object") {
+				if (typeof input === "object" && input !== null) {
 					for (const [key, value] of Object.entries(input)) {
 						params[key] = typeof value === "string" ? value : JSON.stringify(value)
 					}
@@ -144,12 +158,13 @@ export class ToolUseHandler {
 					name: pending.name as ClineDefaultTool,
 					params: params as any,
 					partial: true,
+					signature: pending.signature,
 					isNativeToolCall: true,
 				})
 			}
 		}
-
-		return results
+		// Ensure all returned tool uses are marked as partial
+		return results.map((t) => ({ ...t, partial: true }))
 	}
 
 	reset(): void {
@@ -165,6 +180,7 @@ export class ToolUseHandler {
 			parsedInput: undefined,
 			jsonParser,
 			call_id,
+			signature: undefined,
 		}
 
 		jsonParser.onValue = (info: any) => {

src/core/assistant-message/index.ts

@@ -1,9 +1,10 @@
 import { ClineDefaultTool } from "@shared/tools"
-export type AssistantMessageContent = TextContent | ToolUse
+
+export type AssistantMessageContent = TextStreamContent | ToolUse | ReasoningStreamContent
 
 export { parseAssistantMessageV2 } from "./parse-assistant-message"
 
-export interface TextContent {
+export interface TextStreamContent {
 	type: "text"
 	content: string
 	partial: boolean
@@ -53,4 +54,15 @@ export interface ToolUse {
 	partial: boolean
 	// Whether this tool use was initiated by a native tool call
 	isNativeToolCall?: boolean
+	signature?: string
+}
+
+export interface ReasoningStreamContent {
+	type: "reasoning"
+	reasoning: string
+	details?: any // openrouter has various properties that we can pass back unmodified in api requests to preserve reasoning traces
+	signature?: string
+	redacted?: boolean // whether this reasoning block has been redacted
+	data?: string // redacted data
+	partial: boolean
 }

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

@@ -1,5 +1,5 @@
 import { ClineDefaultTool, toolUseNames } from "@shared/tools"
-import { AssistantMessageContent, TextContent, ToolParamName, ToolUse, toolParamNames } from "." // Assuming types are defined in index.ts or a similar file
+import { AssistantMessageContent, TextStreamContent, ToolParamName, ToolUse, toolParamNames } from "." // Assuming types are defined in index.ts or a similar file
 
 // parseAssistantmessageV1 removed in https://github.com/cline/cline/pull/5425
 
@@ -27,7 +27,7 @@ import { AssistantMessageContent, TextContent, ToolParamName, ToolUse, toolParam
 export function parseAssistantMessageV2(assistantMessage: string): AssistantMessageContent[] {
 	const contentBlocks: AssistantMessageContent[] = []
 	let currentTextContentStart = 0 // Index where the current text block started
-	let currentTextContent: TextContent | undefined
+	let currentTextContent: TextStreamContent | undefined
 	let currentToolUseStart = 0 // Index *after* the opening tag of the current tool use
 	let currentToolUse: ToolUse | undefined
 	let currentParamValueStart = 0 // Index *after* the opening tag of the current param

src/core/context/context-management/context-error-handling.ts

@@ -5,7 +5,8 @@ export function checkContextWindowExceededError(error: unknown): boolean {
 		checkIsOpenAIContextWindowError(error) ||
 		checkIsOpenRouterContextWindowError(error) ||
 		checkIsAnthropicContextWindowError(error) ||
-		checkIsCerebrasContextWindowError(error)
+		checkIsCerebrasContextWindowError(error) ||
+		checkIsBedrockContextWindowError(error)
 	)
 }
 
@@ -70,3 +71,45 @@ function checkIsCerebrasContextWindowError(response: any): boolean {
 		return false
 	}
 }
+
+function checkIsBedrockContextWindowError(error: any): boolean {
+	try {
+		// Bedrock returns ValidationException for context window errors
+		const errorType = error?.name ?? error?.error?.type ?? error?.__type
+		const errorCode = error?.code ?? error?.error?.code ?? error?.$metadata?.httpStatusCode
+
+		// Handle nested error structures (e.g., through Vercel AI SDK)
+		const nestedError = error?.error?.param
+		const nestedErrorCode = nestedError?.statusCode ?? error?.details?.code
+		const nestedMessage = nestedError?.message ?? nestedError?.error
+
+		const message: string = String(error?.message || error?.error?.message || nestedMessage || "")
+
+		// Check for ValidationException with HTTP 400
+		const isValidationException =
+			errorType === "ValidationException" ||
+			errorType === "AI_APICallError" ||
+			String(errorCode) === "400" ||
+			String(nestedErrorCode) === "400" ||
+			error?.code === "stream_initialization_failed"
+
+		if (!isValidationException) {
+			return false
+		}
+
+		// Known Bedrock context window error patterns
+		const BEDROCK_CONTEXT_PATTERNS = [
+			/maximum tokens.*exceeds.*model limit/i,
+			/input length and max_tokens exceed context limit/i,
+			/context length.*exceeds/i,
+			/total number of tokens.*exceeds.*limit/i,
+			/requested.*tokens.*exceeds.*limit/i,
+			/reduce.*length.*messages.*completion/i,
+			/input is too long/i,
+		] as const
+
+		return BEDROCK_CONTEXT_PATTERNS.some((pattern) => pattern.test(message))
+	} catch {
+		return false
+	}
+}

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

@@ -1,6 +1,7 @@
 import { getRuleFilesTotalContent, synchronizeRuleToggles } from "@core/context/instructions/user-instructions/rule-helpers"
 import { formatResponse } from "@core/prompts/responses"
 import { ensureRulesDirectoryExists, GlobalFileNames } from "@core/storage/disk"
+import { StateManager } from "@core/storage/StateManager"
 import { ClineRulesToggles } from "@shared/cline-rules"
 import { fileExistsAtPath, isDirectory, readDirectory } from "@utils/fs"
 import fs from "fs/promises"
@@ -8,27 +9,48 @@ import path from "path"
 import { Controller } from "@/core/controller"
 
 export const getGlobalClineRules = async (globalClineRulesFilePath: string, toggles: ClineRulesToggles) => {
+	let combinedContent = ""
+
+	// 1. Get file-based rules
 	if (await fileExistsAtPath(globalClineRulesFilePath)) {
 		if (await isDirectory(globalClineRulesFilePath)) {
 			try {
 				const rulesFilePaths = await readDirectory(globalClineRulesFilePath)
 				const rulesFilesTotalContent = await getRuleFilesTotalContent(rulesFilePaths, globalClineRulesFilePath, toggles)
 				if (rulesFilesTotalContent) {
-					const clineRulesFileInstructions = formatResponse.clineRulesGlobalDirectoryInstructions(
-						globalClineRulesFilePath,
-						rulesFilesTotalContent,
-					)
-					return clineRulesFileInstructions
+					combinedContent = rulesFilesTotalContent
 				}
 			} catch {
 				console.error(`Failed to read .clinerules directory at ${globalClineRulesFilePath}`)
 			}
 		} else {
 			console.error(`${globalClineRulesFilePath} is not a directory`)
-			return undefined
 		}
 	}
 
+	// 2. Append remote config rules
+	const stateManager = StateManager.get()
+	const remoteConfigSettings = stateManager.getRemoteConfigSettings()
+	const remoteRules = remoteConfigSettings.remoteGlobalRules || []
+	const remoteToggles = stateManager.getGlobalStateKey("remoteRulesToggles") || {}
+
+	for (const rule of remoteRules) {
+		// If alwaysEnabled, always include; otherwise check toggle
+		const isEnabled = rule.alwaysEnabled || remoteToggles[rule.name] !== false
+
+		if (isEnabled) {
+			if (combinedContent) {
+				combinedContent += "\n\n"
+			}
+			combinedContent += `${rule.name}\n${rule.contents}`
+		}
+	}
+
+	// 3. Return formatted instructions
+	if (combinedContent) {
+		return formatResponse.clineRulesGlobalDirectoryInstructions(globalClineRulesFilePath, combinedContent)
+	}
+
 	return undefined
 }
 

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

@@ -6,6 +6,7 @@ import {
 } from "@core/context/instructions/user-instructions/rule-helpers"
 import { formatResponse } from "@core/prompts/responses"
 import { GlobalFileNames } from "@core/storage/disk"
+import { listFiles } from "@services/glob/list-files"
 import { ClineRulesToggles } from "@shared/cline-rules"
 import { fileExistsAtPath, isDirectory } from "@utils/fs"
 import fs from "fs/promises"
@@ -13,7 +14,7 @@ import path from "path"
 import { Controller } from "@/core/controller"
 
 /**
- * Refreshes the toggles for windsurf and cursor rules
+ * Refreshes the toggles for windsurf, cursor, and agents rules
  */
 export async function refreshExternalRulesToggles(
 	controller: Controller,
@@ -21,6 +22,7 @@ export async function refreshExternalRulesToggles(
 ): Promise<{
 	windsurfLocalToggles: ClineRulesToggles
 	cursorLocalToggles: ClineRulesToggles
+	agentsLocalToggles: ClineRulesToggles
 }> {
 	// local windsurf toggles
 	const localWindsurfRulesToggles = controller.stateManager.getWorkspaceStateKey("localWindsurfRulesToggles")
@@ -42,9 +44,16 @@ export async function refreshExternalRulesToggles(
 	const updatedLocalCursorToggles = combineRuleToggles(updatedLocalCursorToggles1, updatedLocalCursorToggles2)
 	controller.stateManager.setWorkspaceState("localCursorRulesToggles", updatedLocalCursorToggles)
 
+	// local agents toggles
+	const localAgentsRulesToggles = controller.stateManager.getWorkspaceStateKey("localAgentsRulesToggles")
+	const localAgentsRulesFilePath = path.resolve(workingDirectory, GlobalFileNames.agentsRulesFile)
+	const updatedLocalAgentsToggles = await synchronizeRuleToggles(localAgentsRulesFilePath, localAgentsRulesToggles)
+	controller.stateManager.setWorkspaceState("localAgentsRulesToggles", updatedLocalAgentsToggles)
+
 	return {
 		windsurfLocalToggles: updatedLocalWindsurfToggles,
 		cursorLocalToggles: updatedLocalCursorToggles,
+		agentsLocalToggles: updatedLocalAgentsToggles,
 	}
 }
 
@@ -117,3 +126,76 @@ export const getLocalCursorRules = async (cwd: string, toggles: ClineRulesToggle
 
 	return [cursorRulesFileInstructions, cursorRulesDirInstructions]
 }
+
+/**
+ * Helper function to find all agents.md files recursively (case-insensitive)
+ * Only searches if a top-level agents.md file exists
+ */
+async function findAgentsMdFiles(cwd: string): Promise<string[]> {
+	try {
+		// First check if top-level agents.md exists
+		const topLevelAgentsPath = path.resolve(cwd, GlobalFileNames.agentsRulesFile)
+		const topLevelExists = await fileExistsAtPath(topLevelAgentsPath)
+
+		// Only search recursively if top-level agents.md exists
+		if (!topLevelExists) {
+			return []
+		}
+
+		// Search recursively for all agents.md files
+		const [allFiles] = await listFiles(cwd, true, 500)
+		return allFiles.filter((filePath) => {
+			const basename = path.basename(filePath).toLowerCase()
+			return basename === GlobalFileNames.agentsRulesFile.toLowerCase()
+		})
+	} catch (error) {
+		console.error(`Failed to find agents.md files in ${cwd}:`, error)
+		return []
+	}
+}
+
+/**
+ * Gather formatted agents rules - searches recursively and combines all agents.md files
+ */
+export const getLocalAgentsRules = async (cwd: string, toggles: ClineRulesToggles) => {
+	const agentsRulesFilePath = path.resolve(cwd, GlobalFileNames.agentsRulesFile)
+
+	// Check if the top-level agents.md file is enabled
+	if (agentsRulesFilePath in toggles && toggles[agentsRulesFilePath] === false) {
+		return undefined
+	}
+
+	try {
+		const agentsMdFiles = await findAgentsMdFiles(cwd)
+
+		if (agentsMdFiles.length === 0) {
+			return undefined
+		}
+
+		// Read and combine all agents.md files
+		const combinedContent = await Promise.all(
+			agentsMdFiles.map(async (filePath) => {
+				try {
+					const fullPath = path.resolve(cwd, filePath)
+					const content = (await fs.readFile(fullPath, "utf8")).trim()
+					if (content) {
+						const relativePath = path.relative(cwd, fullPath)
+						return `## ${relativePath}\n\n${content}`
+					}
+					return null
+				} catch (error) {
+					console.error(`Failed to read agents.md file at ${filePath}:`, error)
+					return null
+				}
+			}),
+		).then((contents) => contents.filter(Boolean).join("\n\n"))
+
+		if (combinedContent) {
+			return formatResponse.agentsRulesLocalFileInstructions(cwd, combinedContent)
+		}
+	} catch (error) {
+		console.error("Failed to read agents.md files:", error)
+	}
+
+	return undefined
+}

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

@@ -1,5 +1,6 @@
 import { ensureRulesDirectoryExists, ensureWorkflowsDirectoryExists, GlobalFileNames } from "@core/storage/disk"
 import { ClineRulesToggles } from "@shared/cline-rules"
+import { GlobalInstructionsFile } from "@shared/remote-config/schema"
 import { fileExistsAtPath, isDirectory, readDirectory } from "@utils/fs"
 import fs from "fs/promises"
 import * as path from "path"
@@ -101,6 +102,36 @@ export async function synchronizeRuleToggles(
 	return updatedToggles
 }
 
+/**
+ * Synchronizes remote rule toggles with current remote config
+ * Removes toggles for rules that no longer exist, adds defaults for new rules
+ */
+export function synchronizeRemoteRuleToggles(
+	remoteRules: GlobalInstructionsFile[],
+	currentToggles: ClineRulesToggles,
+): ClineRulesToggles {
+	const updatedToggles: ClineRulesToggles = {}
+
+	// Create set of current remote rule names
+	const existingRuleNames = new Set(remoteRules.map((rule) => rule.name))
+
+	// Keep toggles only for rules that still exist
+	for (const [ruleName, enabled] of Object.entries(currentToggles)) {
+		if (existingRuleNames.has(ruleName)) {
+			updatedToggles[ruleName] = enabled
+		}
+	}
+
+	// Add default toggles for new rules (default to enabled)
+	for (const rule of remoteRules) {
+		if (!(rule.name in updatedToggles)) {
+			updatedToggles[rule.name] = true
+		}
+	}
+
+	return updatedToggles
+}
+
 /**
  * Certain project rules have more than a single location where rules are allowed to be stored
  */
@@ -268,6 +299,10 @@ export async function deleteRuleFile(
 				const toggles = controller.stateManager.getWorkspaceStateKey("localWindsurfRulesToggles")
 				delete toggles[rulePath]
 				controller.stateManager.setWorkspaceState("localWindsurfRulesToggles", toggles)
+			} else if (type === "agents") {
+				const toggles = controller.stateManager.getWorkspaceStateKey("localAgentsRulesToggles")
+				delete toggles[rulePath]
+				controller.stateManager.setWorkspaceState("localAgentsRulesToggles", toggles)
 			} else {
 				const toggles = controller.stateManager.getWorkspaceStateKey("localClineRulesToggles")
 				delete toggles[rulePath]

src/core/controller/account/requestyAuthClicked.ts

@@ -1,6 +1,6 @@
 import { Empty, StringRequest } from "@shared/proto/cline/common"
-import { toRequestyServiceUrl } from "@shared/providers/requesty"
 import { HostProvider } from "@/hosts/host-provider"
+import { toRequestyServiceUrl } from "@/shared/clients/requesty"
 import { openExternal } from "@/utils/env"
 import { Controller } from ".."
 

src/core/controller/file/refreshRules.ts

@@ -16,7 +16,10 @@ export async function refreshRules(controller: Controller, _request: EmptyReques
 	try {
 		const cwd = await getCwd(getDesktopDir())
 		const { globalToggles, localToggles } = await refreshClineRulesToggles(controller, cwd)
-		const { cursorLocalToggles, windsurfLocalToggles } = await refreshExternalRulesToggles(controller, cwd)
+		const { cursorLocalToggles, windsurfLocalToggles, agentsLocalToggles } = await refreshExternalRulesToggles(
+			controller,
+			cwd,
+		)
 		const { localWorkflowToggles, globalWorkflowToggles } = await refreshWorkflowToggles(controller, cwd)
 
 		return RefreshedRules.create({
@@ -24,6 +27,7 @@ export async function refreshRules(controller: Controller, _request: EmptyReques
 			localClineRulesToggles: { toggles: localToggles },
 			localCursorRulesToggles: { toggles: cursorLocalToggles },
 			localWindsurfRulesToggles: { toggles: windsurfLocalToggles },
+			localAgentsRulesToggles: { toggles: agentsLocalToggles },
 			localWorkflowToggles: { toggles: localWorkflowToggles },
 			globalWorkflowToggles: { toggles: globalWorkflowToggles },
 		})

src/core/controller/file/toggleAgentsRule.ts

@@ -0,0 +1,33 @@
+import type { ToggleAgentsRuleRequest } from "@shared/proto/cline/file"
+import { ClineRulesToggles } from "@shared/proto/cline/file"
+import type { Controller } from "../index"
+
+/**
+ * Toggles an Agents rule (enable or disable)
+ * @param controller The controller instance
+ * @param request The toggle request
+ * @returns The updated Agents rule toggles
+ */
+export async function toggleAgentsRule(controller: Controller, request: ToggleAgentsRuleRequest): Promise<ClineRulesToggles> {
+	const { rulePath, enabled } = request
+
+	if (!rulePath || typeof enabled !== "boolean") {
+		console.error("toggleAgentsRule: Missing or invalid parameters", {
+			rulePath,
+			enabled: typeof enabled === "boolean" ? enabled : `Invalid: ${typeof enabled}`,
+		})
+		throw new Error("Missing or invalid parameters for toggleAgentsRule")
+	}
+
+	// Update the toggle in workspace state
+	const toggles = controller.stateManager.getWorkspaceStateKey("localAgentsRulesToggles")
+	toggles[rulePath] = enabled
+	controller.stateManager.setWorkspaceState("localAgentsRulesToggles", toggles)
+
+	// Get the current state to return in the response
+	const agentsToggles = controller.stateManager.getWorkspaceStateKey("localAgentsRulesToggles")
+
+	return ClineRulesToggles.create({
+		toggles: agentsToggles,
+	})
+}

src/core/controller/file/toggleClineRule.ts

@@ -1,6 +1,6 @@
 import { getWorkspaceBasename } from "@core/workspace"
 import type { ToggleClineRuleRequest } from "@shared/proto/cline/file"
-import { ToggleClineRules } from "@shared/proto/cline/file"
+import { RuleScope, ToggleClineRules } from "@shared/proto/cline/file"
 import { telemetryService } from "@/services/telemetry"
 import type { Controller } from "../index"
 
@@ -11,41 +11,57 @@ import type { Controller } from "../index"
  * @returns The updated Cline rule toggles
  */
 export async function toggleClineRule(controller: Controller, request: ToggleClineRuleRequest): Promise<ToggleClineRules> {
-	const { isGlobal, rulePath, enabled } = request
+	const { scope, rulePath, enabled } = request
 
-	if (!rulePath || typeof enabled !== "boolean" || typeof isGlobal !== "boolean") {
+	if (!rulePath || typeof enabled !== "boolean" || scope === undefined) {
 		console.error("toggleClineRule: Missing or invalid parameters", {
 			rulePath,
-			isGlobal: typeof isGlobal === "boolean" ? isGlobal : `Invalid: ${typeof isGlobal}`,
+			scope,
 			enabled: typeof enabled === "boolean" ? enabled : `Invalid: ${typeof enabled}`,
 		})
 		throw new Error("Missing or invalid parameters for toggleClineRule")
 	}
 
-	// This is the same core logic as in the original handler
-	if (isGlobal) {
-		const toggles = controller.stateManager.getGlobalSettingsKey("globalClineRulesToggles")
-		toggles[rulePath] = enabled
-		controller.stateManager.setGlobalState("globalClineRulesToggles", toggles)
-	} else {
-		const toggles = controller.stateManager.getWorkspaceStateKey("localClineRulesToggles")
-		toggles[rulePath] = enabled
-		controller.stateManager.setWorkspaceState("localClineRulesToggles", toggles)
+	// Handle the three different scopes
+	switch (scope) {
+		case RuleScope.GLOBAL: {
+			const toggles = controller.stateManager.getGlobalSettingsKey("globalClineRulesToggles")
+			toggles[rulePath] = enabled
+			controller.stateManager.setGlobalState("globalClineRulesToggles", toggles)
+			break
+		}
+		case RuleScope.LOCAL: {
+			const toggles = controller.stateManager.getWorkspaceStateKey("localClineRulesToggles")
+			toggles[rulePath] = enabled
+			controller.stateManager.setWorkspaceState("localClineRulesToggles", toggles)
+			break
+		}
+		case RuleScope.REMOTE: {
+			const toggles = controller.stateManager.getGlobalStateKey("remoteRulesToggles")
+			toggles[rulePath] = enabled
+			controller.stateManager.setGlobalState("remoteRulesToggles", toggles)
+			break
+		}
+		default:
+			throw new Error(`Invalid scope: ${scope}`)
 	}
 
 	// Track rule toggle telemetry with current task context
 	if (controller.task?.ulid) {
 		// Extract just the filename for privacy (no full paths)
 		const ruleFileName = getWorkspaceBasename(rulePath, "Controller.toggleClineRule")
+		const isGlobal = scope === RuleScope.GLOBAL
 		telemetryService.captureClineRuleToggled(controller.task.ulid, ruleFileName, enabled, isGlobal)
 	}
 
 	// Get the current state to return in the response
 	const globalToggles = controller.stateManager.getGlobalSettingsKey("globalClineRulesToggles")
 	const localToggles = controller.stateManager.getWorkspaceStateKey("localClineRulesToggles")
+	const remoteToggles = controller.stateManager.getGlobalStateKey("remoteRulesToggles")
 
 	return ToggleClineRules.create({
 		globalClineRulesToggles: { toggles: globalToggles },
 		localClineRulesToggles: { toggles: localToggles },
+		remoteRulesToggles: { toggles: remoteToggles },
 	})
 }

src/core/controller/file/toggleWorkflow.ts

@@ -1,4 +1,4 @@
-import { ClineRulesToggles, ToggleWorkflowRequest } from "@shared/proto/cline/file"
+import { ClineRulesToggles, RuleScope, ToggleWorkflowRequest } from "@shared/proto/cline/file"
 import { Controller } from ".."
 
 /**
@@ -8,34 +8,45 @@ import { Controller } from ".."
  * @returns The updated workflow toggles
  */
 export async function toggleWorkflow(controller: Controller, request: ToggleWorkflowRequest): Promise<ClineRulesToggles> {
-	const { workflowPath, enabled, isGlobal } = request
+	const { workflowPath, enabled, scope } = request
 
-	if (!workflowPath || typeof enabled !== "boolean") {
+	if (!workflowPath || typeof enabled !== "boolean" || scope === undefined) {
 		console.error("toggleWorkflow: Missing or invalid parameters", {
 			workflowPath,
+			scope,
 			enabled: typeof enabled === "boolean" ? enabled : `Invalid: ${typeof enabled}`,
 		})
 		throw new Error("Missing or invalid parameters for toggleWorkflow")
 	}
 
-	// Update the toggles based on isGlobal flag
-	if (isGlobal) {
-		// Global workflows
-		const toggles = controller.stateManager.getGlobalSettingsKey("globalWorkflowToggles")
-		toggles[workflowPath] = enabled
-		controller.stateManager.setGlobalState("globalWorkflowToggles", toggles)
-		await controller.postStateToWebview()
+	// Handle the three different scopes
+	let toggles: Record<string, boolean>
 
-		// Return the global toggles
-		return ClineRulesToggles.create({ toggles: toggles })
-	} else {
-		// Workspace workflows
-		const toggles = controller.stateManager.getWorkspaceStateKey("workflowToggles")
-		toggles[workflowPath] = enabled
-		controller.stateManager.setWorkspaceState("workflowToggles", toggles)
-		await controller.postStateToWebview()
-
-		// Return the workspace toggles
-		return ClineRulesToggles.create({ toggles: toggles })
+	switch (scope) {
+		case RuleScope.GLOBAL: {
+			toggles = controller.stateManager.getGlobalSettingsKey("globalWorkflowToggles")
+			toggles[workflowPath] = enabled
+			controller.stateManager.setGlobalState("globalWorkflowToggles", toggles)
+			break
+		}
+		case RuleScope.LOCAL: {
+			toggles = controller.stateManager.getWorkspaceStateKey("workflowToggles")
+			toggles[workflowPath] = enabled
+			controller.stateManager.setWorkspaceState("workflowToggles", toggles)
+			break
+		}
+		case RuleScope.REMOTE: {
+			toggles = controller.stateManager.getGlobalStateKey("remoteWorkflowToggles")
+			toggles[workflowPath] = enabled
+			controller.stateManager.setGlobalState("remoteWorkflowToggles", toggles)
+			break
+		}
+		default:
+			throw new Error(`Invalid scope: ${scope}`)
 	}
+
+	await controller.postStateToWebview()
+
+	// Return the updated toggles
+	return ClineRulesToggles.create({ toggles: toggles })
 }