grok-mcp
Exposes local Grok Build bridge as MCP stdio server, enabling code agents like Claude Code to run Grok models, check status, and manage runs.
README
grok-mcp
把本机 Grok Build 的 grok-bridge.mjs 暴露为 MCP stdio server，任意支持 MCP 的 code agent（Claude Code / OpenCode / …）可调用。
MCP client
│ stdio
▼
grok-mcp
│
▼
grok-bridge.mjs → grok headless
不是 Claude Code 插件本体；不打包 slash commands / hooks。只包 bridge CLI。
工具
| Tool | Bridge | 说明 |
|---|---|---|
grok_check |
check --json |
就绪探测（node / grok / auth） |
grok_run |
run … |
主路径；background / write / model / effort |
grok_runs |
runs --json |
列表或指定 run |
grok_show |
show --json |
状态/结果 |
grok_stop |
stop --json |
取消 |
grok_bridge_path |
— | 解析到的 bridge 路径 |
审阅类任务：write=false（默认），需要改文件时再 write=true。
前置
- Node ≥ 18
- 本机已装
grokCLI 并登录（grok/ bridgecheck通过） - 可找到
grok-bridge.mjs（见路径解析）
安装
cd ~/karson_ubuntu/grok-mcp
npm install
npm run build
路径解析（GROK_BRIDGE）
按顺序找 grok-bridge.mjs：
- 环境变量
GROK_BRIDGE ~/github_repo/grok-build-plugin-cc/plugins/grok-build/scripts/grok-bridge.mjs~/.claude/plugins/cache/xai-grok-build/grok-build/<ver>/scripts/grok-bridge.mjs（最新 ver）$CLAUDE_PLUGIN_ROOT/scripts/grok-bridge.mjs
接入 Claude Code
claude mcp add grok -s user \
-e GROK_BRIDGE=$HOME/github_repo/grok-build-plugin-cc/plugins/grok-build/scripts/grok-bridge.mjs \
-- node $HOME/karson_ubuntu/grok-mcp/dist/index.js
- 前台
grok_run默认超时 15 分钟（timeout_ms可改）。 - background=true：只等 bridge 入队并返回 jobId（默认最长 60s），不跑满 agent；完成靠
grok_show/ 报告文件。 - Prompt 经
--prompt-file写在 tmpdir（不污染项目 cwd）。
重启 Claude Code 后 /mcp 应见 grok。工具名形如 mcp__grok__grok_check。
接入 OpenCode
在 ~/.config/opencode/opencode.json 的 mcp 段增加等价 stdio 配置（command/args/env 同上）。
multi-model-review
skill 的 Grok 路：优先 MCP grok_run（background=true, write=false, model=grok-4.5）；无 MCP 时回退 Bash 调 bridge。
开发
npm run typecheck
npm run build
node dist/index.js # stdio MCP；不要手动交互跑
手动验 bridge（不经 MCP）：
node "$GROK_BRIDGE" check --json
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
Neon Database
MCP server for interacting with Neon Management API and databases
E2B
Using MCP to run code via e2b.
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.