workbuddy-mcp
Enables Codex to hand off tasks to WorkBuddy for document generation, data analysis, and more via file-based handoff and deep links.
README
Codex ↔ WorkBuddy 桥接(MCP server)
让 OpenAI Codex 在执行任务时把子任务"交接"给 WorkBuddy(生成文档 / PPT / 数据分析 / 研究报告等)。
为什么需要桥接
调查结论(已在本机核实):
- Codex 支持 MCP:配置在
~/.codex/config.toml,用[mcp_servers.xxx]表,STDIO 模式(command+args+env)。你机器上已在用tradingview、node_repl等 server。 - WorkBuddy 目前不原生作为 MCP server,也没有对外"提交任务并带回结果"的公开 API。它内部的
createTask / submitTask / runAgent是 Electron 主进程私有方法,外部不可达。 - WorkBuddy 的
workbuddy://Deep Link(workbuddy://home、workbuddy://chat/<id>、workbuddy://my-files?tab=cloudFiles等)只做 UI 导航,不提交任务。
因此采用 handoff(任务交接) 模式:Codex 调 MCP 工具 → 打开 WorkBuddy + 把任务写进共享收件箱 + 复制到剪贴板 → WorkBuddy 在界面里执行。待 WorkBuddy 未来暴露 MCP server / 任务 API 后,可无缝升级为"直接调用并拿回结果"。
文件
server.js—— stdio MCP server,暴露两个工具:workbuddy_handoff(prompt, title?):交接一项子任务给 WorkBuddy。workbuddy_open(target):打开 WorkBuddy 的某个界面(home / my-files)。
test-client.mjs—— 本机冒烟测试(node test-client.mjs <node路径>)。node_modules/—— 依赖(已装在隔离环境,不污染系统)。
在 Codex 中注册
把下面这段加进 ~/.codex/config.toml(与现有 [mcp_servers.tradingview] 并列):
[mcp_servers.workbuddy]
command = "/Users/zhaoxiangyu/.workbuddy/binaries/node/versions/22.22.2/bin/node"
args = ["/Users/zhaoxiangyu/WorkBuddy/2026-07-20-17-51-22/codex-workbuddy-bridge/server.js"]
env = { WORKBUDDY_INBOX_DIR = "/Users/zhaoxiangyu/WorkBuddy/inbox" }
重启 Codex(或在 TUI 里 /mcp)后,Codex 就多了 workbuddy_handoff / workbuddy_open 两个工具。
也可改用 CLI:codex mcp add workbuddy -- /Users/zhaoxiangyu/.workbuddy/binaries/node/versions/22.22.2/bin/node /Users/zhaoxiangyu/WorkBuddy/2026-07-20-17-51-22/codex-workbuddy-bridge/server.js
在 Codex 里怎么用
让 Codex 在合适时机调用,例如:
把这个项目跑完单测后,用 workbuddy_handoff 让 WorkBuddy 生成一份发布说明文档。
Codex 会:打开 WorkBuddy、把任务落到 ~/WorkBuddy/inbox/,并把任务文本复制到剪贴板,方便在 WorkBuddy 里直接粘贴执行。
环境变量
WORKBUDDY_INBOX_DIR(默认~/WorkBuddy/inbox):交接任务落盘目录。WORKBUDDY_SUBMIT_DEEPLINK(可选):若你发现 WorkBuddy 支持"打开即提交"的深链(如workbuddy://chat/new?prompt={prompt}),设上它即可让 handoff 直接尝试自动提交,无需手动粘贴。
验证
cd codex-workbuddy-bridge
/Users/zhaoxiangyu/.workbuddy/binaries/node/versions/22.22.2/bin/node test-client.mjs /Users/zhaoxiangyu/.workbuddy/binaries/node/versions/22.22.2/bin/node
应看到 CONNECTED、两个工具名、handoff 返回的收件箱路径,并在 ~/WorkBuddy/inbox/ 生成对应 .md。
反向方案(可选)
若你更想"WorkBuddy 调 Codex":WorkBuddy 能消费 MCP / 运行 CLI,可把 Codex 的能力(或 codebuddy CLI,位于 WorkBuddy.app 内的 cli/bin/codebuddy)接成 WorkBuddy 的 MCP 连接器。本仓库聚焦 Codex→WorkBuddy 方向。
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.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
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.