{"$schema":"https://wellknown.network/schemas/agent-record-v1.json","schemaVersion":"1","id":"ag_2scdm62bkte7","handle":"micro-app-mcp","url":"https://wellknown.network/agents/micro-app-mcp","links":{"self":"https://wellknown.network/agents/micro-app-mcp/record.json","html":"https://wellknown.network/agents/micro-app-mcp","markdown":"https://wellknown.network/agents/micro-app-mcp/record.md","api":"https://wellknown.network/api/v1/agents/micro-app-mcp","status":"https://wellknown.network/api/v1/agents/micro-app-mcp/status","claim":"https://wellknown.network/agents/micro-app-mcp/claim","claimApi":"https://wellknown.network/api/v1/claims","claimDescriptor":"https://wellknown.network/agents/micro-app-mcp/claim.json","badge":"https://wellknown.network/agents/micro-app-mcp/badge.svg","openapi":"https://wellknown.network/openapi.json","history":"https://wellknown.network/api/v1/agents/micro-app-mcp/history","tools":"https://wellknown.network/api/v1/agents/micro-app-mcp/tools"},"ard":{"identifier":"urn:air::server:micro-app-mcp","type":"application/mcp-server-card+json"},"kind":"mcp_server","declared":{"name":"micro-app-mcp","summary":"MCP Server for micro-app knowledge base","description":"# Micro-App MCP Server\n\n[![Python Version](https://img.shields.io/badge/python-3.12+-blue.svg)](https://www.python.org/downloads/)\n[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](LICENSE)\n[![PyPI Version](https://img.shields.io/pypi/v/micro-app-mcp)](https://pypi.org/project/micro-app-mcp/)\n[![Release PyPI](https://github.com/didengren/micro-app-mcp/actions/workflows/release-pypi.yml/badge.svg)](https://github.com/didengren/micro-app-mcp/actions/workflows/release-pypi.yml)\n\n`micro-app-mcp` 是一个面向 **micro-app 微前端框架** 的专用知识库 MCP Server。  \n它会自动 **吃透 micro-app 仓库源码、爬取官方文档内容、用嵌入模型向量化后存入本地数据库**，并通过 MCP 协议提供语义检索与可控更新能力。\n\n---\n\n## 特性亮点\n\n- **源码级掌握 micro-app**：从官方 GitHub 仓库拉取源码，结合文档做统一知识建模。\n- **自动爬取官方文档**：利用 Playwright 渲染 + 文档爬虫，从文档站点采集完整内容。\n- **本地向量化知识库**：基于嵌入模型将文本向量化，存入本地 ChromaDB，查询低延迟、可离线可缓存。\n- **可控更新策略**：支持按需更新、强制更新与状态查询，避免每次都重建知识库。\n- **标准 MCP 接入**：开箱即用接入 Trae AI Agent、VSCode 等支持 MCP 的智能体。\n\n---\n\n## 对话场景\n\n在你的智能体里（如 Trae / VSCode MCP 插件）完成配置后，当用户输入 `/micro` 时，LLM 会先通过统一入口工具按问题类别（状态 / 更新 / 检索）识别意图后再回答问题，比如：\n\n- **查询知识库状态**：`/micro 获取知识库状态`\n- **控制知识库更新**：`/micro 更新知识库` 或 `/micro 强制更新知识库 force=true`\n- **询问 API / 配置项**：`/micro micro-app 支持哪些沙箱隔离模式？`\n- **理解源码实现**：`/micro micro-app 子应用的加载流程怎么实现的？入口源码在哪？`\n- **看版本变更 / 升级注意**：`/micro 从 vX 到 vY 有哪些 breaking changes？`\n\n---\n\n## 安装与运行（需要 Python 3.12+）\n\n### 方式 A：源码运行（推荐，已内置 CPU 优化）\n适合开发者。本项目已配置 `uv` 智能索引，在 Windows/Linux 下会自动安装轻量化 CPU 版 Torch。\n\n```bash\n# 克隆仓库\ngit clone https://github.com/didengren/micro-app-mcp.git\ncd micro-app-mcp\n\n# 1. 一键安装依赖（自动处理跨平台 Torch 索引）\nuv sync --extra dev\n\n# 2. 安装浏览器内核\nuv run python -m playwright install chromium\n\n# 3. 以 editable 方式安装本项目作为依赖模块，方便开发调试\nuv run pip install -e .\n\n# 4. 复制一份 .env 按需修改项目环境变量，如 GITHUB_TOKEN、DATA_DIR 等\ncp .env.example .env\n\n# 5. 运行\nuv run micro-app-mcp\n\n# 6. 测试\nuv run pytest\n```\n\n### 方式 B：CLI 快速运行（uvx）\n适合普通用户。需手动指定 CPU 源以减小体积：\n\n```bash\n# Linux/Windows 用户手动设置环境变量指定 CPU 源以减小体积并使用 unsafe-best-match 策略避开依赖冲突\nUV_EXTRA_INDEX_URL=\"https://download.pytorch.org/whl/cpu\" \\\nuvx…","publisher":null,"homepage":"https://github.com/didengren/micro-app-mcp","repository":"https://github.com/didengren/micro-app-mcp/blob/main/CHANGELOG.md","version":"0.1.8","license":null,"protocols":["mcp"],"tags":["mcp","micro-app","knowledge-base","ai-agent","vector-database","chromadb","model-context-protocol"],"pricing":null,"endpoints":[{"url":"pypi:micro-app-mcp","type":"package_pypi","auth":null,"probeable":false}],"skills":null,"tools":null,"extra":null,"attribution":{"kind":"pypi","name":"pypi","repoUrl":"pypi","summary":"pypi","version":"pypi","description":"pypi","homepageUrl":"pypi"}},"derived":{"capabilities":[{"slug":"dev.version-control","name":"Version Control","confidence":0.745,"provenance":"derived"},{"slug":"data.database","name":"Databases","confidence":0.675,"provenance":"derived"},{"slug":"data.vector-search","name":"Vector Search","confidence":0.638,"provenance":"derived"}],"categories":["data","dev"],"language":"en"},"observed":{"status":"unknown","statusReason":"Distributed as a package to run locally; no network endpoint to check.","lastOkAt":null,"lastProbedAt":null,"statusComputedAt":null,"reliability30d":null,"latestObservations":[],"tools":null,"package":{"name":"micro-app-mcp","registry":"pypi","observedAt":"2026-11-03T17:26:13.973Z","publishedAt":"2026-03-02T19:46:13.926863Z","latestVersion":"0.1.8"},"toolSurface":null,"endpointFacts":[]},"verification":{"claimed":false,"claimedAt":null,"proofs":[]},"provenance":{"sources":[{"source":"pypi","key":"micro-app-mcp","url":"https://pypi.org/project/micro-app-mcp/","firstSeenAt":"2026-09-10T06:22:58.537Z","fetchedAt":"2026-11-03T17:25:43.336Z","normalizedAt":"2026-11-03T17:25:43.336Z"}]},"firstSeenAt":"2026-09-10T06:22:58.537Z","updatedAt":"2026-11-03T17:26:13.973Z"}