{"$schema":"https://wellknown.network/schemas/agent-record-v1.json","schemaVersion":"1","id":"ag_7sm8g7kq8zmt","handle":"cangjie-mcp","url":"https://wellknown.network/agents/cangjie-mcp","links":{"self":"https://wellknown.network/agents/cangjie-mcp/record.json","html":"https://wellknown.network/agents/cangjie-mcp","markdown":"https://wellknown.network/agents/cangjie-mcp/record.md","api":"https://wellknown.network/api/v1/agents/cangjie-mcp","status":"https://wellknown.network/api/v1/agents/cangjie-mcp/status","claim":"https://wellknown.network/agents/cangjie-mcp/claim","claimApi":"https://wellknown.network/api/v1/claims","claimDescriptor":"https://wellknown.network/agents/cangjie-mcp/claim.json","badge":"https://wellknown.network/agents/cangjie-mcp/badge.svg","openapi":"https://wellknown.network/openapi.json","history":"https://wellknown.network/api/v1/agents/cangjie-mcp/history","tools":"https://wellknown.network/api/v1/agents/cangjie-mcp/tools"},"ard":{"identifier":"urn:air::server:cangjie-mcp","type":"application/mcp-server-card+json"},"kind":"mcp_server","declared":{"name":"cangjie-mcp","summary":"MCP server for Cangjie programming language - documentation search and code intelligence","description":"# Cangjie MCP Server\n\n[![CI](https://github.com/Zxilly/cangjie-mcp/actions/workflows/ci.yml/badge.svg)](https://github.com/Zxilly/cangjie-mcp/actions/workflows/ci.yml)\n[![PyPI](https://img.shields.io/pypi/v/cangjie-mcp)](https://pypi.org/project/cangjie-mcp/)\n[![Python Versions](https://img.shields.io/pypi/pyversions/cangjie-mcp)](https://pypi.org/project/cangjie-mcp/)\n[![GitHub Release](https://img.shields.io/github/v/release/Zxilly/cangjie-mcp)](https://github.com/Zxilly/cangjie-mcp/releases)\n[![License](https://img.shields.io/github/license/Zxilly/cangjie-mcp)](LICENSE)\n\n仓颉编程语言的 MCP (Model Context Protocol) 服务器，提供文档搜索和代码智能功能。\n\n## 功能\n\n- **文档搜索**: 基于向量检索的仓颉语言文档搜索\n- **代码智能**: 基于 LSP 的跳转定义、查找引用、悬停信息等功能\n- **可选远程文档服务**: 支持连接远程文档/索引服务，减少本地资源占用，适合开箱即用或团队共享\n\n## 安装\n\n你可以根据偏好的包管理器，选择通过 PyPI (Python) 或 npm (Node.js) 进行安装。\n\n### 通过 PyPI 安装 (Python)\n\n```bash\npip install cangjie-mcp\n```\n\n#### 安装并指定 Rust feature\n\n当平台没有预编译 wheel，或你希望按需启用 GPU/NPU 等加速后端时，可强制从 sdist 构建并传入 maturin 构建参数：\n\n```bash\npip install --no-binary cangjie-mcp cangjie-mcp \\\n  --config-settings=build-args=\"--features local\"\n```\n\n可用 feature（传给 `cangjie-mcp-cli`）：\n\n- `local`：本地向量化（CPU；默认构建的二进制已启用）\n- `legacy`：本地向量化改用 `ort-tract` 后端（适配旧 glibc 构建）\n- `local-cuda` / `local-cudnn`：启用 CUDA/CUDNN 后端\n- `local-metal`：启用 Apple Metal 后端\n- `local-mkl` / `local-accelerate`：启用 MKL / Accelerate 后端\n\n### 通过 npm 安装 (Node.js)\n\n```bash\nnpm install -g cangjie-mcp\n```\n\nnpm 默认优先使用预编译二进制；当当前平台没有匹配包，或 Linux 的 glibc 低于 `2.28` 时，会自动回退到本地源码构建。你也可以显式强制源码构建：\n\n```bash\nCANGJIE_MCP_FORCE_BUILD=1 npm install -g cangjie-mcp\n```\n\n## 架构\n\ncangjie-mcp 支持两种运行模式：\n\n### 本地模式（默认）\n\nMCP 服务器在本地加载检索索引（BM25 + 向量索引），直接处理查询。\n\n```bash\ncangjie-mcp               # 或 uvx cangjie-mcp\n```\n\n### 远程服务模式（可选）\n\n如果你不想在本机下载/加载向量模型与索引（或希望多人/多台机器共享同一套索引），可以把检索能力独立部署为 `cangjie-mcp-server`，同时对外提供：\n\n- **HTTP 查询 API**，供本地 `cangjie-mcp --server-url` 连接（LSP 工具仍在本地可用）\n- **Remote MCP 端点**，MCP 客户端无需本地运行 `cangjie-mcp` 即可直接连接\n  - Streamable HTTP（默认 `/mcp`）— [MCP 规范推荐的现代传输方式](https://modelcontex…","publisher":{"name":"Zxilly","url":null},"homepage":null,"repository":"https://github.com/Zxilly/cangjie-mcp","version":"0.5.0","license":null,"protocols":["mcp"],"tags":["cangjie","mcp","model-context-protocol","language-server"],"pricing":null,"endpoints":[{"url":"pypi:cangjie-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","publisherName":"pypi"}},"derived":{"capabilities":[{"slug":"dev.package-management","name":"Packages & Dependencies","confidence":0.722,"provenance":"derived"}],"categories":["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":"cangjie-mcp","registry":"pypi","observedAt":"2026-09-15T11:22:40.467Z","publishedAt":"2026-07-13T15:39:33.117363Z","latestVersion":"0.5.0"},"toolSurface":null,"endpointFacts":[]},"verification":{"claimed":false,"claimedAt":null,"proofs":[]},"provenance":{"sources":[{"source":"pypi","key":"cangjie-mcp","url":"https://pypi.org/project/cangjie-mcp/","firstSeenAt":"2026-09-09T11:29:31.408Z","fetchedAt":"2026-09-15T11:20:38.159Z","normalizedAt":"2026-09-15T11:20:38.159Z"}]},"firstSeenAt":"2026-09-09T11:29:31.408Z","updatedAt":"2026-09-15T11:22:40.467Z"}