MCP服务器，具语义搜索+知识图谱，适用于Claude代码、光标和Copilot
概览
BM25 + 向量嵌入 + RAPTOR 目录摘要 + 图展开——融合到一个单一的 MCP 工具中，让 Claude、Copilot 和 Cursor 真正理解你的代码库。 可与 、（VS Code 1.99+）、、 、 、 一起使用。只需一个命令来索引;Claude Code 插件会从那里保持同步。 由PragmaWorks开发——该学科致力于构建不漂移的AI软件。两台兄弟MCP服务器与此组成，各自解决同一问题的不同一半： 它们是独立的：安装一个，或者全部安装三个。 CodeSeeker 解决了一个检索问题，而这个问题正在缩小。 如果你运行一个基于规范的循环，使用真实的线束——一个足够精确的规范，使得无状态模型得出，以及验证结果与实际系统匹配的质量门——或者一个意图驱动的回路，你的助手就得到了它本应去寻找的结构。
概述（AI 翻译）
AI编码助手的四层混合搜索与知识图谱。 BM25 + 向量嵌入 + RAPTOR 目录摘要 + 图展开——融合到一个单一的 MCP 工具中，让 Claude、Copilot 和 Cursor 真正理解你的代码库。
支持Claude Code、GitHub Copilot（VS Code 1.99+）、Cursor、Windsurf 和 Claude Desktop。 只需一个命令来索引;Claude Code 插件会从那里保持同步。
由PragmaWorks作为生成规范的一部分构建—— 为构建不漂移的AI软件提供纪律。两个兄弟MCP服务器组成 用这个方法，每个人分别解决同一问题的不同一半：
| 它给你的助手带来了什么 | |
|---|---|
| CodeSeeker（此） | 事物所在之处 — 语义搜索与对你代码的知识图谱 |
编年史 npm i -g chronicle-mcp |
之前发生的事 — 分层记忆能在上下文重置中幸存 |
锻造工艺 npm i -g forgecraft-mcp |
应如何构建 — SOLID、测试、架构及CI/CD标准 |
它们是独立的：安装一个，或者全部安装三个。
当你不再需要这个
本页由 AI 翻译，技术细节请以原文为准。
README
以下为原文 README 节选，完整内容请查看原文。
CodeSeeker
Four-layer hybrid search and knowledge graph for AI coding assistants.
BM25 + vector embeddings + RAPTOR directory summaries + graph expansion — fused into a single MCP tool that gives Claude, Copilot, and Cursor a real understanding of your codebase.
Works with Claude Code, GitHub Copilot (VS Code 1.99+), Cursor, Windsurf, and Claude Desktop.
One command to index; the Claude Code plugin keeps it in sync from there.
Built by PragmaWorks as part of Generative Specification — the discipline for building software with AI that doesn’t drift. Two sibling MCP servers compose with this one, each solving a different half of the same problem:
| what it gives your assistant | |
|---|---|
| CodeSeeker (this) | where things are — semantic search and a knowledge graph over your code |
Chronicle npm i -g chronicle-mcp |
what happened before — tiered memory that survives context resets |
Forgecraft npm i -g forgecraft-mcp |
how it should be built — SOLID, testing, architecture and CI/CD standards |
They are independent: install one, or all three.
When you stop needing this
CodeSeeker solves a retrieval problem, and that problem is shrinking.
If you run a spec-driven loop with a real harness — a specification precise enough that a stateless model derives from it, and quality gates that verify the result against a live system — or an intent-driven one, your assistant is being handed the structure it would otherwise have to go find. A sentinel tool that routes actions instead of sprawling across a dozen endpoints removes another slice of the searching. Frontier models with better navigation and larger effective context remove more. Past a certain point, semantic search over your own code stops being the bottleneck, and this server becomes something you could uninstall without noticing.
That is the intended outcome, not a defect: a method that works should make its own scaffolding unnecessary. Most teams do not start there, though, and until the process is in place CodeSeeker is the cheaper substitute — one command instead of a discipline.
If you would rather have the process than the substitute, start with the field guide: Generative Specification — Field Guide (PDF).
The Problem
AI assistants are powerful editors, but they navigate code like a tourist:
安装
npx -y codeseeker serve --mcp配置
{
"mcpServers": {
"codeseeker": {
"command": "npx",
"args": ["-y", "codeseeker", "serve", "--mcp"],
"env": { "CODESEEKER_STORAGE_MODE": "embedded" }
}
}
}