MCP.so
ログイン

Semble

@MinishLab

リポジトリ

Semble について

Fast, accurate, local code search for agents. Indexes any local path or GitHub repo on demand in ~250ms and answers queries in ~1.5ms. Works on CPU, no API keys or external services.

基本情報

カテゴリ

開発者ツール

トランスポート

stdio

公開者

MinishLab

投稿者

Thomas van Dongen

設定

以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。

{
  "mcpServers": {
    "semble": {
      "command": "uvx",
      "args": [
        "--from",
        "semble[mcp]",
        "semble"
      ]
    }
  }
}

ツール

2

Search a codebase with a natural-language or code query. Pass a git URL or local path as `repo` to index it on demand; indexes are cached for the session. Use this to find where something is implemented, understand a library, or locate related code.

Find code chunks semantically similar to a specific location in a file. Use after `search` to explore related implementations or callers. Pass file_path and line from a prior search result.

概要

What is Semble?

Semble is a code search library built for agents. It returns exact code snippets instantly, cutting token usage and waiting time on every step. Indexing and searching a full codebase takes under a second, with ~200x faster indexing and ~10x faster queries than a code-specialized transformer, at 99% of its retrieval quality. Everything runs on CPU with no API keys, GPU, or external services.

How to use Semble?

Install Semble with pip or uv, then run it as an MCP server using uvx --from "semble[mcp]" semble. Agents like Claude Code, Cursor, Codex, and OpenCode are configured with the same command. The server provides two tools: search (for natural-language or code queries, accepting a git URL or local path) and find_related (given a file path and line number, returns semantically similar chunks).

Key features of Semble

  • Indexes a repository in ~250 ms and answers queries in ~1.5 ms, all on CPU.
  • Achieves NDCG@10 of 0.854, on par with code-specialized transformer models.
  • Supports both local paths and remote git URLs.
  • Runs as a drop-in MCP server for Claude Code, Cursor, Codex, OpenCode, and others.
  • Zero setup: no API keys, GPU, or external services required.

Use cases of Semble

  • An agent can search a cloned repository on demand for relevant code snippets.
  • Developers can find code related to a specific file and line number during debugging.
  • CI pipelines can quickly index new repositories and answer natural-language queries about code.
  • Local code editing: index any local project and search by concept or identifier.

FAQ from Semble

Does Semble require an API key or GPU?

No. Everything runs on CPU with no API keys, GPU, or external services.

What agents are compatible with Semble?

Claude Code, Cursor, Codex, OpenCode, and any MCP-compatible agent.

How do I set up Semble as an MCP server?

The setup requires uv to be installed. For Claude Code,

コメント

「開発者ツール」の他のコンテンツ

JetBrains MCP Proxy Server

JetBrains

A model context protocol server to work with JetBrains IDEs: IntelliJ, PyCharm, WebStorm, etc. Also, works with Android Studio

962

Huoshan Test

volcengine

2.4K

Deepwiki MCP Server

regenrek

📖 MCP server for fetch deepwiki.com and get latest knowledge in Cursor and other Code Editors

1.4K

MCP server to deploy code to Google Cloud Run

GoogleCloudPlatform

MCP server to deploy apps to Cloud Run

621

Unity MCP (Server + Plugin)

IvanMurzak

AI Skills, MCP Tools, and CLI for Unity Engine. Full AI develop and test loop. Use cli for quick setup. Efficient token usage, advanced tools. Any C# method may be turned into a tool by a single line. Works with Claude Code, Gemini, Copilot, Cursor and any other absolutely for fr

3.5K