MCPgee
MCPgee
MCPgeeMCPgee

Semble

v1.0.0•Search & Data Extraction•stable

Fast and Accurate Code Search for Agents. Uses ~98% fewer tokens than grep+read

agentscode-searchembeddingsmcpmcp-server
Share:
3,591
Stars
0
Downloads
0
Weekly
0/5
View on GitHub

What is Semble?

Semble is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to fast and accurate code search for agents. uses ~98% fewer tokens than grep+read

Fast and Accurate Code Search for Agents. Uses ~98% fewer tokens than grep+read

This server falls under the Search & Data Extraction category on MCPgee, the world's largest MCP server directory with 33,000+ servers.

Features

  • Fast and Accurate Code Search for Agents. Uses ~98% fewer to

Use Cases

Search code 98% faster using semantic embeddings.
Find relevant code with 98% fewer tokens than grep+read.
MinishLab

Maintainer

LicenseMIT
Languagepython
Versionv1.0.0
UpdatedMay 22, 2026
Statushealthy
Maintenanceactive

Works with

ClaudeOpenAIwindowsmacoslinux
View SourceBrowse All Servers

Installation

Manual Installation

npx semble

Configuration

Configuration Details

Config File

claude_desktop_config.json

Performance

Response Metrics

Response Time< 200ms
ThroughputMedium

Resource Usage

Memory UsageLow
CPU UsageLow

How to Set Up and Use Semble

Semble is a semantic code search MCP server that indexes your codebase using embeddings and lets AI agents find relevant code with natural language queries — using approximately 98% fewer tokens than the traditional grep-then-read approach. It supports both local directories and remote Git repositories, and exposes two MCP tools (search and find_related) that agents can call to retrieve precise code chunks without reading entire files. Development teams use it to let coding agents navigate large unfamiliar codebases efficiently, dramatically reducing context window usage and improving answer quality.

Prerequisites

  • Python 3.9 or later installed
  • uv package manager installed (https://github.com/astral-sh/uv)
  • An MCP-compatible client such as Claude Desktop, Cursor, or Windsurf
  • A local codebase or accessible Git repository URL to index
1

Install semble

Install semble globally using uv tool install so it is available on your PATH.

uv tool install semble
2

Register semble with your agent integration

Run semble install to configure the tool for use with your MCP client. This sets up the integration so your AI agent can discover and call the semble MCP tools.

semble install
3

Index a codebase

Run a search query to trigger initial indexing of a local path or remote Git repository. Semble builds an embedding index on first access and caches it.

semble search "authentication handling" /path/to/your/project
4

Configure file exclusion rules

Create a .sembleignore file in your project root using standard gitignore syntax to exclude files from indexing. By default, .gitignore patterns are also respected. Use ! to force-include file types not indexed by default.

# .sembleignore
node_modules/
*.min.js
dist/
build/
5

Add to Claude Desktop configuration

Register semble as an MCP server in Claude Desktop so the AI assistant can call it during coding sessions.

{
  "mcpServers": {
    "semble": {
      "command": "uvx",
      "args": ["--from", "semble[mcp]", "semble"],
      "env": {
        "SEMBLE_CACHE_LOCATION": "/optional/custom/cache/path"
      }
    }
  }
}
6

Check token savings

After using semble in your sessions, run the savings command to see cumulative token savings compared to grep+read.

semble savings

Semble Examples

Claude Desktop configuration

Full configuration block for registering semble as an MCP server in Claude Desktop using uvx.

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

Prompts to try

Once semble is connected, use these natural language and code queries in Claude Desktop to search your codebase.

- "Search the codebase for how authentication is handled" (natural language)
- "Find all code related to 'save model to disk'" (semantic query)
- "Find code similar to the function on line 42 of src/auth/login.py" (find_related tool)
- "Search for Foo::bar method implementations" (symbol-style lexical query)
- "Search the remote repo https://github.com/owner/repo for config parsing logic"

Troubleshooting Semble

Index takes a long time to build on first search

Semble builds embeddings on first access, which can take minutes for large codebases. Subsequent searches use the cache and are fast. Override the cache location with SEMBLE_CACHE_LOCATION if the default OS cache path has limited space.

After upgrading semble, the MCP server returns stale results

Run: uv cache clean semble, then restart your MCP client (e.g., Claude Desktop). The cache is tied to the installed version and must be cleared after upgrades.

Certain file types are not being indexed

Semble indexes common code file types by default and respects .gitignore. To include additional types (e.g., .proto or .toml), add force-include rules to .sembleignore with a ! prefix: !*.proto

Frequently Asked Questions about Semble

What is Semble?

Semble is a Model Context Protocol (MCP) server that fast and accurate code search for agents. uses ~98% fewer tokens than grep+read It connects AI assistants to external tools and data sources through a standardized interface.

How do I install Semble?

Follow the installation instructions on the Semble GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.

Which AI clients work with Semble?

Semble works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.

Is Semble free to use?

Yes, Semble is open source and available under the MIT license. You can use it freely in both personal and commercial projects.

Learn More About MCP Servers

Getting Started with MCP

Set up your first MCP server in minutes

MCP Setup Guide

Configure MCP in Claude, Cursor & VS Code

All MCP Tutorials

18+ hands-on guides for developers

MCP FAQ

40+ answers about Model Context Protocol

Semble Alternatives — Similar Search & Data Extraction Servers

Looking for alternatives to Semble? Here are other popular search & data extraction servers you can use with Claude, Cursor, and VS Code.

TrendRadar

★ 58.0k

A real-time hotspot monitoring and news aggregation assistant that provides AI-powered analysis of trending topics across multiple platforms via the Model Context Protocol. It enables users to track news and receive automated notifications through va

Scrapling

★ 52.7k

🕷️ An adaptive Web Scraping framework that handles everything from a single request to a full-scale crawl!

PDF Math Translate

★ 33.9k

[EMNLP 2025 Demo] PDF scientific paper translation with preserved formats - 基于 AI 完整保留排版的 PDF 文档全文双语翻译，支持 Google/DeepL/Ollama/OpenAI 等服务，提供 CLI/GUI/MCP/Docker/Zotero

GPT Researcher

★ 27.2k

An autonomous agent that conducts deep research on any data using any LLM providers

Agent Reach

★ 20.1k

Give your AI agent eyes to see the entire internet. Read & search Twitter, Reddit, YouTube, GitHub, Bilibili, XiaoHongShu — one CLI, zero API fees.

Xiaohongshu

★ 13.7k

MCP for xiaohongshu.com

Browse More Search & Data Extraction MCP Servers

Explore all search & data extraction servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.

Search & Data ExtractionBrowse All Servers

Quick Config Preview

{ "mcpServers": { "semble": { "command": "npx", "args": ["-y", "semble"] } } }

Add this to your claude_desktop_config.json or .cursor/mcp.json

Read the full setup guide →

Ready to use Semble?

Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.

33,000+ ServersFree & Open SourceStep-by-Step Guides
Explore All ServersRead Our Guides