All Topics
All Topics
Technology
Technology
Design
Design
Programming
Programming
Science
Science
News
News
Gaming
Gaming
Entertainment
Entertainment
Business
Business
Finance
Finance
Sports
Sports
Health
Health
Food
Food
Travel
Travel
Art
Art
Music
Music
Books
Books
Education
Education
Politics
Politics
Personal
Personal
No algorithm. No AI slop. No ads. Just RSS. Pro-human. Indie writers. Real journalism. Open web. Chronological. Hand toasted.

Semble: A CPU-Based Code Search Library for AI Agents That Reduces Token Usage by 98%

By

Bibabomas

14d ago· 6 min readenCode

Summary

Semble is a code search library designed for AI agents that enables fast and accurate code retrieval using ~98% fewer tokens than traditional grep+read approaches. It indexes and searches entire codebases in under a second, with ~200x faster indexing and ~10x faster queries compared to code-specialized transformers, while maintaining 99% of retrieval quality. The tool runs entirely on CPU with no need for API keys, GPUs, or external services, and can be used as an MCP server or via shell integration with agents like Claude Code, Cursor, Codex, and OpenCode.

Key quotes

· 4 pulled
Semble is a code search library built for agents. It returns the exact code snippets they need instantly, using ~98% fewer tokens than grep+read.
Indexing and searching a full codebase end-to-end 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.
Run it as an MCP server or call it from the shell via AGENTS.md and any agent (Claude Code, Cursor, Codex, OpenCode, etc.) gets instant access to any repo.
Snippet from the RSS feed
Fast and Accurate Code Search for Agents. Uses ~98% fewer tokens than grep+read - MinishLab/semble

You might also wanna read