Lightweight semantic code search engine with hybrid search and MCP server for Claude Code.
codexlens-search is worth checking the docs before setup with strong trust signals. Check agent compatibility and use-case fit before adding it to your workflow.
gh repo view catlog22/codexlens-search --webOpen the official README and confirm the supported install method.
Add the server entry to your MCP client config.
Restart your agent and verify that the server tools appear.
Codexlens-search is a tool that helps you find code in your projects using natural language. It combines vector search, full-text search, AST graph analysis, and regex to give accurate results. It also provides an MCP server so Claude Code can use it directly.
Codexlens-search is a lightweight semantic code search engine designed for developers who need to quickly find relevant code in large codebases. It uses a two-stage hybrid search approach: first, it performs vector search (using embeddings), full-text search (FTS), AST graph search, and ripgrep regex search in parallel; then it fuses the results using Reciprocal Rank Fusion (RRF) and reranks them for optimal accuracy. The tool includes an MCP server that integrates seamlessly with Claude Code, allowing Claude to index projects, search semantically, and locate specific code. It supports multiple ANN backends (USearch, FAISS, hnswlib) and can be installed with optional GPU acceleration (DirectML on Windows, CUDA on Linux). Features include file watching for auto-indexing, gitignore filtering, and focused search that indexes only relevant files on demand. The project is open-source under the MIT license.
Looks usable, but maintenance, license, or security notes deserve a closer look.
Last commit was about 57 days ago.
46 GitHub stars indicate community interest.
2 open issues signal maintenance load.
MIT license detected.
Find a function or class in a large codebase using a natural language description.
Integrate semantic code search into Claude Code for AI-assisted development.
Quickly locate buggy code by searching for patterns or error messages.
Explore unfamiliar codebases by searching for concepts rather than exact names.
Automate code review by searching for similar code patterns across the project.
API keys for embedding and LLM services must be kept secure; avoid committing them to version control.
The tool sends code snippets to external embedding APIs; ensure compliance with your organization's data policies.
GPU acceleration requires additional runtime dependencies (DirectML or CUDA) which may have licensing considerations.
46
Stars
7
Forks
2
Issues
MIT
License
Local codebase intelligence CLI and MCP server for AI coding agents with change-safety gates and audit evidence.
An offline MCP server that indexes your codebase for semantic search, code search, and git history retrieval.
Official MCP reference servers from Anthropic. Includes servers for filesystem, GitHub, Postgres, Slack, and more.
3 security/trust notes recorded.
Setup difficulty is 3/5.