Local-first semantic and hybrid BM25 grep/search tool for AI and humans.
ck is easy to set up with strong trust signals. Check agent compatibility and use-case fit before adding it to your workflow.
gh repo view BeaconBay/ck --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.
ck is a code search tool that understands meaning, not just keywords. It uses AI embeddings to find code based on concepts, so searching for 'error handling' finds try/catch blocks even if those words aren't present. It also supports traditional regex search and hybrid modes, and can be used as an MCP server for AI agents.
ck (seek) is a local-first semantic and hybrid BM25 grep/search tool designed for both AI agents and human developers. It indexes codebases using tree-sitter AST parsing and AI embeddings to enable search by meaning, not just keywords. Key features include: semantic search that understands synonyms and related concepts, traditional regex search for exact pattern matching, hybrid search combining both approaches, an interactive TUI with multiple preview modes, and an MCP server for seamless integration with AI agents like Claude Desktop and Cursor. ck automatically builds and updates indexes, supports pagination, and offers editor integration. It is written in Rust for performance and is available under MIT/Apache-2.0 license.
Strong trust signals; still review the README and permissions before production use.
Last commit was about 3 days ago.
1623 GitHub stars indicate community interest.
23 open issues signal maintenance load.
Apache-2.0 license detected.
Find code related to 'error handling' across a large codebase without knowing exact keywords.
Integrate with AI coding assistants to provide semantic code search capabilities.
Quickly locate authentication logic or database connection patterns in unfamiliar projects.
Use as a traditional grep replacement with enhanced semantic capabilities.
Search through documentation and code simultaneously using hybrid queries.
The tool indexes code locally; no data is sent externally unless configured otherwise.
Running the MCP server exposes search functionality to connected AI agents; ensure trusted environment.
Large codebases may require significant disk space for indexes.
1,623
Stars
68
Forks
23
Issues
Apache-2.0
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 2/5.