grepai is a semantic code search tool designed specifically for AI coding agents that indexes the meaning of code rather than just text patterns. It understands code semantically, allowing developers to search by what code does rather than what it's called.

grepai provides semantic search with natural language queries, call graph tracing capabilities, and operates 100% locally for privacy. The tool features real-time file watching that updates the index automatically as you code, eliminating the need for manual re-indexing. It offers native MCP server integration and JSON output for seamless AI integration.

The tool works by creating vector embeddings of code to understand its semantic meaning, enabling searches based on functionality rather than keyword matching. It can index 10,000 files in seconds and return search results in milliseconds.

grepai helps developers ship faster by making AI tools actually understand their codebase. It enables finding code that traditional grep searches miss, understanding the impact of changes through call tracing, and keeping all code processing local to the machine.

Target users include developers working with AI coding assistants like Claude Code, Cursor, and Windsurf. The tool supports multiple programming languages including Go, TypeScript, JavaScript, Python, PHP, Java, C#, C++, Rust, and Zig.