If you use AI coding agents (Cursor, Claude Code, Windsurf...), you've probably noticed this:
The agent is smart, but it struggles to find the RIGHT files in your codebase. It greps for exact keywords, misses context, and sometimes hallucinates file paths.
The problem: traditional search tools (grep, ripgrep) weren't built for agents. They need exact matches, but agents think in intent "where is the authentication logic?" not "grep login".
grep was built in 1973. It finds text, not meaning. When you search for "authentication logic" but the code is named handleUserSession, grep fails.
grepai uses vector embeddings to understand code semantically. Search by WHAT code does, not what it's called. Built for AI agents like Claude Code and Cursor, find relevant code instantly instead of browsing hundreds of files.
Features:
• Semantic search with natural language
• Call graph tracing
• 100% local
• Real-time file watching
