Sensegrep is a Model Context Protocol (MCP) server designed for semantic and structural code search, specifically optimized for AI-native development environments. It exposes tools such as semantic code search, project indexing, and duplicate code detection to AI agents and coding assistants like Claude, Cursor, and Windsurf. Sensegrep integrates with source code repositories (supporting TypeScript, JavaScript, Python, and more) through tree-sitter AST parsing, and utilizes AI embeddings (Gemini, OpenAI-compatible) with LanceDB for vector search. It is valuable for developers, AI agent workflows, and teams aiming to enhance code search, refactoring, and understanding across modern codebases.
Visit Sensegrep's official website for product details and getting started.