grepai
Semantic Search & Call Graphs for AI Agents (100% Local)
grepai is an MCP server and CLI tool that enables AI agents to perform semantic code search using vector embeddings rather than simple text matching. It allows agents to query codebases with natural language (e.g., "authentication logic") and find conceptually related code even when naming conventions differ. The tool runs entirely locally to preserve privacy, indexes code automatically via a file watcher, and significantly reduces AI agent token consumption by providing only relevant context. It works out-of-the-box with Claude Code, Cursor, and Windsurf.
Key Features
Use Cases
- 01Finding authentication or error handling logic across a codebase without knowing exact function names
- 02Tracing which functions call a specific method before refactoring
- 03Reducing AI agent token usage by providing targeted code context instead of full file dumps
- 04Discovering related code patterns when naming conventions are inconsistent
- 05Keeping AI coding assistants up-to-date with live codebase changes via automatic indexing
- 06Searching legacy codebases where documentation is sparse or outdated
Related MCP Servers
View moremarkitdown
Python tool for converting files and office documents to Markdown.
firecrawl
The context API to search, scrape, and interact with the web at scale. 🔥
prompts.chat
f.k.a. Awesome ChatGPT Prompts. Share, discover, and collect prompts from the community. Free and open source — self-host for your organization with complete privacy.
langflow
Langflow is a powerful tool for building and deploying AI-powered agents and workflows.
grepai — FAQ
What is grepai?+
grepai is a privacy-first CLI tool and MCP server that enables semantic code search using vector embeddings. Unlike grep's exact text matching, it understands code meaning and allows natural language queries to find conceptually related code, while all processing happens locally on your machine.
How do I install grepai?+
On macOS, use Homebrew: `brew install yoanbernabeu/tap/grepai`. On Linux/macOS, run the install script via curl. On Windows, use the PowerShell install script. After installation, you'll need an embedding provider like Ollama (recommended) with the nomic-embed-text model.
Which AI clients does grepai work with?+
grepai works with Claude Code, Cursor, and Windsurf out of the box. It provides an MCP server interface that any MCP-compatible AI agent can use as a tool for code search and call graph analysis.
Do I need API keys or external services?+
You need an embedding provider—either Ollama (free, local, recommended), LM Studio (free, local), or OpenAI (requires API key). Ollama is recommended because it runs entirely locally without any API keys or external dependencies.
Is grepai free to use?+
Yes, grepai is open source under the MIT license and completely free. When using Ollama or LM Studio as the embedding provider, there are no API costs either.
How does grepai reduce AI agent token usage?+
Instead of feeding entire files or raw search results to AI agents, grepai returns only semantically relevant code snippets. This targeted context dramatically reduces input tokens—users have reported saving significant portions of their Claude usage limits.
How do I install grepai?+
Open the source repository on GitHub and follow its README. grepai is a mcp server — MCP Agents Market links you directly to the official repo.
Is grepai free?+
grepai is an open-source project hosted on GitHub. Check the repository for its license and any usage requirements.