#4343rd of 111 in MCP servers for developers
grepai
CLI for semantic code search with call-graph tracing and an MCP server
- Stars
- 2k
- License
- MIT
- Last commit
- Oct 2026
- Last release
- Oct 2026
- Language
- C
1,608 on 21 Apr 20261,950 on 10 Nov 2026
From GitHub's daily star counts.
Checked on GitHub.
Before you pick it
- Known advisory in latest version. 2 known advisories: github.com/yoanbernabeu/grepai v1.2.0, the latest version. Source (opens in a new tab), checked .
Overview
grepai indexes a codebase with vector embeddings and answers natural-language queries such as "authentication logic", instead of exact text or regex matches. A file watcher daemon keeps the index current, and `grepai trace callers` finds callers of a function. It also runs as an MCP server so agents like Claude Code, Cursor and Windsurf can call it as a tool.
Who it is for: Developers who want semantic repo search for themselves or their coding agents
Strengths
- Natural-language queries find code even when naming conventions differ
- Embeddings can come from local Ollama or LM Studio, so code stays on the machine
- File watcher daemon keeps the index updated without manual reindexing
- Call-graph tracing and MCP server support, installable via Homebrew, shell script or PowerShell
Weaknesses
- Needs a separately installed embedding provider (Ollama, LM Studio or OpenAI)
- Using OpenAI as provider sends code to an external service
- Pull requests are not accepted; contributions go through issues only
- Token-saving claims come from user quotes, with no benchmark in the README
What it needs
- Compose
- Needs Ollama, LM Studio, OpenAI
- Models: nomic-embed-text
Quick answers
Is grepai open source?
Yes. Its code is on GitHub under the MIT license.
Can you run grepai with Docker?
Yes. Its repository has a Docker setup: a Dockerfile, a Compose file or a published image.
Also in MCP servers for developers
See all 111| Rank | Project | Score |
|---|---|---|
| 1 | gitnexusCode knowledge graph exposed to coding agents through MCP tools | 96 out of 100 |
| 2 | codegraphLocal code knowledge graph that coding agents query over MCP | 95 out of 100 |
| 3 | playwright-mcpMCP server that lets LLMs drive a browser through Playwright accessibility snapshots | 92 out of 100 |
| #4 | code-review-graphLocal Tree-sitter code graph that gives coding assistants review context over MCP | 92 out of 100 |
| #5 | serenaMCP server giving coding agents symbol-level code search, editing and refactoring | 91 out of 100 |