VecGrep Cursor-style semantic code search as an MCP plugin for Claude Code. Instead of grepping 50 files and sending 30,000 tokens to Claude, VecGrep returns the top 8 semantically relevant code chunks (1,600 tokens). That's a 95% token reduction for codebase queries. Benchmarks Measured on the VecGrep codebase itself (5 source files, 26k tokens raw). Token usage per query returns structured node metadata only (name, kind, file, line range) — no source code — so it's ultra-cheap for structural questions ("where is X defined?", "what calls Y?"). Query latency (median, 5 runs) searchgrap
| Stars | 6 |
| Forks | 1 |
| Language | Python |
| Category | MCP 服务器 |
| License | MIT |
| Quality Score | 37.8/100 |
| Open Issues | 2 |
| Last Updated | 2026-06-09 |
| Created | 2026-02-21 |
| Platforms | claude-code, mcp, python |
| Est. Tokens | ~65k |
Explore other popular mcp 服务器 tools:
vecgrep is Cursor-style semantic code search as an MCP plugin for Claude Code. Instead of grepping 50 files and sending 30,000 tokens to Claude, VecGrep returns the top 8 semantically relevant code chunks (~1,6. It is categorized as a MCP 服务器 with 6 GitHub stars.
vecgrep is primarily written in Python. It covers topics such as claude, claude-code, mcp.
You can find installation instructions and usage details in the vecgrep GitHub repository at github.com/VecGrep/vecgrep. The project has 6 stars and 1 forks, indicating an active community.
vecgrep is released under the MIT license, making it free to use and modify according to the license terms.