grepai-mcp-cursor

Integrate GrepAI semantic code search and call graph tools into Cursor via MCP.

18|2|Updated Jan 28, 2026

Search Tags:#semantic search #mcp #code search #call graph #cursor #grepai

Ready to run

npx skills add https://github.com/yoanbernabeu/grepai-skills --skill grepai-mcp-cursor

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: grepai-mcp-cursor
Source: https://github.com/yoanbernabeu/grepai-skills/tree/main/skills/integration/grepai-mcp-cursor
Command: npx skills add https://github.com/yoanbernabeu/grepai-skills --skill grepai-mcp-cursor

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Cursor users want to leverage GrepAI's semantic search capabilities without leaving their IDE. This skill provides an MCP bridge so Cursor can query GrepAI for meaning-based code search, dependency tracing, and call graph insights.

Core Features & Use Cases

MCP integration: Exposes GrepAI tools (grepai_search, grepai_trace_callers, grepai_trace_callees, grepai_trace_graph, grepai_index_status) to Cursor.
Project-wide search: Enable semantic search across a codebase from within Cursor.
Traceability: Build and inspect call graphs to understand function dependencies.
Use Case: In a Cursor project, ask the AI to find where authentication is implemented and how it propagates.

Quick Start

Use the grepai-mcp-cursor skill to configure MCP with Cursor by creating .cursor/mcp.json in your project root and restarting Cursor, then verify by asking Cursor to "Search the codebase for authentication".

Frequently Asked Questions about grepai-mcp-cursor

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I enable semantic code search inside Cursor?▼

To enable semantic code search inside Cursor, you integrate GrepAI via the Model Context Protocol. This exposes semantic search tools directly within the IDE by configuring an MCP server in your project root.

Can I trace function dependencies and call graphs using semantic search?▼

Yes, you can trace function dependencies using semantic search through specific exposed tools. The integration provides grepai_trace_callers, grepai_trace_callees, and grepai_trace_graph to inspect and build call graphs for your codebase.

What do I need to configure before using GrepAI with MCP?▼

Before using GrepAI with MCP, you need a configured MCP server, a local embedding provider, and a project-indexed workspace. These prerequisites allow the system to expose semantic search tools to your IDE.

How does semantic code search differ from regular text search?▼

Semantic code search differs from regular text search by finding meaning-based matches across a codebase rather than exact string matches. This allows developers to query concepts like authentication implementation and retrieve relevant functional dependencies.

How do I set up the MCP server for Cursor projects?▼

To set up the MCP server for Cursor projects, create a .cursor/mcp.json file in your project root and restart Cursor. After restarting, verify the connection by asking the AI to search the codebase for a specific concept.