Code Graph RAG

Provides multi-language code analysis with semantic intelligence and graph representations for codebases.

javascript

66

GitHub Stars

javascript

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "er77-code-graph-rag-mcp": {
      "command": "npx",
      "args": [
        "@er77/code-graph-rag-mcp",
        "/path/to/your/codebase"
      ],
      "env": {
        "MCP_TIMEOUT": "80000",
        "MCP_SEMANTIC_WARMUP_LIMIT": "25",
        "MCP_DEBUG_DISABLE_SEMANTIC": "0"
      }
    }
  }
}

You can use the Code Graph RAG MCP Server to analyze multi-language codebases, build intelligent graph representations, and perform semantic code analysis across projects. This server exposes a rich set of methods to search, compare, and understand code, making it easier to navigate large codebases and assess changes with confidence.

How to use

Install the MCP server locally or run it via a fast inline command, then connect your MCP client to access 24 methods for semantic code analysis, graph health checks, and cross-language relationships. Start by launching the MCP server for your project and then issue queries through your client to perform tasks such as semantic search, impact analysis, clone detection, and project graph exports.

How to install

# Install globally
npm install -g @er77/code-graph-rag-mcp

# Or use with npx (no installation)
npx @er77/code-graph-rag-mcp /path/to/your/project
```}]}]},{

Notes on usage and setup

The server supports a lightweight setup that you can integrate with Claude Desktop, Gemini, and Codex clients. For Claude Desktop, you can add the MCP server configuration to the inspector to enable quick analyses like discovering entities, finding similar code, and assessing the impact of changes.

Troubleshooting

If you encounter a native module mismatch or a missing migration in older installations, perform a rebuild or reindex to restore compatibility. For one-shot indexing runs, you can supply a JSON-RPC payload to execute tools directly in debug mode, which bypasses the semantic agent for a lighter run.

Available tools

get_graph_health

Check the health and status of the graph database and indexing components.

reset_graph

Safely reset the current graph data to reindex from scratch.

clean_index

Remove and rebuild the index to ensure a clean reindexing process.

lerna_project_graph

Export a workspace dependency graph for a Lerna project; supports optional ingest and cached refresh controls.

get_version

Return server version and runtime details for diagnostics.

get_agent_metrics

Collect runtime metrics across agents to monitor performance and usage.

get_bus_stats

Inspect knowledge bus topics and system health statistics.

clear_bus_topic

Clear a specific knowledge bus topic to reset related state.

list_entity_relationships

Query relationships for an entity name, such as imports or dependencies.