What is code-graph-mcp?

code-graph-mcp is a high-performance server that builds a structured knowledge graph of your codebase for AI coding assistants. It offers multi-language AST parsing, semantic search, call graph traversal, and HTTP route tracing to provide deep code understanding and optimize LLM context usage.

Is code-graph-mcp open source?

Please check the official repository for license details.

code-graph-mcp: A high-performance server that builds a structured knowledge graph of… (★ 54)

Features

01Multi-language AST parsing across various depths (calls, imports, inheritance, HTTP routes) for languages like TypeScript, Python, Go, Rust, and Java.

02Hybrid semantic code search combining BM25 full-text and vector similarity search with Reciprocal Rank Fusion (RRF).

03Advanced code graph analysis including recursive call graph traversal, HTTP route tracing, dead code detection, and impact analysis.

04Incremental indexing with real-time updates via Merkle tree change detection and LLM context optimization through token-aware snippet extraction.

Why choose it

+Multi-language AST parsing across various depths (calls, imports, inheritance, HTTP routes) for languages like TypeScript, Python, Go, Rust, and Java.

+Perform impact analysis to understand the consequences and blast radius of proposed code changes.

+Covers 6 supported environments or platforms, which is helpful for broader deployment needs.

+The latest recorded update is 2026-07-19, which suggests the project is still actively maintained.

Trade-offs

!There are at least 8 related tools in the same category, so the best choice is easier to make after side-by-side comparison.

Compatibility

TypeScript/JavaScript

Language

Verified via docs

Python

Language

Verified via docs

Language

Verified via docs

Rust

Language

Verified via docs

MCP Clients

Protocol

Verified via docs

SQLite

Embedded DB

Verified via docs

Quick start

$ /plugin marketplace add sdsrss/code-graph-mcp

$ /plugin install code-graph-mcp

Use cases

↳Perform impact analysis to understand the consequences and blast radius of proposed code changes.

↳Trace transitive call chains to understand complex function flows and dependencies across your codebase.

↳Conduct semantic searches to find relevant code based on conceptual descriptions, even if exact keywords are not present.

↳Gain a comprehensive overview of project architecture, modules, dependencies, and entry points.

How it compares

≈code-graph-mcp sits in the Code Assistant category, so it makes more sense to evaluate it alongside tools like Context7 instead of in isolation.

≈If your main need is closer to "Perform impact analysis to understand the consequences and blast radius of proposed code changes.", that use case is a better lens for comparison than broad feature checklists alone.

≈code-graph-mcp's licensing and community traction are both easier to judge in category context.