CodeGraph FAQs

Question 1

What kind of architectural insights can CodeGraph provide?

Accepted Answer

It provides powerful structured queries for tracing dependencies, performing impact analysis, finding nodes by name/type, calculating shortest paths between components, and generating high-level architecture summaries.

Question 2

How does CodeGraph integrate with AI coding agents?

Accepted Answer

CodeGraph uses the Model Context Protocol (MCP) to expose its query engine. It supports various transport modes (stdio, SSE, streamable-http) for seamless integration with MCP-compatible AI agents and clients.

Question 3

Which programming languages are supported by CodeGraph?

Accepted Answer

Currently, CodeGraph supports parsing and generating knowledge graphs for both TypeScript (including TSX) and Java codebases, with potential for more languages in the future.

Question 4

What is CodeGraph?

Accepted Answer

CodeGraph transforms your codebase into a machine-readable knowledge graph, enabling AI coding agents to understand and reason about dependencies, impact, and system architecture for enhanced development workflows.

Question 5

How does CodeGraph generate the knowledge graph?

Accepted Answer

CodeGraph scans your repository, parses source files using Tree-sitter (for TypeScript/TSX and Java), and builds a detailed directed graph, which is then stored for structured architectural queries.

CodeGraph

CodeGraph

Key Features

Use Cases

Key Features

Use Cases