code-graph

Navigate AST-based code graphs for symbol lookup and dependency analysis.

705|56|Updated Dec 26, 2025
One-click install
npx skills add https://github.com/alinaqi/maggy --skill code-graph-alinaqi
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: code-graph
Source: https://github.com/alinaqi/maggy/tree/main/skills/code-graph
Command: npx skills add https://github.com/alinaqi/maggy --skill code-graph-alinaqi

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It eliminates slow, error-prone navigation through large codebases by replacing brute-force grepping and repeated file reading with fast AST-backed symbol lookup and dependency-aware blast radius analysis.

Core Features & Use Cases

  • Sub-ms symbol lookup & dependency analysis: Quickly find definitions, callers, and relationships so you can understand impact before you open files.
  • Blast radius detection for safer changes: Identify which parts of the system are affected to prevent accidental breakage.
  • Graph-first workflow for engineering velocity: A practical process for planning, locating, understanding, and verifying changes using the graph MCP tools.
  • Use Case: Before refactoring an authorization middleware function, query the graph to identify all dependents and affected modules, then open only the specific files required to implement the change confidently.

Quick Start

Use the code-graph skill to answer “What are the definition, callers, and blast radius of function X in this repository?” before reading any files.

Frequently Asked Questions about code-graph

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

FAQPage Schema
How do I analyze the blast radius of code changes in a large repository?

Blast radius detection tracks which modules are affected by a code change using AST-based dependency graphs. By querying the code graph, you identify all dependents of a function before modifying it, preventing accidental breakage in large multi-module repositories.

What is the best way to find symbol definitions and callers without grep?

AST-based code graph navigation provides sub-millisecond symbol lookup, replacing brute-force grepping. Querying the dependency graph instantly returns definitions, callers, and structural relationships across large multi-module repositories without reading full files.

How do I trace call paths across multi-module repositories efficiently?

Call path tracing uses MCP tool-driven graph querying workflows via codebase-memory-mcp. By calling trace_call_path on the AST code graph, you map execution routes and locate dependent code across multi-module repositories without inefficient full-file reads.

Do I need MCP tools to perform AST-based dependency analysis?

Yes, AST-based dependency analysis requires MCP tool-driven graph querying workflows via codebase-memory-mcp. It uses search_graph, get_code_snippet, trace_call_path, and detect_changes architecture helpers to navigate the code graph and return structural relationships.

When should I use a code graph instead of full-file reads for refactoring?

Use a code graph instead of full-file reads when working across large multi-module repositories where grep becomes inefficient. AST-backed dependency analysis provides fast blast radius detection, ensuring you only open specific files required for safe code modifications.