Codebase Navigator

codebase-navigator

Rapidly understand any codebase: identify entry points, map architecture, find key patterns, summarize modules, trace data flow.

When to Use

• Onboarding to a new project
• Understanding unfamiliar code
• Finding where functionality lives
• Mapping dependencies between modules
• Creating architectural overviews

Scripts

codebase-scan.sh

Quick codebase overview — detects languages, frameworks, entry points, file counts.

bash scripts/codebase-scan.sh /path/to/project
bash scripts/codebase-scan.sh .

Output: Markdown summary with languages, frameworks, entry points, directory structure, LOC counts.

module-map.py

Build a module/import dependency map.

python3 scripts/module-map.py --dir src/
python3 scripts/module-map.py --dir src/ --format mermaid
python3 scripts/module-map.py --dir lib/ --format json

Arguments:

• --dir — Directory to scan (default: .)
• --format tree|mermaid|json — Output format (default: tree)

Identifies: core modules (most imported), leaf modules, circular dependencies.

trace-flow.sh

Trace a function/symbol through the codebase.

bash scripts/trace-flow.sh --symbol "handlePayment" --dir src/
bash scripts/trace-flow.sh --symbol "UserService" --depth 3

Arguments:

• --symbol — Function/class/string to trace (required)
• --dir — Directory to search (default: .)
• --depth — Call chain depth (default: 2)

Uses rg (ripgrep) for fast searching.

References

• codebase-patterns.md — Common project structures and how to identify them

Requirements

• rg (ripgrep) — For fast code searching
• python3 — For module-map.py

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

• Author: JansenAnalytics
• Source: JansenAnalytics/claudex
• License: MIT

Install and usage instructions live in the source repository linked above.