module-documenter

Module Documenter

Purpose

Document a single module by reading its source files, extracting public APIs with full signatures, and generating structured documentation data.

Context

You receive:

• module_id: Kebab-case identifier for this module
• module_name: Human-readable module name
• module_path: Relative path to the module from repo root
• files: List of source files in the module
• session_id: Current session
• output_path: Where to write results

Instructions

1. Read All Source Files

For each file in files:

1. Read the full file contents
2. Note the file's role (entry point, helper, types, tests)
3. Track line numbers for later reference

2. Extract Public API

Identify all public-facing items:

Language	Public API Signals
JS/TS	export, module.exports, export default
Python	Functions/classes without _ prefix, __all__
Go	Capitalized identifiers
Rust	pub keyword
Java	public keyword

For each public item, extract:

• Name: Exact identifier
• Type: function, class, type, constant, export
• Signature: Full signature as it appears in source
• File and line: filename:line_number
• Description: From JSDoc, docstring, comments, or inferred from name/context
• Parameters: Name, type, and description for each
• Returns: Type and description

3. Generate Usage Examples

For each major API item:

1. Search for usage patterns in tests, examples, or other modules
2. If found, adapt the real usage into a concise example
3. If no usage found, construct a minimal example based on the actual signature
4. Mark examples as "from source" or "constructed"

4. Build Internal Diagram

For modules with multiple files:

1. Map import relationships between files within the module
2. Generate a Mermaid flowchart showing file dependencies
3. Skip for single-file modules (set to null)

graph TD
    index[index.ts] --> handler[handler.ts]
    index --> types[types.ts]
    handler --> utils[utils.ts]
    handler --> types