gds-document-project

Generates structured documentation for existing brownfield codebases through configurable scan workflows.

Updated Mar 14, 2026
One-click install
npx skills add https://github.com/ArchaonHW/MingGoRTS --skill gds-document-project-archaonhw
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: gds-document-project
Source: https://github.com/ArchaonHW/MingGoRTS/tree/main/.agents/skills/gds-document-project
Command: npx skills add https://github.com/ArchaonHW/MingGoRTS --skill gds-document-project-archaonhw

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Brownfield projects often lack up-to-date documentation, making it hard for developers and AI agents to understand the codebase before planning new features. This Skill scans an existing project and produces a complete documentation set (index, architecture, source tree, development guide, API contracts, data models) ready for brownfield PRD workflows. ## Core Features & Use Cases - Three Scan Levels: Choose quick (pattern-based, no source reading), deep (critical directories), or exhaustive (all source files) analysis depending on time and depth needs. - Resumable State Tracking: A project-scan-report.json state file records progress after every step, allowing interrupted runs to resume within 24 hours. - Deep-Dive Mode: Exhaustively documents a single folder, feature, or API group with full file inventories, dependency graphs, and data flow analysis. - Use Case: Point the workflow at a legacy repository with no docs; it detects the project type from manifest files, classifies the stack, and writes index.md, architecture.md, and development-guide.md to the output folder. ## Quick Start Ask the AI to document this project and choose a quick scan to generate the initial documentation set.

Frequently Asked Questions about gds-document-project

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

FAQPage Schema
How do I generate documentation for an existing codebase?

Run the document-project workflow and select a scan level: quick for pattern-based analysis, deep for critical directories, or exhaustive for all source files. The workflow writes index.md, architecture, and development guides to the output folder.

What is the difference between quick, deep, and exhaustive scans?

Quick scan reads only configs and manifests without opening source files. Deep scan reads files in critical directories based on detected project type. Exhaustive scan reads every source file excluding node_modules, dist, and build folders.

Can I resume an interrupted documentation scan?

Yes, the workflow writes a project-scan-report.json state file after each step. If the state file is less than 24 hours old, you can resume from the last completed step; older state files are archived automatically.

How do I document only one module or feature in detail?

Use deep-dive mode, which always runs an exhaustive scan of a selected folder, file, feature, or API group. It produces a deep-dive document with full file inventories, dependency graphs, data flow, and integration points.

Does the documentation workflow support monorepos or multi-part projects?

Yes, it detects multiple parts such as client and server folders, classifies each part separately, and generates per-part architecture files plus an integration-architecture.md and project-parts.json metadata file.