bmad-document-project

Generates comprehensive documentation for brownfield codebases through configurable scan workflows.

Updated May 22, 2026
One-click install
npx skills add https://github.com/Asygnuz-S-A-S/one-star --skill bmad-document-project-asygnuz-s-a-s
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: bmad-document-project
Source: https://github.com/Asygnuz-S-A-S/one-star/tree/main/.claude/skills/bmad-document-project
Command: npx skills add https://github.com/Asygnuz-S-A-S/one-star --skill bmad-document-project-asygnuz-s-a-s

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Brownfield projects often lack documentation that AI agents and new developers need to understand the existing system. This Skill scans an existing codebase and produces a complete, structured documentation set (architecture, API contracts, data models, source tree, development guides) ready for brownfield PRD planning. ## Core Features & Use Cases - Three Scan Levels: Choose quick (pattern-based), deep (critical directories), or exhaustive (every source file) analysis depending on time and depth needs. - Multi-Part Project Support: Detects monorepos and client/server structures, documenting each part separately plus an integration architecture document. - Deep-Dive Mode: Performs exhaustive file-by-file analysis of a specific folder, feature, or API group, producing dependency graphs, data flow traces, and modification guidance. - Resumable State Tracking: Maintains a project-scan-report.json state file so interrupted scans can resume within 24 hours. - Use Case: Point it at a legacy Next.js e-commerce repository and receive an index.md, architecture.md, api-contracts.md, data-models.md, and development-guide.md that an AI agent can use to plan new features. ## Quick Start Ask the AI to document this project and choose a scan level when prompted.

Frequently Asked Questions about bmad-document-project

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

FAQPage Schema
How do I document an existing brownfield codebase for AI agents?▼

Run the document-project workflow, which scans your repository and generates an index.md, architecture docs, source tree analysis, API contracts, and data models. The output is designed so AI agents can plan new features against the existing system.

What scan levels does the document-project workflow support?▼

Three levels: quick scan uses pattern matching without reading source files, deep scan reads files in critical directories per project type, and exhaustive scan reads every source file excluding node_modules, dist, and build folders.

Can the workflow handle monorepos or multi-part projects?▼

Yes. It detects client/server and monorepo structures, documents each part separately with part-suffixed files like architecture-client.md, and generates an integration-architecture.md plus a project-parts.json metadata file describing how parts communicate.

What happens if the documentation scan is interrupted?▼

The workflow writes a project-scan-report.json state file after every step. If the state file is less than 24 hours old, you can resume from the exact step where it stopped; older state files are archived and a fresh scan begins.

What is deep-dive mode and when should I use it?▼

Deep-dive mode exhaustively reads every file in a chosen folder, feature, or API group and produces a document with full export signatures, dependency graphs, data flow, side effects, and modification guidance. Use it before implementing changes in a specific area.