bmad-document-project

This Skill scans existing software projects in quick, deep, or exhaustive modes to produce structured documentation like architecture, API contracts, and data models, giving an AI the deep project context needed to plan accurately instead of guessing.

Brownfield project scanning works by running a workflow resolver to apply customizations, then reading and validating files to output persistent artifacts including source tree analysis, UI inventory, and integration architecture for PRD context.

The best way to document API contracts and data models from an existing system is using an exhaustive deep-dive scan mode that parses the codebase and generates structured, PRD-ready deliverables immediately written to disk.

Yes, you need Python3 installed to run the automated project documentation workflow, because the process relies on executing a customization resolver script to determine the correct activation sequence for your environment.

Yes, you can customize the documentation workflow by providing project override files alongside the skill metadata, which the workflow resolver reads to adjust the activation sequence and architecture generation for your specific project.

A key limitation is that automated brownfield documentation tools require YAML frontmatter-defined skill metadata to function, and they depend on a deterministic instructions router that must successfully read and validate files before generating any persistent artifacts.