bmad-document-project

You can automate existing codebase documentation by running an initial scan to identify technologies and architecture, which then generates comprehensive documentation including architecture and component inventories using Python and Jinja2.

Yes, deep-dive project documentation for specific areas is generated by selecting exhaustive scan levels to target specific directories or components, producing comprehensive analysis outputs without scanning the entire project structure.

Python project documentation generation works with Jinja2 and pandas, alongside yamllib and requests, which are required dependencies for parsing code, analyzing project structure, and rendering documentation templates.

The best way to create AI integration documentation for a brownfield project is to use automated scanning to identify existing technologies and architecture, generating development guides and component inventories required for AI integration.

To define project documentation requirements using a CSV file, map project types to specific documentation needs; the Skill utilizes this CSV input to determine required architecture and development guides for each identified project type.

Automated code analysis for project documentation should not be used when the project lacks a clear directory structure or required dependencies like Python, Jinja2, pandas, yamllib, and requests, which are necessary for parsing and rendering.