acquire-codebase-knowledge

To generate codebase documentation for developer onboarding, you run a project-root scan mapping directory structure, manifests, and entry points into verifiable markdown artifacts. This creates evidence-backed docs without guessing, explicitly marking unknowns as [TODO] or [ASK USER].

Evidence-backed repository documentation means every statement in the generated docs is verifiable from files or terminal output. It works by running a scan script validating documentation against inquiry checkpoints, enforcing strict traceability to evidence and surfacing unresolved items.

Yes, you can map monorepo architecture and discover integrations through automated codebase scanning. The scan captures CI/CD, containers, security config, and performance markers, packaging them into a complete docs bundle covering stack, structure, and conventions.

To document codebase architecture with a focus on specific sections, you request repository-level discovery with an optional focus area that prioritizes certain sections. The scan script populates the required seven documents from verifiable evidence based on that priority.

Automated codebase mapping for repository documentation is limited to stating what is verifiable from files or terminal output. It cannot guess team intent or infer undocumented architecture, meaning unresolved items are explicitly surfaced as [TODO] or [ASK USER] rather than filled with assumptions.