gds-document-project

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

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

Yes. The workflow maintains a project-scan-report.json state file updated after each step. If a state file less than 24 hours old exists, you are prompted to resume from the last completed step; older state files are archived automatically.

Yes. It detects client/server and similar multi-part structures, documents each part separately with part-suffixed files, and generates an integration-architecture.md plus a machine-readable project-parts.json describing how the parts communicate.

Deep-dive mode exhaustively reads every file in a chosen folder, feature, or API group. It produces a complete file inventory with exports and signatures, a dependency graph, data flow analysis, testing coverage, and modification guidance.

It is designed for brownfield projects with existing code. For greenfield projects with only planning artifacts, it warns you and asks before continuing, since there is little source code to analyze.