categorizing-files

File categorization maps your codebase into eight standard categories: Config, Tests, Docs, Scripts, Source Code, Data, AI Tooling, and Other. Run the bundled categorize.py script on your project root to classify files by directory patterns, filenames, and content structure, producing a standardized report of how files map to each category.

The taxonomy defines Config (configuration files), Tests (test suites), Docs (documentation), Scripts (automation and tooling), Source Code (application code), Data (datasets and resources), AI Tooling (machine learning and AI artifacts), and Other (unclassified files). Each category has explicit scope and mapping rules applied consistently across your repository.

Yes. The categorization process scans your repository tree and produces categorized output without changing your files. The CLI script respects directory exclusions and priority rules, allowing you to audit and report on project structure while preserving your original code.

Categorization applies a four-phase classification approach: Phase 1 matches directory and filename patterns, Phase 2 checks frontmatter signals, Phase 3 analyzes content structure, and Phase 4 detects keywords. Layered exclusion handling and per-file priority rules resolve ambiguous cases consistently.

The script generates a standardized per-file output schema mapping each file in your codebase to one of the eight categories. Results are formatted for repository audits and automated analysis, showing which files belong to Config, Tests, Docs, Scripts, Source Code, Data, AI Tooling, or Other.