What problem does it solve? Maintaining a deduplicated catalog of AI coding tools (terminal CLIs, desktop IDEs, workspace apps) requires manually parsing repository URLs, inferring vendor and type metadata, and editing Markdown tables without breaking row structure. This Skill automates that pipeline from a single URL or batch list. ## Core Features & Use Cases - URL and path parsing: Converts GitHub repo URLs, release pages, download pages, and local directories into 16-column catalog rows with stable IDs, vendor inference, and type detection. - Batch processing with dedup: Accepts arrays, comma/newline-separated strings, or listing pages, then deduplicates against the existing master table before upserting. - Spreadsheet-safe table editing: Upserts full rows into catalog-master-table.md via a Markdown spreadsheet helper, preserving escaped pipes and surrounding document content. - Use Case: Given a list of new AI coding CLI GitHub URLs, run the skill in dry-run mode to preview which entries would be added, skipped as duplicates, or rejected as out-of-scope SaaS tools before writing anything. ## Quick Start Ask the agent to add https://github.com/QwenLM/qwen-code to the AI CLI client catalog as a terminal CLI.
