Content
76%Weight 40%Scale 1-5Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The skill body is highly actionable with executable examples and clean structure, but workflow checkpoints and signaling of the bundled citation reference are the main gaps.
Suggestions
Add an explicit validation step in the Typical Workflow (e.g., confirm resolve-gencode-id returned a valid ID before using it, and how to handle an empty/error response).
Reference references/citation.bib from the body so the bundled citation material is discoverable and not orphaned.
Trim the redundancy between the Command Selection Guide and the Commands section, or merge them so each command's purpose and example live in one place.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is largely efficient with useful boundary guidance and no padding about concepts Claude already knows; the Command Selection Guide slightly duplicates the per-command detail that follows. | 4 / 5 |
Actionability | Every subcommand has copy-paste-ready, fully executable invocations with realistic arguments and output paths, covering all common cases. | 5 / 5 |
Workflow Clarity | The resolve-then-query dependency and a 2-step Typical Workflow are clearly sequenced, but there are no validation checkpoints or error-recovery guidance (e.g., what to do if resolve returns nothing or the API errors). | 3 / 5 |
Progressive Disclosure | The body is well-organized into clear sections and correctly references the real scripts/gtex_cli.py inline, but the bundled references/citation.bib is never signaled or navigated from SKILL.md. | 4 / 5 |
Total | 16 / 20 Passed |