Content
57%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 content is well-organized and actionable with strong tool sequences and a useful quick reference, but it is hurt by duplicated pitfalls, implicit validation in destructive workflows, and a monolithic structure with no reference files. Splitting reference material and adding explicit validation checkpoints would meaningfully raise the score.
Suggestions
Add explicit validation checkpoints for destructive/batch workflows (e.g., verify item_id and confirm with the user before DELETE/permission changes, then re-check permissions after).
Move the full Quick Reference table and detailed pitfalls into a separate references/ file, keeping SKILL.md as a concise overview with one-level-deep links.
De-duplicate the "Known Pitfalls" section against the per-workflow "Pitfalls" blocks to remove repeated guidance and reclaim tokens.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is mostly efficient and tool-focused with little concept over-explanation, but the "Known Pitfalls" section substantially duplicates pitfalls already listed under each workflow, adding tokens that could be trimmed. | 3 / 5 |
Actionability | Concrete tool slugs, key parameters, and a Quick Reference table give mostly executable guidance for an MCP-tool skill; minor gaps are the absence of worked end-to-end examples for the common cases. | 4 / 5 |
Workflow Clarity | Each workflow has a clear numbered tool sequence and "When to use", but destructive/batch operations (delete, permission grants, copy) lack explicit validate→fix→retry checkpoints; validation is only implicitly mentioned, capping the score at 3. | 3 / 5 |
Progressive Disclosure | The single 240+ line SKILL.md inlines the full API reference, pitfalls, and quick-reference table that could live in separate reference files; no bundle/reference files exist, so structure is section-based rather than file-based. | 3 / 5 |
Total | 13 / 20 Passed |