Content
38%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 delivers a concrete activation protocol and useful operational commands, but it is bloated by a triple-repeated command list and an inlined persona YAML whose external dependency references point to files not included in the bundle. Consolidating the command definitions, moving the agent YAML to a referenced file, and fixing the stale command names in the workflow would materially raise the score.
Suggestions
Deduplicate the command definitions: keep a single canonical command list and have the Star Commands table, YAML block, and Quick Commands section all derive from it rather than restating it three times.
Move the full agent YAML definition into a referenced bundle file and replace the inline block with a one-level-deep pointer, which would also surface the external dependency references clearly instead of burying them in YAML.
Fix the Typical Workflow to use current command names (replace *rls-audit with *security-audit and *explain with *analyze-performance) and add explicit validation checkpoints (e.g. dry-run before apply-migration, security-audit before deploy) between steps since migrations are destructive operations.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | At 569 lines it is noticeably verbose: the command list is repeated three times (Star Commands table, YAML commands block, Quick Commands section) and the persona YAML carries padding (zodiac '♊ Gemini', archetype, Spanish vocabulary list) that does not advance the task. | 2 / 5 |
Actionability | The activation protocol and CodeRabbit CLI invocations are concrete and executable, but the core database operations are delegated to dependency files (e.g. db-apply-migration.md, security-audit.md) that are not present in the bundle, leaving key execution details missing. | 3 / 5 |
Workflow Clarity | A Typical Workflow is sequenced (Design → Bootstrap → Migrate → Secure → Optimize → Test) and the activation steps are ordered, but validation checkpoints are only implicit and the workflow references stale command names (*rls-audit, *explain) that no longer exist in the command table. | 3 / 5 |
Progressive Disclosure | No bundle files exist, yet the body inlines a ~400-line full agent YAML definition that clearly belongs in a separate file, and references to external dependency files are buried inside that YAML rather than clearly signaled from navigable section headers. | 2 / 5 |
Total | 10 / 20 Passed |