Content
87%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a well-crafted skill that provides concrete, actionable guidance for evaluating package health. It efficiently covers multiple scenarios (single package, version comparison, package comparison) with specific tool calls and CLI commands. The main weakness is the lack of explicit validation between dependent steps—particularly verifying the package UUID retrieval before using it in subsequent queries.
Suggestions
Add a validation checkpoint between Step 1/2 confirming the package UUID was retrieved before querying the scorecard metric, with a fallback action if it's missing.
Add a brief note about what to do when Step 1 (vulnerabilities) succeeds but Step 2 (metrics) fails or vice versa, so Claude knows how to present partial results.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The content is lean and efficient. It doesn't explain what package health means or how ecosystems work. Every section serves a purpose—input parsing, workflow steps, score categories, error handling—without padding or unnecessary context. | 3 / 3 |
Actionability | Provides concrete MCP tool names with exact parameters, executable CLI commands with proper flags, specific filter syntax, and clear recommendation thresholds. The guidance is copy-paste ready with real tool/API calls. | 3 / 3 |
Workflow Clarity | Steps are clearly sequenced (1-5) with logical progression from data gathering to presentation. However, there are no explicit validation checkpoints—no verification that the package UUID was successfully retrieved before using it in Step 2's scorecard query, and no feedback loop for handling partial data (e.g., metrics available but vulnerabilities check failed). | 2 / 3 |
Progressive Disclosure | Content is well-structured with clear sections (Input Parsing, Workflow steps, Next Steps, Error Handling). References to related skills (/endor-check, /endor-upgrade-impact, /endor-scan) and a data sources file (references/data-sources.md) are clearly signaled and one level deep. The skill is appropriately sized for a single file. | 3 / 3 |
Total | 11 / 12 Passed |