Content
77%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.
A well-structured, highly actionable troubleshooting skill with a clear validated workflow and rich domain specifics. Its main weakness is progressive disclosure: several referenced platform files are absent from the bundle and large reference sections are inlined rather than split out.
Suggestions
Provide the referenced platforms/{ios,android,kmp,flutter,react-native}.md files in the bundle (e.g. under ./references/) so the one-level-deep navigation actually resolves.
Move the 'Reference: SDK error codes', 'Reference: known platform issues', and 'Reference: platform configuration checklists' sections into a separate reference file (e.g. references/error-codes.md) and link to it from SKILL.md to reduce inlined bulk.
Confirm the cross-skill reference to the 'revenuecat-store-state' skill is intentional and that the skill is available, since it is invoked for store-state changes.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Dense with domain-specific, non-obvious knowledge (specific MCP tool names, iOS-version bugs, error-code tables) rather than concepts Claude already knows; minor trimming possible in the diagnostic-report template and some advisory checklist items, but most tokens earn their place. | 4 / 5 |
Actionability | Highly executable guidance throughout — named MCP tools with parameters (e.g. list-offerings with expand=items.package.product), exact logLevel settings, settings paths, and code snippets — with minor gaps where some checklist steps are advisory rather than runnable commands. | 4 / 5 |
Workflow Clarity | Clearly sequenced five-phase workflow (detect platform → universal checklist → dashboard inspection Phase A/B/C with [ ] checklists → platform specifics → verify the fix) with explicit validation in Section 5 and a user-confirmation gate before dashboard mutations, satisfying the destructive-operation feedback-loop requirement. | 5 / 5 |
Progressive Disclosure | Platform-specific content is appropriately deferred to one-level-deep references (platforms/ios.md etc.) and clearly signaled, but those files are not present in the provided bundle and substantial reference material (error-code tables, known issues, configuration checklists) is inlined in SKILL.md rather than split into separate files. | 3 / 5 |
Total | 16 / 20 Passed |