Content
77%Weight 40%Scale 1-3Reviews 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 skill body with clear workflows and validation, undermined mainly by repetition across sections and a missing bundled script that the content depends on.
Suggestions
Consolidate the policy.state/preview and context-window guidance into one place (e.g., Model Selection Criteria) and reference it from Troubleshooting/Anti-Patterns instead of restating it.
Add the referenced scripts/fetch-models.sh to the scripts/ bundle (or note that it is generated/optional) so the primary action path is not broken.
Trim the 'When to Query Models' list, which largely duplicates the description's triggers, to keep the body token-lean.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is mostly code and lean commands, but policy.state/preview and context-window guidance is restated across Decision Framework, Anti-Patterns, and Troubleshooting, and the 'When to Query Models' list overlaps the description — efficient but could be tightened. | 2 / 3 |
Actionability | It provides fully executable, copy-paste-ready commands — fetch-models.sh flags, curl+jq filters, opencode run — and the Manual API Query block is a complete runnable fallback if the script is unavailable. | 3 / 3 |
Workflow Clarity | The query → filter → validate → set-default sequence is explicit, with a concrete validation checkpoint ('opencode run "Echo back: model working"') and clear per-command vs default guidance. | 3 / 3 |
Progressive Disclosure | No bundle files exist, so the frequently referenced scripts/fetch-models.sh path is not a real file and all detail lives inline in one SKILL.md; only the external-doc References are cleanly one level deep. | 2 / 3 |
Total | 10 / 12 Passed |