Content
86%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 dense, high-signal skill body that focuses exclusively on non-obvious gotchas and conventions rather than restating known concepts. The main improvement would be adding a complete end-to-end create_issue example and explicit feedback loops on every batch/status flow.
Suggestions
Add one complete, copy-paste-ready create_issue example (with teamId, title, priority, labels) so the most common action is fully executable from the skill.
Add an explicit validate→fix→retry loop for status transitions, e.g. 'if update_issue returns stateId must be a UUID, re-read tracker-config.md for the UUID and retry'.
Surface the priority→label mapping steps as a short numbered checklist so the create-and-label flow is sequenced rather than described.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Lean and efficient; assumes Claude's intelligence by never explaining what Linear, sprints, or priorities are, and every line covers a non-obvious API gotcha or convention Claude would not already know. | 5 / 5 |
Actionability | Provides executable JSON examples, an exact GraphQL query, menu paths, and concrete title/priority formats, but lacks a complete copy-paste create_issue call covering the common end-to-end case. | 4 / 5 |
Workflow Clarity | Clear sequencing with validation checkpoints for risky flows ('verify before delegating', 're-read every issue status before acting', 'search first; if absent create'), though not every flow carries an explicit validate→fix→retry feedback loop. | 4 / 5 |
Progressive Disclosure | A short (~40 line), single-purpose skill with well-organized sections and one-level-deep external references (tracker-config.md, Linear docs URL); the simple-skill exception applies. | 5 / 5 |
Total | 18 / 20 Passed |