Content
85%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 body is a lean, well-sequenced investigation workflow with concrete gh CLI commands and explicit verification steps for flakiness and breaking-commit claims. Its main gaps are minor: two steps lack an executable command (notably how to fetch the full logs), and the repeated emphasis phrasing adds a few redundant tokens.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body explains nothing Claude already knows and every step carries task-specific guidance, but emphasis phrasing like 'THE EXACT SAME failing job' / 'THIS SPECIFIC JOB' repeats across steps 2-3 and could be trimmed. Matches anchor 4: efficient with minor instances that could be tightened. | 4 / 5 |
Actionability | Provides concrete commands ('gh run list --workflow=<workflow-name>', 'gh run view <run-id> --json jobs', 'gh pr list --state open --search "<keywords>"'), but steps 1, 3, and 4 rely on direction without a log-retrieval command. Matches anchor 4: mostly executable guidance with minor gaps. | 4 / 5 |
Workflow Clarity | A clear five-step sequence with explicit validation checkpoints ('Verify by checking: does THIS JOB fail in ALL runs after that commit?', 'Verify your hypothesis against the logs and failure logic') plus a defined final report structure. This is a read-only investigation, so the destructive/batch cap does not apply; matches anchor 5. | 5 / 5 |
Progressive Disclosure | The skill is under 50 lines, single-purpose, needs no external references, and is well-organized into numbered investigation steps plus a report spec — exactly the simple-skill case the rubric says can score 5. | 5 / 5 |
Total | 18 / 20 Passed |