Content
92%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.
An unusually tight operational skill: every section gives environment-specific instructions Claude could not infer, with executable commands, explicit exit-code contracts, and validation checkpoints around the risky merge operation. The only improvement space is trimming a couple of low-value framing lines.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is dense and imperative and assumes Claude's competence throughout (no explanation of what PRs, CI, or review bots are), but lines like "Many repos have AI review bots. They're helpful, even if they are not always right." and "Screenshots and videos help as well..." are minor padding that could be trimmed. | 4 / 5 |
Actionability | Guidance is copy-paste ready: a complete pr-wait invocation with flags and enumerated exit codes (0/1/2), exact commands (gh api rate_limit --jq .resources.graphql, gh api repos/<owner>/<repo>/pulls/<n>, git fetch), and concrete tool directives (Monitor tool with timeout_ms: 3600000). | 5 / 5 |
Workflow Clarity | The sequence is clear (handle existing checks first, start exactly one watcher, act on events, rebase when behind, gate the merge) with explicit validation before the destructive merge step ("run git fetch and check the PR again", "Never merge when it is BEHIND or DIRTY") and error-recovery loops (rate-limit reset handling, restart after an exit 1 or a push). | 5 / 5 |
Progressive Disclosure | A well-sectioned single body whose only bundle reference, scripts/pr-wait (a real 556-line executable), is one level deep and clearly signaled with usage instructions ("Wait with scripts/pr-wait from this skill's directory (use its absolute path)"); the bulk of the script is appropriately kept out of SKILL.md. | 5 / 5 |
Total | 19 / 20 Passed |