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 excellent orchestration skill: fully actionable with concrete commands, agent prompts, and explicit disposition vocabulary, and a clearly gated phase sequence with feedback loops. Its only inefficiency is the repeated restatement of the tiny-work rule at each phase boundary.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is dense and directive with zero concept explanations or padding, but the tiny-work exception ("for tiny work the PR description") is restated across §1, §2, §3, and §6 and could be consolidated, fitting the efficient-with-minor-trims anchor rather than the every-token-earns-its-place level. | 4 / 5 |
Actionability | Guidance is fully executable: exact watcher commands (`timeout 1800 gh pr checks <n> --required --watch --fail-fast`), a copy-paste reviewer prompt in §3, explicit disposition labels (TRACKED FOLLOW-UP, DECLINED, NOT A FINDING, READY TO MERGE), and precise sub-skill invocations (`/prp-plan publish <path>`, `/prp-review --verify-corrections`) covering the common cases. | 5 / 5 |
Workflow Clarity | Six clearly sequenced phases with explicit validation gates (VALIDATION: GREEN before review, complete canonical report before correction, READY TO MERGE, green CI on the final head) and feedback loops (correction → focused re-verify; CI failure → CI-correction mode → re-review), matching the top anchor. | 5 / 5 |
Progressive Disclosure | The bundle is a single SKILL.md with no references/, scripts/, or assets/, and none are needed: the body is one coherent orchestration contract whose detail lives in the composed /prp-plan, /prp-implement, and /prp-review skills it delegates to, with well-organized sections throughout. | 5 / 5 |
Total | 19 / 20 Passed |