Content
100%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 model procedural skill: lean, fully executable, with an ordered workflow, explicit user-decision checkpoints, and conditional handling for existing PRs, missing tickets, and oversized diffs. No dimension shows meaningful weakness.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is purely procedural with no explanation of concepts Claude already knows; even the 'avoid gh pr edit' note is non-inferable repo-specific knowledge. The size-threshold sentence ('Reviews get harder past ~10 files / ~300 changed lines') directly informs the split-vs-proceed decision, so every token earns its place — the score-5 anchor. | 5 / 5 |
Actionability | Every step ships copy-paste-ready commands: git context-gathering commands, 'python3 .claude/hooks/pr-size-check.py --report', 'gh pr create --draft --title ... --body ...', and the 'gh api ... --method PATCH --field body="$body"' update. Placeholders like <PR_NUMBER> are parameterization, not pseudocode, matching the fully-executable score-5 anchor. | 5 / 5 |
Workflow Clarity | Steps 0–9 are strictly ordered with explicit checkpoints and error-recovery branches: existing-PR detection before creating, ask-user fallback when the branch lacks a DOCSP ticket, size-verdict branching, wait-for-reply before template use, and the draft-to-ready gate. This matches the score-5 anchor (clear sequence, validation steps, feedback loops). | 5 / 5 |
Progressive Disclosure | There are no bundle files, and the ~140 well-sectioned lines all belong inline as one procedural workflow; nothing here is reference material that should be split out. The external 'staging-preview' skill and template files are one level deep and clearly signaled, matching the score-5 anchor for appropriately placed, easy-to-navigate content. | 5 / 5 |
Total | 20 / 20 Passed |