Content
72%Weight 40%Scale 1-3Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The body is well-organized, actionable, and appropriately self-contained for a focused single-purpose skill, with executable gh/sl commands and a useful output example. The main gap is the absence of an explicit verification step after destructively rewriting the PR body.
Suggestions
Add an explicit post-edit verification checkpoint, e.g., re-read the updated PR body with `gh pr view` to confirm formatting and that no preserved content (images, links) was lost.
Tighten the more prosey directives ('It is critically important to...', 'It is generally frowned upon to...') into direct imperatives to improve token efficiency.
Consider a brief ordered checklist for the body-rewrite flow (locate PR -> read existing body -> preserve key info -> rewrite why/what -> verify) to make the sequence and checkpoint explicit.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body avoids explaining concepts Claude already knows and stays focused, but some prose could be tightened (e.g., 'It is critically important to explain _why_...', 'It is generally frowned upon to discuss...'), so it is mostly efficient rather than fully lean. | 2 / 3 |
Actionability | It provides concrete, executable commands such as `gh pr view <branch> --repo openai/codex --json number --jq '.number'`, `sl log --template '{github_pull_request_url}' -r .`, and `sl sl`, plus a real example transcript with annotated output, matching the fully-executable score-3 anchor. | 3 / 3 |
Workflow Clarity | The process (determine PR, check existing body to preserve info, rewrite why/then what, format, reference related items) is sequenced and includes a preservation checkpoint, but rewriting a PR body is destructive and there is no explicit validate/verify step after editing, capping workflow clarity at 2 per the batch/destructive guideline. | 2 / 3 |
Progressive Disclosure | The skill is self-contained with no bundle files and is organized into clear, well-signaled sections (Determining the PR(s), PR Body Contents, Working with Stacks, Sapling) with no nested references, which satisfies the simple-skill allowance for a top score. | 3 / 3 |
Total | 10 / 12 Passed |