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 instruction-only skill body: lean, non-obvious, opinionated guidance organized into a clearly sequenced workflow with strong validation checkpoints and error-recovery stop conditions. The only improvement would be showing the complete `gh pr create` command (e.g. with --base, --title, --body flags) to close the last actionability gap.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is lean, imperative, and assumes Claude's competence throughout — it never explains what a PR or git is, and every instruction adds non-obvious policy ("Never assume main", "do not rebase or force-push as part of this skill"). The only near-redundancy is the duplicated "$ARGUMENTS" note, which is trivial; this matches the 'every token earns its place' anchor. | 5 / 5 |
Actionability | Guidance is highly concrete: an ordered base-resolution procedure, an exact fallback PR body template in a copy-paste markdown block, explicit flag rules ("--base <branch>", "Use --draft only when the user explicitly requests a draft"), and concrete bad/good title examples. Not a 5 because the actual `gh pr create` invocation is described but never shown as a complete command line, leaving a minor gap for the single most important execution step. | 4 / 5 |
Workflow Clarity | Five clearly sequenced sections (establish target → validate work → write PR → push and create → verify and report) with explicit validation checkpoints: re-reading the merge-base diff, comparing against the requested outcome, checking for existing PRs, and verifying the created PR's number/base/head before reporting success. Stop-and-ask and stop-and-report error paths cover recovery, matching the top anchor including feedback loops. | 5 / 5 |
Progressive Disclosure | This is a single-file skill with no references/, scripts/, or assets/ bundle; nothing needs to be split out, sections are well-organized with a clear hierarchy (Scope → numbered steps → Title/Description subsections), and there are no nested or dead references. Per the rubric's guidance to score against the actual bundle structure, a self-contained skill with well-organized sections warrants the top score. | 5 / 5 |
Total | 19 / 20 Passed |