Content
75%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.
This is a strong, execution-ready skill: fully specified gh/git commands, a clear six-step sequence, explicit blocker checks, and a user-confirmation gate before the destructive merge. Remaining improvements are small — trimming minor over-explanations, fixing the duplicated heading, and adding failure handling for the merge step itself.
Suggestions
Remove or relocate the duplicated "# Merge PR" H1 nested under "## Command Template" so the document has a single heading hierarchy.
Drop the sentence explaining "git pull --prune" (Claude already knows this flag) and the re-statement of the method choices in step 4.
Add brief failure handling for the merge step, e.g., if "gh pr merge" fails or the chosen method is disabled, re-run "gh pr view" and report the new state to the user.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is efficient — commands and templates dominate — but has minor over-explanation: the trailing sentence explaining what "git pull --prune" does and re-stating that <method> is squash/rebase/merge when the options were already enumerated in step 3. These are trimmable without loss, so anchor 4 rather than the every-token-earns-its-place anchor 5. | 4 / 5 |
Actionability | The "gh pr view" command with its full --jq filter and the cleanup commands are copy-paste ready, and the merge command's <number>/<method> placeholders are concretely sourced from earlier steps. Minor gaps keep it from a 5: no guidance if the chosen merge method is disabled on the repo or the merge command itself fails. | 4 / 5 |
Workflow Clarity | Six steps are clearly sequenced with real validation checkpoints for a destructive operation: stop if no PR exists, report blockers (CI, reviews, conflicts) and stop, and a user confirmation gate with an explicit cancel option before merging. It falls short of anchor 5 because there is no error-recovery loop after a failed merge attempt (e.g., re-checking mergeable state or retry guidance). | 4 / 5 |
Progressive Disclosure | Sections are well-organized and all content is appropriately inline for a skill of this size with no external references needed (no references/, scripts/, or assets/ directories exist, and the body cites none). The duplicated nested "# Merge PR" H1 under "## Command Template" is a structural oddity that mildly hurts navigation, keeping this at anchor 4 rather than 5. | 4 / 5 |
Total | 16 / 20 Passed |