Content
88%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 content is an exemplary operational skill: concrete, executable, and safe — with validation checkpoints before and after merging, verbatim user-facing menus, and hard confirmation gates on destructive paths. Its only weaknesses are mild padding in a few rationale sentences and the option to split the rationalizations table into a reference file.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body assumes Claude's competence throughout — no git or worktree concepts are explained — and every section is operational, but a few rationale sentences (e.g. "Confirm before merging: merging into the wrong base is expensive to undo") and the 11-row rationalizations table could be trimmed. | 4 / 5 |
Actionability | Fully executable guidance throughout: copy-paste bash for environment detection, merge, push, and worktree removal; exact menu wording to present verbatim; and an exact typed confirmation token ('discard') gating destructive deletion. Placeholders like <base-branch> are the only parameters and are unavoidable. | 5 / 5 |
Workflow Clarity | Six clearly sequenced steps with explicit validation checkpoints: a green test suite gates the menu, the merged result is re-tested before any cleanup, and a refused worktree removal surfaces uncommitted files with a fix/move/delete choice loop. Destructive operations are confirmation-gated, so the destructive-operation cap does not apply. | 5 / 5 |
Progressive Disclosure | No bundle files exist and none are referenced; the single-file body is well organized with clear section headers, state and quick-reference tables, and easy navigation. It scores 4 rather than 5 because at ~226 lines (above the 50-line simple-skill exception) content like the Common Rationalizations section could arguably live in a separate reference file. | 4 / 5 |
Total | 18 / 20 Passed |