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.
This is a strong, highly actionable skill body: every step is executable, the destructive batch operation is gated by guards, classification rules, and an explicit confirmation, and failure paths are handled. The only weaknesses are minor repetition in the output templates and the absence of any progressive-disclosure split for the longer boilerplate sections.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is dense with executable commands, tables, and non-obvious gotchas (e.g., "Do NOT use `-d` (lowercase) since squash-merged branches won't be detected as merged by git") with essentially no explanation of concepts Claude already knows. Not score 5 because the summary-table templates repeat identical table headers across sections and a few procedural sentences could be trimmed. Not score 3 — over-explanation is minor, not 'some unnecessary explanation'. | 4 / 5 |
Actionability | Every step gives copy-paste-ready bash: `git for-each-ref --format=...`, the `gh pr view ... --jq` invocation, a concrete regex for worktree classification with a worked example, and exact `git worktree remove` / `git branch -D` commands. Guidance is fully executable and covers the common cases. | 5 / 5 |
Workflow Clarity | Eleven clearly numbered steps with explicit validation checkpoints: a detached-HEAD guard before anything runs, an ancestor/PR-state check before classification, a category table, a mandatory yes/no/pick confirmation before any destructive action, and error-recovery handling ("If removal fails (dirty worktree), warn the user and skip that branch — do not force-remove without explicit approval"). Despite being a destructive batch operation, validation and feedback loops are present throughout. | 5 / 5 |
Progressive Disclosure | The body is organized into well-labeled numbered sections that each map to a phase of the workflow, and no bundle files exist to reference. Not score 5 because the 5 anchor requires well-signaled one-level-deep references to separate files, and at ~220 lines some content (e.g., the full summary-output templates) could arguably live in a reference file. Not score 3 — structure is genuinely good, not merely present. | 4 / 5 |
Total | 18 / 20 Passed |