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.
A well-sequenced, highly actionable procedural skill with strong error-recovery feedback loops and no concept over-explanation. Minor verbosity in recovery prose and a dangling rationale.md reference keep it just short of top marks.
Suggestions
Tighten the step-3 "Tool error" and "Denied" recovery prose to bullet-level brevity; the operational detail can be compressed without losing the routing logic.
Resolve the maintainer reference to rationale.md — either add the file to a references/ bundle or remove the HTML comment, so no dangling reference remains.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is dense and operational, assuming Claude knows what a worktree is, but the step-3 recovery and "Denied" prose run long and could be trimmed without losing clarity. | 4 / 5 |
Actionability | Concrete, copy-paste-ready commands and tool calls throughout (`EnterWorktree({name: "<branch>"})`, `wt -C <repo> switch --create <branch> --no-cd --format=json`, `git stash push -u`, `cd <path>`) cover the common cases. | 5 / 5 |
Workflow Clarity | Steps 1–3 are clearly sequenced with explicit validation branches ("Branch already exists", tool-error `cd` recovery, "Shell cwd was reset") and feedback loops that route back to user action before continuing. | 5 / 5 |
Progressive Disclosure | Content is well organized into Arguments / What to do / Cleanup / Scope sections in a single appropriately self-contained file, but the maintainer note references a non-existent `rational.md`/`rationale.md`, a minor dangling-reference gap. | 4 / 5 |
Total | 18 / 20 Passed |