Content
81%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-structured, actionable workflow with strong validation and clear sequencing appropriate to its destructive nature. The main improvement opportunity is tightening redundancy between the process steps and the Common Mistakes/Red Flags sections.
Suggestions
Consolidate 'Common Mistakes' and 'Red Flags' into a single concise checklist, or fold each rule into its relevant step, to reduce redundancy with the process body.
Replace the generic '<test command>' placeholder and the 'npm test / cargo test / pytest / go test' list with a single instruction to run the project's configured test command, keeping examples in a comment.
Tighten the Overview/Core principle lines so they do not restate what the Quick Reference table already conveys.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Mostly lean with tables and tight code blocks, but the 'Common Mistakes' and 'Red Flags' sections partly restate guidance already embedded in the process steps, adding some redundancy. | 4 / 5 |
Actionability | Concrete, executable bash commands throughout (git checkout, merge, push, worktree remove, prune) with clear placeholders; minor gaps like the generic '<test command>' and the 'npm test / cargo test / ...' list keep it just short of fully copy-paste ready. | 4 / 5 |
Workflow Clarity | A clearly sequenced 6-step process with explicit validation checkpoints ('Stop. Don't proceed to Step 2', 'Verify tests on merged result', 'Only after merge succeeds'), typed discard confirmation, provenance checks, and Always/Never checklists for this destructive workflow. | 5 / 5 |
Progressive Disclosure | Self-contained SKILL.md with well-organized sections (Overview, Steps 1–6, Quick Reference, Common Mistakes, Red Flags) and a summary table; no bundle files exist, and the inline reference-like material (Common Mistakes/Red Flags) introduces minor organization gaps rather than true multi-file structure. | 4 / 5 |
Total | 17 / 20 Passed |