Content
78%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.
Highly actionable and well-structured with a clean single-level reference, but the destructive/batch recovery workflows lack explicit validation checkpoints, which caps workflow clarity.
Suggestions
Add explicit validation steps before destructive operations, e.g. run tests or `git log` to confirm before `git push --force-with-lease` and `git reset --hard`.
In the Reflog recovery section, add a verify checkpoint such as `git log --oneline` or checking `git status` before finalizing the recovery.
Trim minor filler phrases ('Swiss Army knife', 'Your safety net') to tighten the body further.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Largely lean with command tables and code blocks assuming Claude's competence; minor over-explanation ('the Swiss Army knife of Git history editing', 'Your safety net - tracks all ref movements') could be trimmed. | 4 / 5 |
Actionability | Provides copy-paste-ready, fully executable git commands across every technique (rebase ops, cherry-pick ranges, bisect run scripts, worktree add/remove, reflog recovery) covering the common cases. | 5 / 5 |
Workflow Clarity | Section 4 'Reflog' gives a recovery sequence without a verify/confirm checkpoint, and destructive/batch operations (force-push, reset --hard) lack explicit validate-before-proceeding feedback loops; the destructiveness cap limits this to 3. | 3 / 5 |
Progressive Disclosure | Clear overview in SKILL.md with a single well-signaled one-level-deep reference ('Read that file when the navigation tier above is insufficient') to the real references/details.md; content is appropriately split. | 5 / 5 |
Total | 17 / 20 Passed |