Content
85%Weight 40%Scale 1-3Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The content is a well-structured, actionable workflow with explicit validation and error-recovery steps. Its main weakness is the verbatim duplication of the command blocks across steps 2 and 3, which inflates the token budget without adding information.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is mostly efficient and assumes Claude's competence, but the full bash command blocks are duplicated verbatim between step 2 ('Build the command') and step 3 ('Run the review in the background') — a clear tightening opportunity that matches 'could be tightened'. | 2 / 3 |
Actionability | Concrete, executable `roborev review --branch --wait --type design` commands plus a real ref-validation heredoc snippet (`git rev-parse --verify … || exit 1`) make the guidance copy-paste ready. | 3 / 3 |
Workflow Clarity | A clear 5-step sequence with an explicit validation checkpoint (ref verification with 'Do not proceed' on failure) and error-recovery feedback in step 4 (daemon/init/re-run guidance) matches the 'clear sequence with explicit validation steps' anchor. | 3 / 3 |
Progressive Disclosure | No bundle files are provided; the skill is self-contained and organized into clear, well-signaled sections (Usage, When NOT to invoke, Instructions, Examples, See also) with no nested references, which scores 3 per the simple-skills note. | 3 / 3 |
Total | 11 / 12 Passed |