Content
65%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The body is highly actionable with copy-paste-ready commands and config, and steps are clearly sequenced. It loses points for a decorative diagram and restated output that inflate the token budget, and for a batch git-hook step missing an explicit validation feedback loop.
Suggestions
Remove the ASCII "Two-Layer Review Strategy" diagram (or fold it into one sentence) since the two-layer concept is already clear from the steps, and trim the "Output" section that restates deliverables.
Add a validation/feedback checkpoint to the pre-push hook — e.g., capture and check `cr review`'s exit code and surface a clear pass/fail decision rather than silently running non-blocking.
Move the full .coderabbit.yaml and VS Code tasks.json examples into reference files (e.g. references/config.yaml, references/vscode-tasks.json) and link to them, so the main body stays a lean overview.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Mostly lean executable code, but the decorative ASCII "Two-Layer Review Strategy" diagram, the restated "Output" section, and the inline VS Code tasks.json add tokens that could be tightened without losing clarity. | 2 / 3 |
Actionability | Fully executable, copy-paste-ready commands and complete config files throughout — install/auth commands, `cr review` variants, a real pre-push hook script, and a complete .coderabbit.yaml. | 3 / 3 |
Workflow Clarity | Steps are clearly sequenced and Step 1 includes a `cr --version` verify, but the batch pre-push git-hook operation (`xargs cr review`) has no validation checkpoint or failure feedback loop, capping this dimension at 2. | 2 / 3 |
Progressive Disclosure | Sections are well-organized, but at ~170 lines the skill has no reference/detail files and keeps full config and IDE-integration content inline that could be split out for a skill this length. | 2 / 3 |
Total | 9 / 12 Passed |