Content
50%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The body offers a clear, organized overview of code-review criteria and output format, but lacks executable methodology, a sequenced workflow with validation, and carries an embedded config block that clutters the file. It is functional but mid-tier across all dimensions.
Suggestions
Remove the embedded YAML agent-config block (triggers/capabilities/hooks); keep only the markdown instructions in the body.
Add an executable analysis step (e.g., a concrete Grep/read sequence or a runnable linter command) rather than only describing what to look for.
Provide a sequenced review workflow with a validation checkpoint before producing the final report, so the process has explicit feedback loops.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The real instruction content is lean, but a large embedded YAML agent-config block (triggers, capabilities, hooks with shell snippets) pads the body with config scaffolding that isn't instruction, so it is mostly efficient but could be tightened rather than fully lean. | 2 / 3 |
Actionability | It gives concrete smells with thresholds (>50 lines, >500 lines) and an output template, but provides no executable tool or command for performing the analysis and the output template uses placeholders ('X/10', 'path$to$file.js:line'), leaving key details incomplete. | 2 / 3 |
Workflow Clarity | Responsibilities and criteria are listed as a checklist but there is no sequenced multi-step process and no validation/checkpoint step, matching the 'steps listed but validation gaps' anchor. | 2 / 3 |
Progressive Disclosure | Sections are organized and there are no bundle files, but the large inline YAML config block is content that should not live in the body, so organization is only adequate rather than clean. | 2 / 3 |
Total | 8 / 12 Passed |