Content
53%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.
The body is a well-structured agent persona with concrete tool references and a clear workflow sequence, but it leans on high-level direction over executable detail and lacks validation checkpoints in its workflow.
Suggestions
Replace abstract workflow steps with concrete, runnable commands (e.g., exact jscpd and ast-grep invocations) to lift actionability.
Add an explicit verification/checkpoint step in the workflow, such as cross-checking detected anti-patterns against the naming and duplication results before reporting.
Trim the persona framing and generic "When analyzing code" bullets to reduce token overhead without losing actionable guidance.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Mostly efficient persona-and-responsibilities prose that avoids explaining basic concepts, but the "You are a Code Pattern Analysis Expert" framing and the generic "When analyzing code" bullets could be tightened. | 3 / 5 |
Actionability | Names concrete tools and signals (Grep, ast-grep, jscpd with --min-tokens 50, TODO/FIXME/HACK/XXX), but the per-responsibility guidance remains high-level direction rather than executable, copy-paste steps. | 3 / 5 |
Workflow Clarity | A clear six-step sequence is present, but steps are abstract (e.g., "Analyze naming conventions by sampling representative files") with no explicit validation or verification checkpoints. | 3 / 5 |
Progressive Disclosure | The single file is well-organized into distinct sections (responsibilities, workflow, deliverables, considerations) with no nested references; slightly over 50 lines so not a perfect 5. | 4 / 5 |
Total | 13 / 20 Passed |