Content
52%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 has a clear four-phase debugging workflow and a well-signaled one-level-deep reference, but it is padded with generic debugging wisdom Claude already knows and contains no executable commands or code itself. Trimming known concepts and moving pattern detail toward the reference (or surfacing key commands inline) would tighten it considerably.
Suggestions
Cut sections that restate what Claude already knows — the scientific method, rubber duck debugging, the "Don't Assume" quotes, and the typo/null/cache checklist — and keep only novel, opinionated guidance.
Replace the fenced ```markdown pseudo-checklists with real markdown and inline a few executable commands (e.g., the git bisect sequence) so the body is actionable without opening the reference.
Move the pattern-by-issue-type detail (intermittent/performance/production) into the existing reference file or surface the reference link at the top of the relevant sections so it is not buried mid-document.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body spends significant space on concepts Claude already knows — the scientific method, rubber duck debugging ("Explain your code and problem out loud... Often reveals the issue"), the "It works on my machine" caveat, and a checklist of typos/null values/cache issues — matching 'noticeably verbose; several unnecessary explanations or padded sections'. | 2 / 5 |
Actionability | The checklists ask concrete questions ("Full stack trace", "Git history", "N+1 queries") but the body contains no executable code or commands — "Use Version Control: Git bisect, blame, history" only names tools, and the actual commands live in the reference file — fitting 'some concrete guidance but incomplete'. | 3 / 5 |
Workflow Clarity | The debugging process is clearly sequenced (Reproduce → Gather Information → Form Hypothesis → Test & Verify) with mostly present checkpoints ("Analyze: Did it prove/disprove your theory?"), but fix verification only appears implicitly in the mistakes list — 'clear sequence with most checkpoints present; minor validation gaps'. | 4 / 5 |
Progressive Disclosure | The single reference file exists, is one level deep, and is clearly signaled with trigger conditions ("Before selecting debugging tools or performing bisect, differential, trace, or memory analysis, read references/tools-and-techniques.md"), but the body inlines ~200 lines of pattern/checklist content and the reference pointer is buried mid-document — 'good structure; most content appropriately placed; references mostly clear; minor organization gaps'. | 4 / 5 |
Total | 13 / 20 Passed |