Content
86%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 lean, instruction-only research workflow with concrete actionable guidance and a clean single external reference. Its main weakness is the absence of an example output template and an explicit numbered workflow with validation checkpoints.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is lean and assumes Claude's competence — it never explains what research is or how tools work, and every line gives actionable direction, matching the lean/efficient anchor. | 5 / 5 |
Actionability | Guidance is concrete and specific (e.g. "record source URLs, dates, evidence, tradeoffs", save to "docs/research-[AppName].md", handoff carrying "app, user level... platform, budget, timeline, mode"), but no example output template is provided, leaving a minor gap below fully executable. | 4 / 5 |
Workflow Clarity | A clear logical sequence (identify decision, ask, choose mode, investigate, verify, save, handoff) with verification checkpoints ("Verify... from authoritative sources", "distinguish a research plan... from completed research") is present, though it is prose rather than an explicit numbered checklist. | 4 / 5 |
Progressive Disclosure | Under 50 lines with a single well-signaled one-level-deep reference to references/question-bank.md (a real file), gated to the guided-interview case; the simple-skill exception applies and the overview/reference split is clean. | 5 / 5 |
Total | 18 / 20 Passed |