Content
80%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.
A tight, well-structured instruction-only workflow that assumes Claude's competence and names concrete tools and fields. Its main gap is the absence of validation/verification checkpoints in a batch screening process, plus a lack of example queries.
Suggestions
Add explicit validation checkpoints, e.g., reconcile counts against a PRISMA flow diagram after screening.
Include example broad/narrow search query strings or concrete API invocations for the named databases.
Add a brief verification step to confirm inclusion/exclusion criteria were applied consistently.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is lean and directive with no padding or explanations of concepts Claude already knows; every line earns its place, matching the 'lean and efficient' anchor. | 5 / 5 |
Actionability | Provides concrete, specific guidance (named databases Semantic Scholar/arXiv/OpenAlex, exact extraction fields method/dataset/metrics) but lacks example queries or actual API/command invocations, leaving minor gaps. | 4 / 5 |
Workflow Clarity | Steps are clearly sequenced 1-8, but there are no validation checkpoints (e.g., PRISMA flow reconciliation, consistency checks); screening many papers is a batch operation, so the missing validation cap applies. | 3 / 5 |
Progressive Disclosure | Under 50 lines with no external references needed and a well-organized single section plus a clean numbered list, satisfying the simple-skill exception for progressive disclosure. | 5 / 5 |
Total | 17 / 20 Passed |