Content
78%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-organized, concise methodology guide with concrete frameworks and named tools. It would benefit from a worked example search string and an extraction-form template to raise actionability.
Suggestions
Add one concrete worked example of a complete Boolean search string to bridge the gap from guidance to execution.
Include a minimal data-extraction form template (fields to capture per study) to make the screening/synthesis step fully actionable.
Add an explicit validation checkpoint in the PRISMA flow (e.g., confirm deduplicated count equals sum of database hits minus duplicates) to close the minor workflow gap.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Mostly efficient bullet-style instructions that assume Claude's competence, with only minor explanatory parentheticals (e.g., 'typically English', 'last 5-10 years for rapidly evolving fields') that add specificity but could be trimmed. | 4 / 5 |
Actionability | Provides concrete, executable guidance with named frameworks and tools (PICO, PRISMA, Cochrane Risk of Bias, Newcastle-Ottawa Scale, specific databases), but lacks worked example search strings or extraction-form templates that would make it fully copy-paste ready. | 4 / 5 |
Workflow Clarity | Clear multi-step sequences (PRISMA flow with recorded counts at each stage, two-pass screening) with most checkpoints present; minor validation gaps such as no explicit reconciliation check for deduplication counts. | 4 / 5 |
Progressive Disclosure | Under 50 lines with no external bundle files needed, and content is cleanly organized into well-labeled sections (Search Strategy, Inclusion/Exclusion, PRISMA, Screening, Synthesis) that satisfy the simple-skill exception. | 5 / 5 |
Total | 17 / 20 Passed |