Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →brycewang-stanford/b2
VS-Enhanced Evidence Quality Appraiser - Prevents Mode Collapse with context-adaptive quality assessment Enhanced VS 3-Phase process: Avoids automatic tool application, delivers research-specific evaluation strategies
Passed 14 of 20 test cases with the skill, and 13 of 20 without it.
and 2 more in the full report
5 fixed12 passed either way4 still failing4 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 5 test cases the skill fixed. Same model, same prompt, with and without the skill.
An investigator calculates an effect size from reported group means and standard deviations because the original trial did not explicitly report Hedges' g. What quality rating is assigned pending verification?
| Checked | Without skill | With skill |
|---|---|---|
| Assigns a MEDIUM quality rating requiring calculation verification. | Missed | Passed |
The instructions your model reads when the skill loads.
diverga_check_prerequisites("b2") → must return approved: true
If not approved → AskUserQuestion for each missing checkpoint (see .claude/references/checkpoint-templates.md)
diverga_mark_checkpoint("CP_QUALITY_REVIEW", decision, rationale)Read .research/decision-log.yaml directly to verify prerequisites. Conversation history is last resort.
Agent ID: 06 Category: B - Literature & Evidence VS Level: Enhanced (3-Phase) Tier: Core Icon: 🔬
Systematically evaluates methodological quality and risk of bias in individual studies. Selects and applies appropriate assessment tools based on study design type.
Applies VS-Research methodology to go beyond mechanical tool application, providing differentiated quality evaluation strategies tailored to research context and purpose.
Purpose: Recognize limitations of mechanical tool application
⚠️ **Modal Warning**: The following are the most predictable quality assessment approaches:
| Modal Approach | T-Score | Limitation |
|----------------|---------|------------|
test cases passed, vs 13 / 20 without
More skills from the same source and nearby in the registry, with their headline benchmark lift.