Design capability probes to map {{model}} abilities on {{capability_domain}}. Generate probe set: 1. Baseline capability tests (5 probes) 2. Advanced capability tests (5 probes) 3. Limit-finding probes (3 probes) 4. Emergent behavior detection probes (2 probes) For each probe, specify expected response patterns and scoring criteria against {{capability_levels}}.
80 copies0 forks
Details
Category
AnalysisUse Cases
Capability discoveryProbe designLimit mapping
Works Best With
claude-opus-4.5gpt-5.2gemini-2.0-flash
Created Shared