Rubric & Evaluator Templates

Configure multi-agent perspectives, weights, LLM models, and adjudication thresholds.

Available Templates

0 preset configurations

Loading templates…

Create New Template

Define LLM agents, consensus perspectives, and rubric weights.

= 20%

Triggers human review or adjudicator LLM when evaluator scores diverge past this gap.

High-capability reasoning model used to resolve agent disagreements.

Evaluator Agents (1)

Each agent focuses on an independent perspective with custom weight and prompting.

1accuracy Perspective
Total Agent Weight Sum:35 (Recommend 100)