Collected sources and patterns will appear here. Add from search or the patterns library.
TaskCompletionState + EvaluationSchema -> EvaluationGrade
Evaluate agent task completions using direct LLM API calls to bypass agent-specific personality or prompt injection wrappers.
Problem it solves
Agent-system prompts and wrappers can contaminate the objectivity of evaluation metrics.
Consumes
Emits
The real projects this mechanism was found in. Attribution is the point — this is how the best teams actually do it.