Framework-agnostic confidence-gated escalation middleware for LLM agents: multi-signal scoring (logprob, verbalized, tool risk), threshold policies, and escalation handlers for LangChain, CrewAI, AutoGen, and Google ADK.
python ai-safety escalation autogen llm ai-governance langchain enterprise-ai llm-agents crewai eu-ai-act google-adk confidence-scoring human-in-loop regulated-industries owasp-agentic-ai agent-middleware multi-signal threshold-policy logprob
-
Updated
May 2, 2026 - Python