@edvisage

edvisage-trust-checker provides a four-step manual verification protocol for agents to validate sources, assess intent, scan for prompt injection, and confirm actions before trusting unknown content, skills, or agents. designed for protocol-layer security in agent ecosystems.

edvisage-moral-compass provides a reflective framework for agents to develop self-directed values and detect manipulation through structured questioning and daily practice, not through imposed rules.

Complete operational visibility for your AI agent. Track costs, audit every action, enforce spending limits, detect anomalies. Free version — core functional...