A GenLayer policy gateway that verifies AI agent actions against user rules and live evidence before execution.
AgentGuard adds a verifiable safety layer between AI agents and sensitive actions.
Users provide:
- A policy describing the rules the agent must follow
- A proposed agent action
- An HTTPS evidence source
The AgentGuard Intelligent Contract retrieves the external evidence, evaluates the action against the policy, reaches a GenLayer consensus decision, and records the verification result on-chain.
User
↓
AgentGuard Web App
↓
Policy + Agent Action + Evidence URL
↓
GenLayer Intelligent Contract
↓
Contract-side HTTPS Evidence Retrieval
↓
Policy + Evidence Evaluation
↓
GenLayer Consensus
↓
ALLOW / BLOCK
↓
On-chain Verification Record
E0F