Skip to content
Tervos Protocol

Tervos Protocol

Identity, policy enforcement, and a full audit trail for your AI agents.
Quickstart Protocol Specification

What is Tervos Protocol?

Section titled “What is Tervos Protocol?”

Tervos Protocol defines the organizational layer for agentic AI. It sits above agent communication protocols (MCP, A2A) and below application-level agent frameworks, providing the missing infrastructure for agent-driven organizations.

What it gives you:

  • Identity — every agent gets a cryptographic DID, not a string label
  • Policy enforcement — every action passes through Cedar policy evaluation before executing; default-deny
  • Audit trail — every state change is an immutable event in an append-only log
  • Progressive autonomy — agents start supervised and earn autonomy through explicit policy grants

Get started

Section titled “Get started”
Terminal window
npx tervos init --template demo
tervos start --dev
tervos audit

Three agents, 20+ events, full lifecycle — in under 5 minutes. See the Quickstart for a walkthrough.