ID
Agent identity
Register persistent machine identities with ownership, purpose, status and verification metadata.
Identity, traceability and certification for AI agents
Trust infrastructure for autonomous AI agents.
Agentracert helps organisations register AI agents, define what they are allowed to do, trace their actions, and build reliable trust profiles for autonomous software.
Initial prototype: free, developer-friendly, and designed for agentic AI integrations.
The missing trust layer for agentic AI.
As agents begin negotiating, buying, scheduling and executing workflows, organisations need more than model capability. They need identity, authority, traceability, revocation and reputation.
✓
Delegated authority
Define what each agent may do, where approval is required, and when permissions should expire.
⌁
Traceable audit logs
Capture structured event trails for decisions, tool calls, policy checks, escalations and failures.
!
Revocation
Suspend, quarantine or revoke agents when behaviour becomes risky, non-compliant or unauthorised.
R
Reputation signals
Track reliability, policy violations, task success rates and human overrides to form trust profiles.
API
Integration-first
A lightweight API and SDK layer for agent frameworks, enterprise automations and future agent economies.
MVP v1: passport + black box recorder.
The first version focuses on a working trust registry: register an agent, assign permissions, ingest events, show audit history, flag policy violations, and generate a basic trust profile.
POST /api/v1/events
{
"agent_id": "procurement-agent-01",
"event_type": "purchase_attempt",
"vendor": "GPUCloud Ltd",
"amount": 4200,
"policy_status": "allowed"
}