Enforce
Agent Identity
Every agent gets a verifiable identity bound to the workload it runs as, the human or service that owns it, and the scope it was issued. Credentials are short-lived and attested, so a leaked token is a narrow, expiring problem.
The problem
Agents typically inherit a developer's credentials or a shared service account. When something happens, the audit trail says a human did it, and the blast radius is that human's entire access.
Capabilities
How Agent Identity works
- 01
Identity bound to workload
An agent identity is issued against attested workload properties, not a long-lived string copied into an environment variable.
- 02
Ownership chain
Every agent has an accountable owner. The chain from action to person survives delegation, so escalation has a destination.
- 03
Trust graph
Which agents may invoke which other agents, and with what reduction in scope, is declared explicitly and rendered as a graph you can inspect.
- 04
Federated with your IdP
Human approvers, groups, and roles come from the directory you already run — Okta, Entra, Google — via OIDC. Cirvix does not become a second source of truth for who works here.
In practice
What it looks like
{ "agent_id": "agt_7Q2X9F", "name": "release-notes-writer", "owner": { "type": "user", "id": "u_4821", "email": "priya@example.com" }, "runtime": { "framework": "claude-code", "version": "2.1.x" }, "scopes": ["repo:read", "issues:read", "slack:post"], "environments": ["staging"], "credential": { "type": "short-lived", "ttl_seconds": 900, "attested": true }, "delegation": { "may_invoke": ["agt_1B4K7C"], "scope_reduction": "required" }}Limits
What this does not do
- Workload attestation is only as strong as the platform underneath it. On a developer laptop the identity is bound to a device-held key, which is weaker than a hardware-rooted attestation in a managed cluster, and the console labels it as such.
- Cirvix does not replace your identity provider. If a person's directory account is compromised, their approval authority is compromised with it.
Get started
Let the agents do real work.
Start with the free local scanner — it needs no account and tells you what is already ungoverned on your machine. Move to a governed runtime when you are ready.
Cirvix AgentControl is in private beta. Pricing is published and plans are not yet chargeable — early access confirms your place and we will tell you before anything bills.