AI Agent Authorization Demo

Secure AI agent authorization, demonstrated.

See how the Curity Identity Server secures AI agents acting on a user's behalf. Each hop between agents, tools and APIs gets its own narrowly scoped token, privileged actions require step-up MFA and no workload holds a standing credential to the resources it reaches.

SRE Copilot
Secure AI agent operations

Ask the copilot

Pick a request. Privileged actions need step-up MFA, and some users are refused.

Every hop is authorized by the Curity Identity Server

Before each hop, the workload swaps its token at the Curity Identity Server for one scoped to the next hop only. The APIs only verify. The LLM provider never sees a user token.

  • request + its token
  • token exchange
  • request now
  • reached
  • refused
  • outside trust domain

Inside the exchange

Waiting · send a request to watch it travel
    Scope carried by the request
    Nothing in flight.

    Result

    The answer, the identity presented and the tools called.

    Send a request to see the answer.

    What the agent cannot do

    Enforced by policy, not by prompts or guardrails. Agents can only reach what the Curity Identity Server's policy allows, whoever the user is and whether or not they used MFA.

    Workload identities

    read path · 4 workloads

    Each workload has its own short-lived SPIFFE identity, used as the actor in every token exchange.

    The tools this user's token unlocks on each tier

    On-behalf-of chain

    One token per hop. act nests as authority is delegated; may_act names who may spend it next.

    No request in flight.