Agents changed the shape of the problem. A human signing in is a single decision made once; an agent acting on someone's behalf is a decision that has to hold for every call it makes afterwards, often hours later, often without anyone watching.
This piece walks through how we think about it at Paycux, what we have changed our minds about, and where the sharp edges are.
Why the model–agent layer is high risk
That brings us to why the model–agent layer is high risk. The audit trail matters more here than in any human flow. When something goes wrong, the question is never "did the user intend this" in the abstract — it is which agent, acting under whose authority, made which call, and whether the record can prove it.
Consent is the part most implementations get wrong. Asking once at install time and then acting indefinitely is not consent; it is a standing grant with no expiry and no visibility.
Prompt injection & command confusion
Prompt injection & command confusion deserves its own treatment. An agent's credential should describe what it may do, not who owns it. Scope it to the narrowest set of operations that make the task possible, bind it to a single principal, and give it a lifetime measured in minutes rather than days.
An agent's credential should describe what it may do, not who owns it. Scope it to the narrowest set of operations that make the task possible, bind it to a single principal, and give it a lifetime measured in minutes rather than days.
- Scope every agent credential to one principal and one task
- Give tokens minutes of life, not days
- Record which agent acted, under whose authority, on what
- Make revocation a single call that takes effect immediately
Over-privileged agents
That brings us to over-privileged agents. Consent is the part most implementations get wrong. Asking once at install time and then acting indefinitely is not consent; it is a standing grant with no expiry and no visibility.
Consent is the part most implementations get wrong. Asking once at install time and then acting indefinitely is not consent; it is a standing grant with no expiry and no visibility.
Man-in-the-Middle (MitM) risks
Man-in-the-Middle (MitM) risks deserves its own treatment. Consent is the part most implementations get wrong. Asking once at install time and then acting indefinitely is not consent; it is a standing grant with no expiry and no visibility.
Consent is the part most implementations get wrong. Asking once at install time and then acting indefinitely is not consent; it is a standing grant with no expiry and no visibility.
Consent is the part most implementations get wrong.
Replay & cross-session attacks
Replay & cross-session attacks is where this gets concrete. The audit trail matters more here than in any human flow. When something goes wrong, the question is never "did the user intend this" in the abstract — it is which agent, acting under whose authority, made which call, and whether the record can prove it.
Consent is the part most implementations get wrong. Asking once at install time and then acting indefinitely is not consent; it is a standing grant with no expiry and no visibility.
Cross-agent lateral movement
Cross-agent lateral movement deserves its own treatment. The audit trail matters more here than in any human flow. When something goes wrong, the question is never "did the user intend this" in the abstract — it is which agent, acting under whose authority, made which call, and whether the record can prove it.
Consent is the part most implementations get wrong. Asking once at install time and then acting indefinitely is not consent; it is a standing grant with no expiry and no visibility.
Data exfiltration (model memorization & deliberate leak)
Consider data exfiltration (model memorization & deliberate leak). An agent's credential should describe what it may do, not who owns it. Scope it to the narrowest set of operations that make the task possible, bind it to a single principal, and give it a lifetime measured in minutes rather than days.
Consent is the part most implementations get wrong. Asking once at install time and then acting indefinitely is not consent; it is a standing grant with no expiry and no visibility.
Where this leaves us
If there is one thing worth taking away, it is that the expensive decisions are the ones made implicitly. Making them on purpose costs an afternoon.
If you are working through the same problem and want to compare notes, the docs cover the mechanics and the console shows the behaviour on your own data.
Everything here, already built
Sign-in, enterprise SSO, directory provisioning, roles and an audit trail behind one API. Start with the quickstart and have a working sign-in this afternoon.