The uncomfortable part of building for agents is that the old assumptions quietly stop holding. There is no browser to redirect, no person to read a consent screen at the moment it matters, and no obvious place to put the word "no".
This piece walks through how we think about it at Paycux, what we have changed our minds about, and where the sharp edges are.
The Gold Rush phase (aka the part where everyone forgets security)
Consider the gold rush phase (aka the part where everyone forgets security). 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.
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.
Famous last words: “It’s fine, it’s just dev tooling”
Consider famous last words: “it’s fine, it’s just dev tooling”. 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.
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.
- 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
Supply chain reality check: CVE-2025-6514
That brings us to supply chain reality check: cve-2025-6514. 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.
The first malicious MCP server in the wild
The first malicious MCP server in the wild 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.
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.
The audit trail matters more here than in any human flow.
Favorite horror story: drive-by localhost attacks
Consider favorite horror story: drive-by localhost attacks. 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.
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.
The problem is not MCP. It’s the gap between power and discipline
The problem is not MCP. It’s the gap between power and discipline is where this gets concrete. 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.
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.
What secure MCP actually looks like
That brings us to what secure mcp actually looks like. 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.
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.
Where this leaves us
None of this is exotic. It is the ordinary discipline of deciding what you own, writing down what you assume, and making the failures loud enough to notice.
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.