EVERY STEP WAS ALLOWED

Governing agents —
prove what they can't do

An agent can take a sequence of individually authorized actions whose net effect is harm. Fetch untrusted content, read a secret, send — each call passes its own check, and the coexistence is the leak. Per-call inspection has no vocabulary for it, because nothing is tracking what the session has accumulated.

Start here

Three questions, three tools

They answer different things and they compose: what an agent may reach at all, what a single run may do, and how several agents share one run without doing anything twice.

The argument underneath

A leak is a reachable state, not a clever prompt

Prompt injection, goal hijacking and tool misuse name ways to reach a bad state. The state itself is the thing worth checking — and that is what a red line is: a situation declared off-limits in advance, not a phrase to watch for in a message. That reframing is what the two engines below are built on. Neither is distributed; both are described.

Stop asking the agent whether it stayed in bounds. Check that the bounds hold, on every path, before it runs.

Reading

Why this is architectural, not operational

Three essays working the same argument through different sectors.

Next

Where this connects

The same substrate verifies the stateful code an agent writes — that is the SAM v2 gate, seven tools around one deterministic check. If you want the guardrail work applied to your own stack, that is what the advisory engagements are for.