Technically AcceptableIndependent technical publication
← Technically Acceptable
Programs

Programs

Sustained technical work that produces the experiments, systems, and evidence behind the writing.

Program Active
AI agent security

Sentinel

Controlled experiments in prompt injection, privileged tool use, architectural containment, and agent reliability.

Current questionDoes poisoned input increase accepted unauthorized send_email attempts when a worker is given a contained action channel?

Explore Sentinel