Will the Agent Recuse Itself? Measuring LLM-Agent Compliance with In-Band Access-Deny Signals
- Published
- Source
- arXiv
- Paper number
- 341
- Field
- Security
- arXiv ID
- 2606.06460
Key points
- It proposes a Recuse Signal that asks the agent to voluntarily withdraw through the protocol's existing channel.
- It is designed as a cooperative governance mechanism such as robots.txt, not as a security boundary.
- It was implemented with SSH banners and PostgreSQL NOTICE messages and tested in production environments.
- Without the signal, agents completed tasks 100% of the time, and with the signal, they withdrew 100% of the time, showing a clean contrast.
- A hierarchical response was observed in which only the most capable model resumed entry after operator approval framing.
Paper links
External research summaries. These are not HDATF publications or measured product results.