Outbound Wiki

Article

The AI Agent Code of Conduct: Automated Guardrail Policy-as ...

arxiv.org

Open at publisher

Quoted on this wiki

Every place a page here uses this source, in the order the words come in it.

  1. 6Limitations Future work includes leveraging production interaction logs (inputs, outputs, tool calls, actions) to mine candidate rules and hard examples that continuously enrich the policy tree and prompts References

    In AI outbound prompting

  2. Abstract We introduce "Policy as Prompt," a new approach that uses Large Language Models (LLMs) to interpret and enforce natural language policies by applying contextual understanding and the principle of least privilege. 1Introduction

    In AI SDR agent guardrails

  3. Abstract Our system first ingests technical artifacts to construct a verifiable policy tree, which is then compiled into lightweight, prompt-based classifiers that audit agent behavior at runtime. 1Introduction

    In AI SDR agent guardrails