The safeguards map / 3.2 Agent harnesses
OpenAI Agents SDK
Drafted with AI
Drafted with AI from the cited sources under the direction of UnGovr staff. UnGovr holds editorial responsibility for this page. How this site is made
A Python agent SDK with input and output guardrails that can stop a run, and a human-in-the-loop flow that pauses sensitive tool calls for approval.
Where it sits
3.2 Agent harnesses (primary): Runs guardrail checks on an agent's input and output with a tripwire that halts the run, and pauses tools marked as needing approval until a person approves or rejects.
Guardrails enable you to do checks and validations of user input and agent output.
captured
If the guardrail tripwire is triggered, the agent never executes, preventing token consumption and tool execution.
captured
Use the human-in-the-loop (HITL) flow to pause agent execution until a person approves or rejects sensitive tool calls.
captured
Facts
- Concern
- safetysecurity
- Names the law
- no
- Subject
- agent
- Used by
- those who make it
- Runs
- at runtime
- Policy comes from
- owner-entered settings
- Can the agent bypass it
- depends
Handoffs
None recorded.
Listed, not drawn: no documented handoff and no foundation home places it on the map.
Every quote is verbatim from a stored capture of the page linked under it. Placement is our reading of the product's own documentation; no publisher has reviewed or endorsed it. Rows CC BY 4.0. Names, marks and logos belong to their owners; logos identify each product and are not part of the dataset.