Dokumentationerreichbar
OpenAI Agents SDK, Guardrails
openai.github.io (externe Seite)
Der Werkzeugkasten von Open
geprüft 10.11.2026
Worauf sich diese Seite beruft, wörtlich, abgerufen am 07.08.2026:
Guardrails enable you to do checks and validations of user input and agent output.
bestätigt 10.11.2026Input guardrails run only for the first agent in the chain.
bestätigt 10.11.2026Output guardrails run only for the agent that produces the final output.
bestätigt 10.11.2026Input tool guardrails run before the tool executes and can skip the call, replace the output with a message, or raise a tripwire.
bestätigt 10.11.2026Blocking execution guarantees that the expensive model does not start; with parallel execution, the expensive model may already have started before the guardrail completes.
bestätigt 10.11.2026If you need checks before and/
bestätigt 10.11.2026or after each custom function-tool call in a workflow that includes managers, handoffs, or delegated specialists, use tool guardrails instead of relying only on agent-level input/ output guardrails. If a function tool requires approval, input tool guardrails normally run after approval and immediately before execution.
bestätigt 10.11.2026