Security
Decision-support red teaming — with clear limits
AgentRedTeam processes the agent descriptions and probe inputs you submit. This page states what we handle and what we do not claim.
What we handle
Agent descriptions, capability lists, and threat-focus notes used to generate adversarial probe findings.
Data handling commitments
- pages.secData1
- pages.secData2
- pages.secData3
Honesty rule: no guarantee language, no 100% claims, no invented customer metrics.
Our compliance posture
- pages.secPosture1
- pages.secPosture2
Subprocessors & payments
- pages.secSub1
- pages.secSub2
refs: OWASP Top 10 for LLM Applications · OWASP Top 10 for AI Agents · NIST AI RMF