remediating-with-aws-security-agent

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (medium risk: 0.65). Outsider-authored free text is ingested into the LLM context via AWS Security Agent API responses in Stage 2/3—specifically the exported finding fields (e.g., description, reasoning, attackScript, remediationCode) returned by list-findings/batch-get-findings and then read back from .security-agent/ for triage, where those fields can contain exploit/reproduction text authored by the scanning system rather than the operating user.