hunting-saas-sso-token-abuse

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

• Third-party content exposure detected (high risk: 0.85). The required workflow pulls Okta System Log events via the outsider-controlled Okta API (scripts/agent.py → fetch_logs() → requests.get("https://{org}/api/v1/logs"...).json()), and those events contain free-text fields like client.userAgent.rawUserAgent and actor.alternateId that are then ingested into the agent’s runtime context for analysis/output.