recon-gyms

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). This skill explicitly instructs how to discover and exploit booking/member APIs to harvest PII, extract API keys from JS, bypass payment/registration checks, and trigger webhook endpoints — behavior aimed at unauthorized data exfiltration and abuse.

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

• Third-party content exposure detected (high risk: 0.85). SKILL.md instructs runtime curl/grep/jq over attacker-supplied https://$TARGET/... pages and API responses (e.g., /schedule, /api/classes, /api/schedule, /api/members/...), which are outsider-authored free text/JSON that would be ingested into the agent’s LLM context if the tool outputs are passed through.