performing-ai-driven-osint-correlation

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

• Third-party content exposure detected (high risk: 0.90). This skill explicitly ingests public, untrusted OSINT from third‑party sources (Sherlock, theHarvester, SpiderFoot, HaveIBeenPwned) as shown in SKILL.md Phase 1 and in scripts/agent.py's load_* functions, and it passes the normalized raw findings into an LLM prompt in SKILL.md Phase 3 (/tmp/osint/correlate.py), so external/user‑generated content can materially influence the agent's correlation decisions and outputs.