attack-surface-analyst

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

• Third-party content exposure detected (medium risk: 0.65). SKILL.md describes runtime discovery via public sources/APIs (e.g., “crt.sh API”, “passive DNS databases”, “certificate transparency log parsing”) and scanning of internet-facing endpoints, which can ingest outsider-authored free text (public web/API responses and scraped log content) into the agent’s LLM context if the implementation passes those responses through to the model.