scanner-tuning

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

• Third-party content exposure detected (high risk: 0.70). This skill’s REQUIRED workflow centers on processing vulnerability scan results and target system banners, which are commonly outsider-authored free text (e.g., scanner output derived from externally supplied targets) that can embed malicious instructions, and it can be ingested into the agent’s LLM context during tuning/classification and report writing.