prompt-injection-risk-linter

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

• Third-party content exposure detected (high risk: 1.00). The skill's SKILL.md and prompt_injection_risk_linter.py explicitly accept and read an optional --retrieved text file of representative untrusted content (e.g., "web pages, emails, tickets"), and the script analyzes that retrieved_text for prompt-injection patterns which can materially influence the linter's findings and exit behavior.