quality-auto-test

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

• Third-party content exposure detected (high risk: 0.75). During S_SOURCE the skill parses and extracts scenarios from outsider-authored files like REQ-*.md (spec route) and verification/coverage artifacts, and those markdown/text contents are ingested into the LLM context to generate scenarios.csv and downstream test-writing prompts.