requirements-patterns

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

• Third-party content exposure detected (high risk: 0.75). The skill reads required outsider-authored free text from projects/<ProjectName>/_context/features.md and business_rules.md (and optionally requirements_analysis_report.md) and then extracts conditions/actions to generate LLM output, so any third-party-authored content in those files can be ingested into the agent’s LLM context.