spec-driven

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

• Third-party content exposure detected (high risk: 0.80). The skill's design and research workflows (design.md Step 5 and research.md) explicitly instruct the agent to "use available documentation lookup and web search tools" and to "research and create cache (.artifacts/research/{topic}.md)", meaning it fetches and interprets open web documentation/public content as part of required design steps that can influence decisions and actions.