refs

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

• Third-party content exposure detected (high risk: 0.85). The skill syncs “awesome-*” reference repositories from public/community sources into local files and then other SDLC skills can Read/Grep those README/body texts, so outsider-authored free text from those repos can be ingested into the agent’s LLM context via the local reference repo contents.