add-skill

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

• Third-party content exposure detected (high risk: 0.75). The required workflow for add-skill includes “Gather: web-search the topic for current facts, official docs, and prior art” and “Read .local/ recursively,” so at runtime the agent may ingest outsider-authored free text from public web pages (and any downloaded/forwarded .local/ files) into the LLM context.