add-skill

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

• Third-party content exposure detected (high risk: 0.85). The REQUIRED workflow is “Read .local/ recursively — every subfolder and file is source material” and “research beyond it: web-search …” (runtime), which can ingest outsider-authored free text into the agent’s LLM context; exclusion doesn’t apply because neither .local/ nor web search content is guaranteed to be authored by the operating user.