planning-with-files

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

• Third-party content exposure detected (high risk: 0.90). The skill's main workflow explicitly ingests open web/search results into findings.md (see SKILL.md: "Write web/search results to findings.md" and the "Security Boundary" / hooks that read task_plan.md, progress.md, and findings.md) and requires the agent to read findings.md before making decisions, so untrusted third‑party content can materially influence the agent's actions.