research-plan-implement

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

• Third-party content exposure detected (high risk: 1.00). The skill explicitly spawns a "web-researcher" subagent in Phase 1 (Step 2) that is instructed to research external web sources and write findings to docs/plans/YYYY-MM-DD--external.md, so it fetches and ingests untrusted public web content that will be read and used to drive planning and implementation.