docs

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

• Third-party content exposure detected (high risk: 0.80). The /docs research workflow (references/research.md, "Scan for candidates" — e.g., "Research competitors, adjacent tools, and market shifts" and "Check issue trackers, feature request backlogs, and support channels") instructs the agent to read external public third‑party content (competitor sites, public issue/support channels), which the agent is expected to interpret and use to produce recommendations and roadmap actions, so untrusted third‑party content can materially influence decisions.