poke-webhooks

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly ingests and asks the agent to read/interpret webhook "data" (including error messages, logs, and URLs) delivered from external event sources (see references/payload-shape.md and the vercel/github-actions/cron patterns), so arbitrary/untrusted third‑party content in those payloads can influence the agent's summaries and actions.