observability

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

• Third-party content exposure detected (high risk: 0.80). The skill's SKILL.md explicitly instructs the agent to stream and parse runtime logs and NDJSON events from Vercel's public REST endpoint (e.g., https://api.vercel.com/v3/deployments//events) and to ingest drain payloads, which are untrusted, user-generated third‑party content that the agent is expected to read and that can materially influence decisions (error scans, remediation, downstream processing).