datadog

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

• Third-party content exposure detected (high risk: 0.80). The SKILL.md explicitly queries Datadog's external APIs (e.g., the "Search logs" call to https://api.${DD_SITE}/api/v2/logs/events/search and the "Get monitor details" outputting the monitor "message"), which returns arbitrary log/event/message text from a third‑party service that the agent is expected to read and could materially influence actions like muting monitors or posting events.