datadog

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt explicitly requires the agent to ask the user for DD_API_KEY and DD_APP_KEY and then store/use them in API calls/commands, which forces the LLM to receive and potentially emit those secret values verbatim.

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

• Third-party content exposure detected (high risk: 0.85). The skill’s runtime workflow calls the Datadog Logs API and then parses/summarizes the returned log event fields (including message/stderr text) into the agent context; those log contents are outsider-authored (produced by external systems/users in your environment, not authored by the operating user), so the LLM ingests them via the API response.