The Agent Skills Directory

[SAFE]: The skill provides instructions for incident diagnosis using the Firetiger platform. All external resources, including domains (firetiger.com, firetigerapi.com) and source code repositories, are official vendor assets.
[INDIRECT_PROMPT_INJECTION]: The skill processes data from autonomous agent sessions via the read_agent_messages tool. While this creates a surface for indirect prompt injection (e.g., malicious content in logs influencing the agent's reasoning), this is an inherent aspect of a diagnostic tool. The skill does not provide specific sanitization instructions, but it operates within a scoped platform environment.
[COMMAND_EXECUTION]: The skill provides SQL query templates for the query tool to analyze OpenTelemetry traces and logs. These queries are scoped to diagnostic tables and follow standard telemetry analysis patterns.

firetiger-investigate