The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze external data such as logs, metrics, and traces. While this creates an attack surface for indirect prompt injection if the processed data contains malicious instructions, the skill includes defensive instructions for the agent to audit input credibility, separate facts from assumptions, and maintain evidence-based conclusions. The severity is low as the skill's primary function is analytical and it does not possess tools to execute commands or perform network operations on its own.
[SAFE]: The skill is entirely composed of Markdown and YAML configuration files providing instructional guidance. No code, scripts, or binaries are included. The author metadata and source URLs are consistent with the provided vendor context.

root-cause-analysis