The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to process potentially untrusted data from external sources.
Ingestion points: Reads application logs and execution output via kubectl logs and kubectl exec as described in references/details.md.
Boundary markers: No explicit delimiters or instructions are provided to the agent to ignore or isolate embedded commands within the ingested log data.
Capability inventory: The skill utilizes administrative tools including kubectl, git, and terraform for environment management as shown in SKILL.md.
Sanitization: No specific validation, filtering, or sanitization of external data is defined before it enters the agent's context.
[SAFE]: Credential management follows industry best practices by using clearly defined placeholders (e.g., 'xxx', 'your-secret-key') for API keys, tokens, and secrets.
[SAFE]: External references and infrastructure examples target well-known technology providers such as Grafana, PagerDuty, and Slack for legitimate configuration purposes.

alerting-oncall