The Agent Skills Directory

[SAFE]: The skill provides structured templates and instructional workflows for devops incident management. No malicious patterns, obfuscation, or unauthorized data access were detected.- [COMMAND_EXECUTION]: The skill describes the use of standard deployment and orchestration tools (e.g., kubectl, helm, vercel, fly) for incident mitigation. These commands are consistent with the skill's intended purpose and are executed through the explicitly allowed Bash tool.- [INDIRECT_PROMPT_INJECTION]: The skill defines a surface for processing incident data. Evidence Chain: 1. Ingestion points: Alert data and incident reports; 2. Boundary markers: Markdown templates provide structural separation; 3. Capability inventory: Bash, WebFetch, and Read tools; 4. Sanitization: No explicit code-based sanitization is mentioned, but the workflow is inherently human-in-the-loop and focused on documentation.

production-incident-playbook