deployment-documentation

Pass

Audited by Gen Agent Trust Hub on Mar 4, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill includes configuration templates that reference resources from well-known services.
  • References official GitHub Actions from trusted organizations including GitHub, AWS, Azure, and Slack.
  • Includes integration templates for the Datadog API.
  • [COMMAND_EXECUTION]: The skill contains extensive templates for shell commands and automation scripts intended for infrastructure management.
  • Provides templates for package installation via brew, container builds via docker, and cluster management via kubectl.
  • Includes infrastructure-as-code (IaC) templates for Terraform.
  • [PROMPT_INJECTION]: The skill provides templates for deployment and infrastructure scripts that incorporate user-provided data, creating an indirect injection surface.
  • Ingestion points: Placeholders in SKILL.md such as [Application Name], your-registry, and various environment variable fields.
  • Boundary markers: No explicit boundary markers or instructions to ignore embedded commands are included in the markdown templates.
  • Capability inventory: The templates generate commands for kubectl, docker, terraform, and curl which have high privileges over production infrastructure.
  • Sanitization: No sanitization or validation logic is provided for the data interpolated into these templates.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 4, 2026, 05:21 PM