github-actions-templates

Pass

Audited by Gen Agent Trust Hub on Sep 6, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill acts as a template generator for CI/CD workflows, which creates an attack surface for indirect prompt injection if the agent incorporates untrusted user requirements into the generated YAML files without proper sanitization.
  • Ingestion points: Instructions in SKILL.md direct the agent to "Clarify goals, constraints, and required inputs" from the user.
  • Boundary markers: There are no explicit instructions for the agent to ignore or delimit potentially malicious instructions within the user-supplied data.
  • Capability inventory: The skill is designed to generate configurations that execute shell commands (npm, pip, kubectl, docker) and write to repository files.
  • Sanitization: The instructions do not specify any validation, escaping, or filtering for user-provided inputs used to populate the templates.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 6, 2026, 07:36 AM
Security Audit — agent-trust-hub — github-actions-templates