cicd-automation-workflow-automate
Pass
Audited by Gen Agent Trust Hub on Nov 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill includes logic designed to analyze project files, such as READMEs and CI configurations, to identify automation opportunities. Because these files are external to the skill and potentially attacker-controlled, they represent a surface for indirect prompt injection.
- Ingestion points: The
WorkflowAnalyzerclass inresources/implementation-playbook.mdreads content fromREADME.md,README.rst, andREADME.txtto search for specific process keywords. - Boundary markers: The implementation lacks boundary markers or explicit instructions to the AI to ignore natural language instructions that might be embedded within the project files being analyzed.
- Capability inventory: The skill allows the agent to generate high-privilege configuration files including shell scripts (
setup-dev-environment.sh), GitHub Actions workflows (which handle secrets), Terraform files, and Kubernetes manifests. - Sanitization: There is no evidence of sanitization or validation performed on the raw text extracted from project documentation before it is processed by the agent to generate recommendations.
Audit Metadata