machine-learning-ops-ml-pipeline
Pass
Audited by Gen Agent Trust Hub on Apr 14, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill implements a multi-agent workflow where user-supplied content from the
$ARGUMENTSvariable is directly embedded into the prompts of several sub-agents. - Ingestion points: The
$ARGUMENTSplaceholder is used in thepromptfield of<Task>blocks fordata-engineer,data-scientist, andml-engineersub-agents inSKILL.md. - Boundary markers: There are no delimiters (such as XML tags or Markdown code blocks) or instructions telling the sub-agents to treat the
$ARGUMENTScontent as untrusted data or to ignore embedded instructions. - Capability inventory: The sub-agents are tasked with generating high-impact outputs, including production Python code, Kubernetes manifests, CI/CD pipeline configurations (GitHub Actions/GitLab CI), and Infrastructure as Code (Terraform).
- Sanitization: The skill does not perform any validation, filtering, or escaping of the user input before passing it to the sub-agents, creating a surface for indirect prompt injection where an attacker could provide input that misdirects the sub-agents to generate malicious infrastructure or backdoored code.
Audit Metadata