machine-learning-ops-ml-pipeline
Pass
Audited by Gen Agent Trust Hub on Aug 3, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill facilitates an orchestration workflow where user input from the
$ARGUMENTSvariable is directly inserted into the prompts for specialized sub-agents (Data Engineer, Data Scientist). This creates a surface for indirect prompt injection if the user input contains instructions designed to manipulate the sub-agents' behavior. - Ingestion points: User input is ingested via the
$ARGUMENTSplaceholder inSKILL.mdand passed to multiple<Task>prompts. - Boundary markers: The skill lacks delimiters (such as triple quotes or XML tags) or explicit instructions for the agents to ignore embedded commands within the
$ARGUMENTSstring. - Capability inventory: The orchestrated agents are tasked with generating executable Python code, Hyperparameter optimization scripts (Optuna), Kubernetes manifests (Helm charts), and Infrastructure as Code (Terraform), which could be compromised by malicious input.
- Sanitization: There is no evidence of input validation, filtering, or escaping of the user-provided content before it is processed by the sub-agents.
Audit Metadata