feature-store-connector

Fail

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: HIGHPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [Indirect Prompt Injection] (HIGH): The skill is designed to process external ML deployment patterns and configurations while holding high-privilege capabilities.
  • Ingestion points: External files and user requests regarding feature store patterns and ML configurations.
  • Boundary markers: Absent; the skill does not define delimiters to protect against embedded instructions in processed data.
  • Capability inventory: The skill requests Bash, Write, and Edit tools, which enable arbitrary command execution and filesystem modification.
  • Sanitization: No sanitization or validation mechanisms are described to filter malicious input.
  • [Command Execution] (MEDIUM): The skill requests access to the Bash tool. While common in MLOps, this provides a powerful vector for exploitation if the agent is misled by malicious data from an indirect prompt injection attack.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Feb 16, 2026, 01:02 PM