clinical-trial-schema-designer

Pass

Audited by Gen Agent Trust Hub on Oct 22, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted clinical trial protocols to generate data schemas, creating a surface for potential instruction injection via data.\n
  • Ingestion points: The protocol_text parameter in skill.yaml.\n
  • Boundary markers: No delimiters are specified to separate input data from instructions.\n
  • Capability inventory: The skill is intended to perform study variable extraction and generate SQL/Pydantic schemas.\n
  • Sanitization: No sanitization or validation of the input text is mentioned in the skill definition.\n- [NO_CODE]: The skill consists entirely of configuration and documentation files; no executable scripts or binaries are present.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 22, 2026, 03:51 PM