data-model-creation

Pass

Audited by Gen Agent Trust Hub on Oct 24, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest business requirements from user prompts and translate them into structured Mermaid class diagrams and data models.
  • Ingestion points: Untrusted data enters the agent context via user requests for entity modeling and business logic descriptions as described in SKILL.md.
  • Boundary markers: The instructions do not specify any explicit boundary markers or delimiters (like XML tags or triple quotes) to separate user input from instructions during the modeling phase.
  • Capability inventory: The skill uses tools like modifyDataModel and manageDataModel which possess the capability to create and manage database schemas within the CloudBase environment.
  • Sanitization: There are no instructions for sanitizing, escaping, or validating the user-provided entity names or field descriptions before they are used to generate the model structure.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 24, 2026, 09:54 AM