generate-cds-unit-test

Pass

Audited by Gen Agent Trust Hub on Jun 22, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes untrusted CDS DDL source code from an external SAP system.\n
  • Ingestion points: The skill fetches source code and dependency metadata using SAPRead and SAPContext tool calls in Step 1.\n
  • Boundary markers: The instructions do not specify the use of delimiters or explicit 'ignore embedded instructions' warnings when providing the external DDL content to the model.\n
  • Capability inventory: The skill possesses powerful system-modification tools including SAPWrite (to create/update classes), SAPActivate (to activate code), and SAPDiagnose (to execute tests) on the target SAP system.\n
  • Sanitization: There is no evidence of validation or sanitization of the fetched source code before it is used to influence the model's code generation logic.\n
  • Mitigation: The skill incorporates a mandatory 'Preview and Confirm' checkpoint in Step 5, requiring explicit human review and approval before any write operations are executed.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 22, 2026, 11:09 AM
Security Audit — agent-trust-hub — generate-cds-unit-test