integration-e2e-testing
Pass
Audited by Gen Agent Trust Hub on Nov 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill requires the agent to ingest external acceptance criteria and behavior descriptions to generate test code skeletons. This process creates a surface for indirect prompt injection where adversarial instructions in the input could influence code generation.\n
- Ingestion points: Governing artifacts, task descriptions, and acceptance criteria files.\n
- Boundary markers: Not defined in the skeleton templates or ingestion instructions.\n
- Capability inventory: Writing TypeScript test files to the file system and performing API POST/DELETE requests within test fixtures.\n
- Sanitization: No requirement or guidance for sanitizing or escaping the external input before it is used to populate test skeletons or logic.
Audit Metadata