decision-table-test-design
Pass
Audited by Gen Agent Trust Hub on Nov 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process user-provided business-rule specifications (e.g., in Step 1) to derive manual test cases. While the skill provides a structured method for extraction, it does not explicitly instruct the agent to use boundary markers or to ignore potentially malicious instructions embedded within the source specifications.
- Ingestion points: External business-rule specifications provided as input for test case derivation.
- Boundary markers: The instructions do not define delimiters or specific warnings to ignore instructions within the input data.
- Capability inventory: The skill is primarily used for text generation (deriving test steps and expected results); it does not include commands for network access, sensitive file modification, or arbitrary code execution.
- Sanitization: No input validation or escaping mechanisms are specified for the agent to apply to the input data.
Audit Metadata