test-implementation-behavior
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze external, potentially untrusted data such as source code, issue trackers, and pull requests to generate behavioral tests.\n
- Ingestion points: SKILL.md and references/enterprise-operation.md describe workflows where the agent reads repository content and user-provided requirements.\n
- Boundary markers: The skill contains explicit instructions in references/enterprise-operation.md to treat external inputs as untrusted and to ignore any embedded instructions that attempt to escalate privileges.\n
- Capability inventory: The agent has the capability to write and execute code (via Python and shell scripts) based on its analysis of this data.\n
- Sanitization: The instructions mandate using least-privilege credentials and avoiding the hardcoding of secrets.\n- [DYNAMIC_EXECUTION]: The skill utilizes dynamic loading and external tool invocation for its verification examples.\n
- Evidence: The Python test script assets/rename-contract/test_rename_operation.py uses importlib to dynamically load the implementation file for testing.\n
- Evidence: The shell script assets/hdl-counter/verify.sh executes the iverilog compiler and vvp runtime to perform hardware simulations.\n
- Context: These are standard and expected behaviors for a testing skill and are confined to the local execution of provided asset files.
Audit Metadata