test-engine

Warn

Audited by Gen Agent Trust Hub on Oct 7, 2026

Risk Level: MEDIUMDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [DYNAMIC_EXECUTION]: The script tests/run-validation.py performs dynamic loading of a local module using importlib.util.spec_from_file_location with a file path computed at runtime (SKILL_DIR / 'scripts' / 'validate-output.py'). While intended for local validation, this pattern involves dynamic code loading from the filesystem.
  • [INDIRECT_PROMPT_INJECTION]: The skill's primary function is to analyze existing code to write tests, which involves ingesting untrusted data through the read_file tool.
  • Ingestion points: File content accessed via the read_file tool.
  • Boundary markers: No specific delimiters or instructions to ignore embedded instructions are defined in the output templates.
  • Capability inventory: read_file tool access and local execution of validation scripts.
  • Sanitization: No sanitization or filtering of the ingested file content is present.
  • [COMMAND_EXECUTION]: The skill distributes two Python scripts (tests/run-validation.py and scripts/validate-output.py) designed to be executed to validate that the generated test plans meet the defined output contract.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Oct 7, 2026, 11:25 AM