test-executor

Pass

Audited by Gen Agent Trust Hub on Jun 19, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface by design. It retrieves and executes arbitrary instructions from external files (e.g., questions.md).
  • Ingestion points: The skill reads external question files or directory contents to extract test cases (Workflow Steps 1 and 2).
  • Boundary markers: There are no instructions to wrap external content in delimiters or warnings to ignore embedded instructions; it is told to execute the prompt exactly as a user task.
  • Capability inventory: The skill utilizes bash, await_command, notebook tools, file system tools, and sub-agents (Workflow Step 3).
  • Sanitization: No sanitization or validation of the ingested test content is performed before execution.
  • [COMMAND_EXECUTION]: The skill workflow explicitly directs the agent to perform command execution via bash and await_command, and to manage files and notebooks based on the contents of the test files provided at runtime.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 19, 2026, 08:56 PM
Security Audit — agent-trust-hub — test-executor