prompt-engineer-toolkit

Pass

Audited by Gen Agent Trust Hub on Jul 18, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The scripts/prompt_tester.py utility supports an optional --runner-cmd argument to execute external CLI tools for prompt evaluation. The implementation uses shlex.split and passes arguments as a list to subprocess.run, effectively preventing shell injection attacks while providing the intended functionality.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes prompt definitions and test cases that may be sourced from external files or user input.
  • Ingestion points: Prompt content is read from files (--prompt-a-file, --prompt-b-file) or provided via CLI arguments, and test cases are loaded from JSON files (--cases-file) in scripts/prompt_tester.py.
  • Boundary markers: The tool does not implement specific boundary markers between prompt templates and test case data within the testing script.
  • Capability inventory: The skill possesses command execution capabilities through subprocess.run to call external evaluators.
  • Sanitization: Input data is handled as strings and passed as arguments to sub-processes; shlex.split ensures that input data cannot break out of the command structure into the system shell.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 18, 2026, 04:19 PM
Security Audit — agent-trust-hub — prompt-engineer-toolkit