The Agent Skills Directory

[COMMAND_EXECUTION]: The utility script scripts/agent.py uses subprocess.run to execute the promptfoo command-line tool. This is a functional requirement to automate the red-teaming process. The implementation uses argument lists rather than shell strings, which is the recommended practice to prevent command injection.
[EXTERNAL_DOWNLOADS]: The skill guides users to install well-known and reputable security frameworks: promptfoo via npm and deepteam via pip. These tools are widely recognized in the AI security community for model evaluation and adversarial testing.
[CREDENTIALS_UNSAFE]: The documentation and example CI/CD workflows correctly instruct the use of environment variables and secrets for sensitive API keys, following standard security protocols for credential management.
[DATA_EXFILTRATION]: No unauthorized network operations or sensitive file access patterns were detected. The skill's primary focus is on generating and evaluating local test cases for security analysis.

continuous-llm-red-teaming-with-promptfoo