The Agent Skills Directory

[COMMAND_EXECUTION]: The scripts/agent.py script invokes the certipy binary via subprocess.run. The implementation uses a list of arguments and disables shell execution, which is a secure method for executing external commands.
[CREDENTIALS_UNSAFE]: The skill uses plaintext passwords and NT hashes as command-line arguments in both documentation and the automation script. While standard for command-line security tools, this can lead to credential exposure in process logs or shell history.
[EXTERNAL_DOWNLOADS]: The skill directs users to download certipy-ad from standard package managers. The tool is a well-regarded open-source security project for Active Directory assessments.

exploiting-adcs-with-certipy