The Agent Skills Directory

[COMMAND_EXECUTION]: The Bash implementations of wait_for and retry_with_backoff in references/implementation-patterns.md use the eval command to execute condition and command strings. This pattern is susceptible to command injection if the input strings are derived from untrusted sources.
[COMMAND_EXECUTION]: The Python function check_command in references/implementation-patterns.md utilizes subprocess.run(command, shell=True). Using shell=True is a security risk as it allows command injection if the command string is constructed using unvalidated external input.

condition-based-waiting