The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the Bash tool but limits its scope to 'ls' and 'find' commands via the 'allowed-tools' configuration, ensuring safe directory traversal and file identification.
[DATA_EXPOSURE]: The workflow involves reading source code and configuration files to trace data paths and understand logic. This access is localized to the agent's environment and is necessary for the skill's stated purpose of code exploration.
[SAFE]: No evidence of prompt injection, obfuscation, remote code execution, or persistence mechanisms was found. The skill adheres to its described functionality as a diagnostic and research tool.

code-explore