The Agent Skills Directory

[COMMAND_EXECUTION] (SAFE): The provided script scripts/validate.py is limited to local file system operations, such as checking for the existence of required directories and parsing YAML configuration. It does not use subprocess, os.system, or any other mechanism for arbitrary command execution.- [DATA_EXFILTRATION] (SAFE): Analysis of all files, including SKILL.md and scripts/validate.py, confirms the absence of network-capable code (e.g., requests, urllib, socket) or access to sensitive local files (e.g., SSH keys, environment secrets).- [Indirect Prompt Injection] (INFO): The skill is designed to provide information on 'topics' requested by the user. While it ingests untrusted user input, it possesses no side-effect capabilities like file modification or network communication, meaning any embedded instructions in the processed data cannot be executed or exfiltrated.- [EXTERNAL_DOWNLOADS] (SAFE): The skill does not perform any remote package installations or script downloads. The dependencies identified are standard libraries or expected for the skill's documented purpose.

patterns