The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses local Python scripts (humanize.py and detect.py) to analyze and modify text files. These scripts utilize standard library modules for regex processing and statistics without invoking external shell commands or unauthorized binaries.
[DATA_EXFILTRATION]: Analysis of the source code confirms there are no network operations or imports (e.g., requests, urllib) that would allow for data to be sent to external servers. All text processing is performed locally on the user's machine.
[CREDENTIALS_UNSAFE]: No hardcoded API keys, tokens, or sensitive file path references (such as .env or SSH keys) were found within the skill's files or scripts.
[OBFUSCATION]: There is no evidence of obfuscated code, encoded payloads, or hidden commands. The skill explicitly includes logic to identify and remove obfuscation techniques like zero-width characters (U+200B, U+200C, etc.) and Unicode tag characters from input text as a feature to humanize it.
[REMOTE_CODE_EXECUTION]: The skill has no external dependencies and does not download or execute code from remote sources. It relies exclusively on the provided Python scripts and a local JSON lexicon.

ig-human