The Agent Skills Directory

[SAFE]: The skill implements a localized set of Python scripts for static analysis of React component code. It parses JSX and CSS using regular expressions to identify structure and semantics.
[COMMAND_EXECUTION]: While the skill uses CLI commands, these are restricted to executing its own included Python scripts (e.g., analyzer.py, component_generator.py) for processing project files. No shell injection or arbitrary command execution on user input was detected.
[EXTERNAL_DOWNLOADS]: No external downloads or remote script execution (e.g., curl | bash) are performed. The skill relies on locally provided code and optionally installed standard libraries like cssutils.
[DATA_EXFILTRATION]: No network operations or exfiltration patterns were identified. The scripts read local source code and write generated components to the local file system without contacting external domains.
[PROMPT_INJECTION]: The instructions in SKILL.md and reference documents are strictly task-oriented and do not contain patterns intended to bypass AI safety filters or override system prompts.

code-splitter