The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The script scripts/pydeps.py contains a reference to a shell command for installing a package manager (curl -LsSf https://astral.sh/uv/install.sh | sh). Technical analysis confirms this command is stored as a string constant meant to be printed as part of a help message to the user, and is not executed by the skill itself.\n- [COMMAND_EXECUTION]: The skill utilizes subprocess.run and os.execv within its scripts (e.g., scripts/pydeps.py, scripts/pipeline.py) to orchestrate its various internal components. These commands are used to invoke the skill's own Python scripts and are not used to execute untrusted system commands.\n- [DYNAMIC_EXECUTION]: The skill employs importlib.import_module across several scripts (such as scripts/brand.py and scripts/logolib.py) to dynamically load its own internal modules. This is a standard architectural pattern used for modularity within the skill's codebase.\n- [INDIRECT_PROMPT_INJECTION]: The functionality provided by site_palette.py involves fetching and analyzing content from user-provided URLs to extract color palettes and font information. While this represents a data ingestion surface, the content is parsed by specialized scripts for technical metadata rather than being directly processed as instructions by the agent.

brand-identity