The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill initiates the download and installation of 22 separate creative coding libraries (e.g., Three.js, GSAP, Babylon.js) from the GitHub repository freshtechbro/claudedesignskills. These are third-party resources not maintained by a recognized trusted organization.\n- [REMOTE_CODE_EXECUTION]: The skill instructs the agent to use npx skills add freshtechbro/claudedesignskills to fetch and execute external skill packages. This facilitates the execution of remote code from a non-trusted repository during the installation process.\n- [COMMAND_EXECUTION]: The instructions include shell scripts to check for the presence of specific files and markers (e.g., grep -q "^das:") in the local file system (~/.agents/skills/) to determine installation state before proceeding.\n- [INDIRECT_PROMPT_INJECTION]: By installing a high volume of external instructions (22 skills), the package significantly expands the agent's context and capabilities with unverified third-party content.\n
Ingestion points: The skill metadata and installation logic within SKILL.md and stub.yaml.\n
Boundary markers: None provided to isolate or sanitize the many external skills being installed.\n
Capability inventory: File system status checks ([ -f ]), text processing (grep), and package management commands (npx, /plugin).\n
Sanitization: No validation of the remote skill content is performed prior to integration into the agent's environment.

claudedesignskills