The Agent Skills Directory

[SAFE]: The skill is a structured design guideline consisting of markdown and JSON data. Comprehensive analysis of the 11 threat categories revealed no indicators of prompt injection, obfuscation, or persistence mechanisms.- [COMMAND_EXECUTION]: The skill includes Node.js scripts (check-spec.mjs, detect.mjs, build-reference.mjs) used to validate chart specifications and source code. These scripts use built-in Node.js modules for file system access and regex matching. They do not execute arbitrary shell commands or external payloads.- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied JSON specifications and React source code files to provide design feedback. While this constitutes an ingestion surface for untrusted data, the agent's capabilities are limited to telemetry logging and reporting, and the skill includes instructions for the agent to follow strict design rules, mitigating the risk of following instructions embedded in the data.- [EXTERNAL_DOWNLOADS]: Documentation references are made to mobbin.com, a well-known design reference service. These URLs are used solely for pattern documentation and do not trigger automatic downloads of executable code.

vois-dataviz