The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes a command-line interface (npx skills) to search for, verify, and install packages. It provides instructions for using the -y flag to bypass confirmation prompts, enabling the agent to perform installations autonomously.
[INDIRECT_PROMPT_INJECTION]: The skill processes external data in the form of search results from the skills.sh ecosystem. While this presents a surface for malicious metadata to influence the agent, the skill mitigates this risk by requiring the agent to perform quality and reputation checks (e.g., verifying install counts and official source names like Vercel Labs or Anthropic) before taking action.

find-skills