find-skills

Fail

Audited by Gen Agent Trust Hub on Sep 17, 2026

Risk Level: HIGHREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPRIVILEGE_ESCALATION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The skill's primary function is to facilitate the installation and execution of external code from GitHub repositories using the npx skills add <owner/repo@skill> command. This allows for the introduction of arbitrary executable content into the agent's environment.
  • [COMMAND_EXECUTION]: The instructions direct the agent to execute shell commands using npx, a tool that downloads and runs Node.js packages. This is used for searching (npx skills find) and modifying the system's installed capabilities (npx skills add, npx skills update).
  • [EXTERNAL_DOWNLOADS]: The skill references and fetches data from https://skills.sh/ and various GitHub repositories. While it mentions trusted sources like vercel-labs as examples, the tool itself allows for downloads from any unverified third-party GitHub repository.
  • [PRIVILEGE_ESCALATION]: The skill explicitly instructs the agent to use the -y flag to skip confirmation prompts and the -g flag to install packages globally (npx skills add <package> -g -y). This removes human-in-the-loop oversight for code installation and ensures the installed code has a broader impact on the host system.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Sep 17, 2026, 07:45 AM
Security Audit — agent-trust-hub — find-skills