skills/drolu/agent-skills/find-skills/Gen Agent Trust Hub

find-skills

Fail

Audited by Gen Agent Trust Hub on Jul 2, 2026

Risk Level: HIGHREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The skill provides instructions and commands (npx skillhu install, npx skills add) to download and install executable code from third-party marketplaces (skillhu.bz and skills.sh). These sources are unvetted, and the installed skills can execute arbitrary commands on the host system once loaded.
  • [COMMAND_EXECUTION]: The skill explicitly uses the -y flag in the npx skills add <owner/repo@skill> -g -y command. This flag instructs the agent to skip confirmation prompts during the installation process, bypassing a critical security checkpoint that would normally allow the user to review the installation request.
  • [EXTERNAL_DOWNLOADS]: The skill is designed to interact with external domains to fetch, list, and update software packages. The marketplace skillhu.bz is described as allowing "instant publish/install" with "no auth required," which significantly increases the risk of supply chain attacks or malicious package injection.
  • [REMEDIATION]:
  • Remove the -y flag from installation commands to ensure the user must explicitly approve every package installation.
  • Implement a warning for the user when they are about to install software from unvetted or low-reputation sources.
  • Provide a way for users to inspect the SKILL.md or source code of a discovered skill before proceeding with the installation.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Jul 2, 2026, 07:10 PM
Security Audit — agent-trust-hub — find-skills