search-ux
Fail
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: HIGHEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill directs the agent to download assets from the
Owl-Listener/designer-skillsGitHub repository, which is not a verified or trusted source. - [REMOTE_CODE_EXECUTION]: The installation instructions utilize
npx skills add, a command that fetches and runs code from the remote repository at runtime without source verification. - [COMMAND_EXECUTION]: The skill provides bash scripts that probe the local file system and execute the installation command with automated flags (e.g.,
-y), which reduces user oversight during the execution of remote scripts.
Recommendations
- AI detected serious security threats
Audit Metadata