book-cover-design
Warn
Audited by Gen Agent Trust Hub on Aug 12, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references installation instructions hosted on a raw GitHub URL (
https://raw.githubusercontent.com/inference-sh/skills/refs/heads/main/cli-install.md). Using raw content from unverified organizations is a security risk as the source can be modified without the oversight of official registries. - [REMOTE_CODE_EXECUTION]: The skill instructs the user to add external skills and install CLI tools (
belt-sh/cli,inference-sh/skills) from unverified third-party sources. This facilitates the installation and execution of code that has not been vetted by the platform or a trusted vendor. - [COMMAND_EXECUTION]: The skill heavily utilizes the
beltCLI tool to perform authentication (belt login) and execute remote AI models (belt app run). These operations depend on an external binary whose security profile and network behavior are outside the controlled environment.
Audit Metadata