character-design-sheet

Warn

Audited by Gen Agent Trust Hub on Aug 8, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill requests broad permission for the 'belt' CLI via allowed-tools: Bash(belt *) in the SKILL.md frontmatter. This gives the agent unrestricted access to all subcommands of the tool.
  • [EXTERNAL_DOWNLOADS]: The skill instructs the user to install external components using npx skills add belt-sh/cli and npx skills add inference-sh/skills@... patterns in SKILL.md. This involves fetching and installing code from remote registries.
  • [COMMAND_EXECUTION]: Multiple examples throughout SKILL.md demonstrate executing shell commands (e.g., belt login, belt app run) that transmit user prompts and potentially local file paths to remote inference servers (inference.sh).
  • [EXTERNAL_DOWNLOADS]: The skill references and links to external installation scripts hosted on GitHub (e.g., https://raw.githubusercontent.com/inference-sh/skills/refs/heads/main/cli-install.md). While from a known service, it encourages the execution of external content not bundled with the skill.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Aug 8, 2026, 11:51 PM
Security Audit — agent-trust-hub — character-design-sheet