agent-tools

Pass

Audited by Gen Agent Trust Hub on Aug 3, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs users to install a CLI tool using a remote script from https://cli.inference.sh. This is documented as the standard installation method for the service.
  • Evidence: curl -fsSL https://cli.inference.sh | sh found in SKILL.md, references/authentication.md, and references/cli-reference.md.
  • Analysis: Per the trust scope rules, since the download is from the official domain of the service the skill is designed to integrate with (inference.sh), this is considered a standard functional component rather than a security risk.
  • [REMOTE_CODE_EXECUTION]: The skill provides instructions to execute remote code via shell piping to install the belt CLI.
  • Evidence: curl -fsSL https://cli.inference.sh | sh in SKILL.md.
  • Analysis: This is the intended behavior for setting up the environment. The skill also provides a manual installation method with checksum verification (sha256sum -c checksums.txt) as a safer alternative for security-conscious users.
  • [COMMAND_EXECUTION]: The skill uses Bash(belt *) to interact with the inference.sh API through the installed CLI.
  • Analysis: This capability is restricted to the specific CLI tool required for the skill's stated purpose of running AI applications.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 3, 2026, 03:53 PM
Security Audit — agent-trust-hub — agent-tools