p-image

Pass

Audited by Gen Agent Trust Hub on Sep 7, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references external installation guides and resources hosted on inference.sh and its associated GitHub repository. These downloads are limited to documentation and the service's official CLI tool.
  • [COMMAND_EXECUTION]: The skill utilizes the Bash tool to execute belt CLI commands. The skill's configuration uses allowed-tools to restrict execution to only the belt command, which is a positive security measure following the principle of least privilege.
  • [INDIRECT_PROMPT_INJECTION]: The skill accepts user-provided prompts that are passed to external image generation models. The risk is mitigated by the fact that the output is an image and the input is passed within a structured JSON object to the CLI tool.
  • [CREDENTIALS_SAFE]: While the skill mentions belt login, it does not contain hardcoded credentials. It follows standard practices by instructing the user to authenticate through the official CLI.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 7, 2026, 10:34 AM
Security Audit — agent-trust-hub — p-image