gpt-image

Pass

Audited by Gen Agent Trust Hub on Sep 21, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs the user to install the belt-sh/cli package and other related skills using npx. It also references documentation and installation scripts hosted on GitHub under the inference-sh organization.
  • [COMMAND_EXECUTION]: The skill utilizes the belt CLI tool through the Bash tool to perform image processing operations, such as generating images from prompts and applying edits to reference images.
  • [INDIRECT_PROMPT_INJECTION]:
  • Ingestion points: The skill accepts untrusted data through the prompt, images, and mask parameters defined in the SKILL.md examples and parameter table.
  • Boundary markers: There are no identified boundary markers or instructions to the agent to treat input data as data only and ignore any potential embedded instructions.
  • Capability inventory: The skill possesses the capability to execute shell commands using the belt tool via the Bash tool.
  • Sanitization: There is no evidence of sanitization, filtering, or validation performed on the user-provided prompts or the contents of external URLs before they are passed to the command-line interface.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 21, 2026, 08:14 PM
Security Audit — agent-trust-hub — gpt-image