background-removal

Pass

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs the user to install the 'belt' CLI tool using npx skills add belt-sh/cli and references installation guidelines hosted on a GitHub repository (inference-sh/skills). These are standard installation procedures for the mentioned platform.- [COMMAND_EXECUTION]: The skill uses the belt command-line utility to execute remote AI models (e.g., belt app run infsh/birefnet). These commands are scoped to the intended functionality of background removal and image editing.- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied data such as image_url and prompt strings within JSON payloads passed to the CLI tool.
  • Ingestion points: External image URLs and text prompts provided by the user in SKILL.md examples.
  • Boundary markers: Commands use JSON structures to delimit input data.
  • Capability inventory: The skill uses belt app run to process data through external AI models.
  • Sanitization: Standard for image processing skills; the agent relies on the underlying platform's handling of the JSON input.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 2, 2026, 02:18 PM
Security Audit — agent-trust-hub — background-removal