background-removal

Pass

Audited by Gen Agent Trust Hub on Mar 19, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill requests permission to run infsh commands via the Bash tool. This is the primary method for performing background removal and image editing tasks.
  • [EXTERNAL_DOWNLOADS]: The instructions suggest using npx to fetch and install additional skills from the inference-sh organization, which is the provider of the underlying infrastructure.
  • [DATA_EXFILTRATION]: User-provided images and prompts are sent to external cloud platforms (inference.sh and fal.ai) to perform inference. This behavior is necessary for the skill's intended functionality.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to the ingestion of untrusted image URLs and natural language prompts in SKILL.md examples. It lacks boundary markers or sanitization while having the capability to execute commands via Bash as seen in SKILL.md.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 19, 2026, 01:18 PM
Security Audit — agent-trust-hub — background-removal