NYC
skills/vm0-ai/vm0-skills/imgur/Gen Agent Trust Hub

imgur

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFEDATA_EXFILTRATIONCOMMAND_EXECUTION
Full Analysis
  • [DATA_EXFILTRATION] (LOW): The skill performs network requests to api.imgur.com (a non-whitelisted domain) to upload image data. This is the primary function of the skill, but users must be aware that any uploaded content becomes publicly accessible via the returned URL.
  • [COMMAND_EXECUTION] (SAFE): The skill utilizes standard curl commands for API interaction. It correctly handles credentials using environment variables via the vm0_secrets configuration rather than hardcoding sensitive information. It includes a specific implementation note to use bash -c for preserving environment variables across pipes in certain agent environments, which is an operational instruction rather than a security vulnerability.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:05 PM