Skills
▥AGENT LAB: SKILLS
skills/inference-sh-3/skills/image-upscaling/Gen Agent Trust Hub

image-upscaling

Fail

Audited by Gen Agent Trust Hub on Feb 19, 2026

Risk Level: CRITICALREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • REMOTE_CODE_EXECUTION (CRITICAL): The skill's 'Quick Start' section documentation instructs users to run curl -fsSL https://cli.inference.sh | sh. This pattern is extremely dangerous as it executes remote scripts directly in the shell without any verification or source integrity checks, originating from a domain not listed in the trusted sources. \n- EXTERNAL_DOWNLOADS (MEDIUM): The skill encourages the installation of additional skills from the inference-sh/skills registry using npx skills add, which facilitates the download and execution of unverified third-party code. \n- COMMAND_EXECUTION (LOW): The skill requests access to the Bash tool to execute infsh commands, providing the agent with the ability to interact with the underlying system. \n- PROMPT_INJECTION (LOW): The skill presents an indirect prompt injection surface. 1. Ingestion points: image_url parameters in tool calls. 2. Boundary markers: Absent. 3. Capability inventory: Shell command execution via infsh. 4. Sanitization: Absent. Malicious data returned from processed external images could potentially influence subsequent agent actions.
Recommendations
  • HIGH: Downloads and executes remote code from: https://cli.inference.sh - DO NOT USE without thorough review
  • AI detected serious security threats
Audit Metadata
Risk Level
CRITICAL
Analyzed
Feb 19, 2026, 03:41 AM