Skills
skills/1nfsh/skills/book-cover-design/Gen Agent Trust Hub

book-cover-design

Fail

Audited by Gen Agent Trust Hub on Feb 18, 2026

Risk Level: CRITICALREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [REMOTE_CODE_EXECUTION] (CRITICAL): Found in SKILL.md. The command curl -fsSL https://cli.inference.sh | sh performs a download and execution of a remote script from an untrusted domain. This allows for arbitrary code execution on the host system.
  • [EXTERNAL_DOWNLOADS] (MEDIUM): Found in SKILL.md. The skill uses npx to install additional components from inference-sh/skills, which is not a verified or trusted source.
  • [COMMAND_EXECUTION] (MEDIUM): Found in SKILL.md. The skill requests permission for Bash(infsh *), which allows for execution of system commands via a CLI tool that was installed through an insecure remote script.
  • [PROMPT_INJECTION] (LOW): Found in SKILL.md. The skill is vulnerable to indirect prompt injection through its handling of user-provided prompt strings.
  • Ingestion points: User-supplied text strings passed to the infsh command for image generation.
  • Boundary markers: Partial mitigation through the use of JSON-formatted arguments in CLI calls.
  • Capability inventory: Uses the Bash tool to execute system-level image generation utilities.
  • Sanitization: No evidence of input sanitization or validation for user-provided prompt content.
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 18, 2026, 01:53 AM
Security Audit — agent-trust-hub — book-cover-design