AGENT LAB: SKILLS

prompt-engineering

Fail

Audited by Snyk on Feb 19, 2026

Risk Level: CRITICAL
Full Analysis

CRITICAL E005: Suspicious download URL detected in skill instructions.

  • Suspicious download URL detected (high risk: 0.90). The URLs belong to an unverified .sh domain that explicitly instructs users to curl | sh (download-and-execute) a remote CLI installer — a high-risk distribution pattern for malware; the cli subdomain likely serves an executable script and the cloud subdomain can host lure files even if the image itself appears benign.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

  • Potentially malicious external URL detected (high risk: 0.90). The Quick Start instructs running "curl -fsSL https://cli.inference.sh | sh", which fetches and executes remote shell code at runtime (https://cli.inference.sh) and the skill's examples depend on the resulting CLI, so this external URL directly controls execution and is a required runtime dependency.
Audit Metadata
Risk Level
CRITICAL
Analyzed
Feb 19, 2026, 03:40 AM