AGENT LAB: SKILLS

book-cover-design

Fail

Audited by Snyk on Feb 18, 2026

Risk Level: CRITICAL
Full Analysis

CRITICAL E005: Suspicious download URL detected in skill instructions.

  • Suspicious download URL detected (high risk: 0.90). Both domains point to an unfamiliar service that publishes an install script (curl | sh) / CLI on a non‑well‑known domain — executing remote .sh from an untrusted source is a common malware distribution vector and therefore suspicious.

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

  • Potentially malicious external URL detected (high risk: 0.90). The skill explicitly installs and runs a remote script at runtime with "curl -fsSL https://cli.inference.sh | sh", so https://cli.inference.sh is fetched and executed as a required CLI dependency allowing remote code execution and potential control over agent behavior.
Audit Metadata
Risk Level
CRITICAL
Analyzed
Feb 18, 2026, 11:00 PM