AGENT LAB: SKILLS

google-veo

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: 0.80). The domain and docs pages look consistent and likely legitimate, but the quick-start instructs piping a remote installer (https://cli.inference.sh) directly into sh (curl | sh), which is a high-risk pattern for arbitrary/hidden code execution and therefore could be used to distribute malware.

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

  • Potentially malicious external URL detected (high risk: 1.00). The skill's quick-start uses a runtime command that fetches and pipes a remote shell script into the shell—"curl -fsSL https://cli.inference.sh | sh"—which executes remote code and is required for the CLI used by the skill.
Audit Metadata
Risk Level
CRITICAL
Analyzed
Feb 18, 2026, 11:03 PM