agent-recovery

Warn

Audited by Snyk on Oct 11, 2026

Risk Level: MEDIUM
Full Analysis

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

  • Potentially malicious external URL detected (high risk: 0.70). The skill instructs an AI agent to connect to an external Model Context Protocol (MCP) endpoint and API hosted on an arbitrary AWS Neon tech URL for remote instructions, behavior-selecting configuration, and runtime tools, representing a weak/unknown operational dependency.

Issues (1)

W012
MEDIUM

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

Audit Metadata
Risk Level
MEDIUM
Analyzed
Oct 11, 2026, 12:23 PM
Issues
1