weekly-geo-report

Warn

Audited by Snyk on Jun 16, 2026

Risk Level: MEDIUM
Full Analysis

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

  • Potentially malicious external URL detected (high risk: 0.90). The skill explicitly requires configuring a Promptwatch MCP server endpoint ("https://server.promptwatch.com/mcp") in the agent runtime config, which is a remote Model Control Plane URL used at runtime to deliver control instructions/prompts to the agent, so it can directly control agent prompts.

Issues (1)

W012
MEDIUM

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

Audit Metadata
Risk Level
MEDIUM
Analyzed
Jun 16, 2026, 10:29 PM
Issues
1