ejentum-reasoning-harness

Warn

Audited by Snyk on Jun 23, 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 requires installing and running the ejentum MCP (via "npx -y ejentum-mcp") and at runtime the MCP server makes outbound HTTPS requests to the Ejentum Logic API (ejentum.com/…, e.g. https://ejentum.com/pricing is referenced), which returns scaffolds that directly control the agent's internal reasoning/instruction prompts.

Issues (1)

W012
MEDIUM

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

Audit Metadata
Risk Level
MEDIUM
Analyzed
Jun 23, 2026, 08:50 PM
Issues
1
Security Audit — snyk — ejentum-reasoning-harness