skills/commandcodeai/agent-skills/building-ai-agent-on-cloudflare/Snyk

building-ai-agent-on-cloudflare

Warn

Audited by Snyk on Oct 3, 2026

Risk Level: MEDIUM

Full Analysis

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

Third-party content exposure detected (medium risk: 0.30). The required workflow involves building an AI agent that ingests user chat messages via WebSockets (onMessage), which can be influenced by outsider-authored inputs when deployed.

Issues (1)

W011

MEDIUM

Third-party content exposure detected (indirect prompt injection risk).

Audit Metadata

Risk Level

MEDIUM

Analyzed

Oct 3, 2026, 07:34 PM

Issues

1