cloudflare-tunnel

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt instructs embedding a Cloudflare tunnel token directly into CLI/docker commands (e.g., --token <YOUR_TOKEN>), which would require the agent to accept and output the secret verbatim—an insecure pattern.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill's runtime instructions pull and run the Docker image "cloudflare/cloudflared:latest" (fetched from Docker Hub), which downloads and executes remote code and is required for the skill to operate.