edge-computing-patterns

CRITICAL E004: Prompt injection detected in skill instructions.

• Potential prompt injection detected (medium risk: 0.30). The skill is a legitimate edge-computing guide with no obfuscation, pretext, or impersonation; its templates contain careless security flaws, an open proxy that forwards the user's Authorization header to arbitrary URLs and an auth helper that never verifies the JWT signature, which weaken protections but show no malicious intent.