Mirror Mode

Observe every AI request without adding latency or blocking anything.

How it works

In normal mode, the SDK evaluates guardrails before your request is sent and returns the decision, so your application can stop a blocked request from reaching the provider.

In mirror mode, requests go directly to the AI provider first. SignalVault receives a copy of the request and response asynchronously and logs everything to your audit trail. No latency is added and requests are never blocked.

When to use it

• Getting started — deploy mirror mode first to see what your AI traffic looks like before enabling blocking rules
• Latency-sensitive paths — keep the audit trail without any overhead on the request path
• Gradual rollout — run mirror mode alongside normal mode on different environments to compare

Node.js

const client = new SignalVaultClient({
  apiKey: 'sk_live_...',
  openaiApiKey: process.env.OPENAI_API_KEY,
  mirrorMode: true,
});

Python

client = SignalVaultClient(
    api_key="sk_live_...",
    openai_api_key=os.environ["OPENAI_API_KEY"],
    base_url="https://api.signalvault.io",
    mirror_mode=True,
)