Unmaintained: this verifiable inference plugin needs to be ported

@elizaos/adapter-opacity

This adapter integrates Opacity proofs into ElizaOS, enabling verifiable inference results from various AI model providers available through the CloudFlare AI Gateway. It implements the IVerifiableInferenceAdapter interface, making it compatible with other verifiable inference solutions.

Installation

pnpm add @elizaos/adapter-opacity

Configuration

Add the following environment variables to your .env file:

OPACITY_TEAM_ID=f309ac8ae8a9a14a7e62cd1a521b1c5f
OPACITY_CLOUDFLARE_NAME=eigen-test
OPACITY_PROVER_URL=https://opacity-ai-zktls-demo.vercel.app
# Verifiable Inference Configuration
VERIFIABLE_INFERENCE_ENABLED=true # Set to true to enable verifiable inference
VERIFIABLE_INFERENCE_PROVIDER=opacity # Options: opacity

(make sure to VERIFIABLE_INFERENCE_ENABLED to true!)

Usage

import { OpacityAdapter } from "@elizaos/adapter-opacity";
import { VerifiableInferenceOptions } from "@elizaos/core";

// Initialize the adapter
const opacityAdapter = new OpacityAdapter(runtime, {
    teamId: process.env.OPACITY_TEAM_ID,
    teamName: process.env.OPACITY_CLOUDFLARE_NAME,
    baseUrl: process.env.OPACITY_PROVER_URL
});

// Generate text with verifiable results
const options: VerifiableInferenceOptions = {
    // Optional: Override the default endpoint
    endpoint: "https://custom-api.example.com",
    // Optional: Add custom headers
    headers: {
        "X-Custom-Header": "value",
    },
    // Optional: Provider-specific options
    providerOptions: {
        temperature: 0.7,
    },
};

const result = await opacityAdapter.generateText(
    "What is Rust?",
    "gpt-4",
    options
);

console.log("Response:", result.text);
console.log("Proof:", result.proof);

// Verify the proof
const isValid = await opacityAdapter.verifyProof(result);
console.log("Proof is valid:", isValid);

Features

• Implements IVerifiableInferenceAdapter interface for standardized verifiable inference
• Support for multiple AI model provider, in the future may be expanded to accommodate for all gateways supported by CloudFlare
• Customizable options for each request
• Built-in proof verification

Response Format

The adapter returns a VerifiableInferenceResult object containing:

{
    text: string;           // The generated text response
    proof: unknown;         // The proof data
    provider: string;       // The provider name (e.g., "opacity")
    timestamp: number;      // Generation timestamp
    metadata?: {           // Optional metadata
        modelProvider: string;
        modelClass: string;
        endpoint: string;
    }
}

How it Works

The Opacity adapter wraps AI model API calls to CloudFlare, then performs MPC-TLS on the logged responses.

This allows you to:

1. Make verifiable API calls to AI model providers
2. Generate proofs of the responses
3. Verify the authenticity of the responses
4. Ensure the responses haven't been tampered with

Step By Step

sequenceDiagram
    autonumber
    participant Eliza
    participant Cloudflare
    participant OpenAI
    Eliza ->> Cloudflare : Prompt Req
    Cloudflare ->> OpenAI : Prompt Req
    OpenAI ->> Cloudflare : Prompt Response
    Cloudflare ->> Cloudflare : Log Prompt Response
    Cloudflare ->> Eliza : Prompt Response
    create participant Opacity
    Eliza ->> Opacity : Generate Proof for Prompt Response
    Opacity ->> Cloudflare : Fetch Prompt Response Log in MPC-TLS
    Cloudflare ->> Opacity : Respond with Prompt Response Log
    Opacity ->> Eliza : Return Proof of Prompt Response Log

Loading

License

MIT