mirror of
https://github.com/hanzoai/agents.git
synced 2026-10-30 18:48:36 +00:00
zeekayandHanzo Dev
ff7d100496
The server mounts /v1, /v1/ui and /v2/ui. The UI and its dev proxy, the helm and kubernetes health probes, functional tests, project templates, examples and scripts still called /api/v1, so the probes failed and the UI's requests missed. Co-authored-by: Hanzo Dev <[email protected]>
87 lines
4.1 KiB
TypeScript
87 lines
4.1 KiB
TypeScript
import 'dotenv/config';
|
|
import { Agent } from '@hanzo-agents/sdk';
|
|
import { reasonersRouter } from './reasoners.js';
|
|
|
|
/**
|
|
* Verifiable Credentials Example
|
|
*
|
|
* This example demonstrates how to use DID (Decentralized Identifiers) and
|
|
* Verifiable Credentials (VCs) in Hanzo Agents to create cryptographically
|
|
* verifiable audit trails for agent executions.
|
|
*
|
|
* Each reasoner in this example generates a VC that:
|
|
* - Records input/output hashes for tamper detection
|
|
* - Links to caller and target DIDs for accountability
|
|
* - Is signed by the control plane's issuer DID
|
|
* - Can be verified independently for compliance/audit
|
|
*
|
|
* Prerequisites:
|
|
* 1. Control plane running with DID enabled (features.did.enabled: true)
|
|
* 2. Keystore directory exists (./data/keys on control plane)
|
|
*
|
|
* Usage:
|
|
* pnpm dev:vc
|
|
*
|
|
* Then test with:
|
|
* curl -X POST http://localhost:8080/v1/execute/vc-demo.vc_process \
|
|
* -H "Content-Type: application/json" \
|
|
* -d '{"input": {"text": "Hello, Verifiable World!"}}'
|
|
*/
|
|
|
|
async function main() {
|
|
const agent = new Agent({
|
|
nodeId: process.env.AGENT_ID ?? 'vc-demo',
|
|
controlPlaneUrl: process.env.HANZO_AGENTS_URL ?? 'http://localhost:8080',
|
|
port: Number(process.env.PORT ?? 8006),
|
|
version: '1.0.0',
|
|
devMode: true,
|
|
|
|
// DID/VC is enabled by default, but we explicitly set it here for clarity
|
|
didEnabled: true,
|
|
|
|
// Optional: AI config for the AI-powered reasoner
|
|
aiConfig: {
|
|
provider: 'openai',
|
|
model: 'gpt-4o-mini',
|
|
apiKey: process.env.OPENAI_API_KEY,
|
|
},
|
|
});
|
|
|
|
// Include the VC-enabled reasoners
|
|
agent.includeRouter(reasonersRouter);
|
|
|
|
await agent.serve();
|
|
|
|
console.log(`
|
|
╔════════════════════════════════════════════════════════════════════╗
|
|
║ Verifiable Credentials Demo Agent Started ║
|
|
╠════════════════════════════════════════════════════════════════════╣
|
|
║ Agent ID: ${agent.config.nodeId.padEnd(50)}║
|
|
║ Port: ${String(agent.config.port).padEnd(50)}║
|
|
║ DID Enabled: ${String(agent.config.didEnabled).padEnd(50)}║
|
|
╠════════════════════════════════════════════════════════════════════╣
|
|
║ Available Reasoners: ║
|
|
║ • vc_process - Basic processing with VC generation ║
|
|
║ • vc_analyze - AI analysis with VC audit trail ║
|
|
║ • vc_transform - Data transformation with VC proof ║
|
|
║ • vc_chain - Multi-step workflow with chained VCs ║
|
|
╠════════════════════════════════════════════════════════════════════╣
|
|
║ Test Commands: ║
|
|
║ ║
|
|
║ # Basic VC generation: ║
|
|
║ curl -X POST http://localhost:8080/v1/execute/vc-demo.vc_process \\
|
|
║ -H "Content-Type: application/json" \\ ║
|
|
║ -d '{"input": {"text": "Hello World"}}' ║
|
|
║ ║
|
|
║ # Check workflow in UI for VC badge (green checkmark) ║
|
|
╚════════════════════════════════════════════════════════════════════╝
|
|
`);
|
|
}
|
|
|
|
if (import.meta.url === `file://${process.argv[1]}`) {
|
|
main().catch((err) => {
|
|
console.error('Failed to start agent:', err);
|
|
process.exit(1);
|
|
});
|
|
}
|