Framework Integrations
Integrate Beltic with popular AI agent frameworks to make your agents verifiable
Framework Integrations
Make your AI agents verifiable by integrating Beltic's HTTP request signing into popular agent frameworks. When your agent calls external APIs, those services can verify the request came from a trusted, verified agent.
Why Sign Agent Requests?
When AI agents interact with external services, those services need to answer a critical question: "Can I trust this agent?"
Beltic solves this by providing:
- Verifiable Identity - Each agent has cryptographic credentials tied to a verified developer
- Request Signing - Every HTTP request is signed with the agent's private key
- Trust Chain - Services can verify the full chain: Agent → Developer → Beltic Platform
Supported Frameworks
Claude Agent SDK
Sign requests from Claude-powered autonomous agents
LangChain.js
Create signed API tools for LangChain agents
Integration Patterns
| Framework | Integration Method | Package |
|---|---|---|
| Claude Agent SDK | MCP Server or Hooks | @belticlabs/agent-signer |
| LangChain.js | Custom Tool | @belticlabs/agent-signer |
| Express/Node (Server) | Middleware | @belticlabs/verifier |
Quick Example
Here's the core pattern - wrap HTTP requests with Beltic signing:
import { signHttpRequest, importPrivateKey } from '@belticlabs/agent-signer';
async function signedFetch(url: string, options: RequestInit = {}) {
const privateKey = await importPrivateKey(process.env.AGENT_PRIVATE_KEY!, 'EdDSA');
const signedHeaders = await signHttpRequest({
method: options.method || 'GET',
url,
headers: options.headers as Record<string, string>,
body: options.body as string,
privateKey,
keyId: process.env.AGENT_KEY_ID!,
keyDirectoryUrl: process.env.AGENT_KEY_DIRECTORY_URL!,
});
return fetch(url, {
...options,
headers: { ...options.headers, ...signedHeaders },
});
}Prerequisites
Before integrating with any framework, you'll need:
- Beltic CLI installed -
brew install belticlabs/tap/beltic - Agent credentials - Run
beltic loginand create an agent - Keypair - Run
beltic keygento generate signing keys - Key directory hosted - Your agent's public key must be accessible via HTTPS
See the CLI Guide for detailed setup instructions.
Next Steps
Choose your framework:
- Claude Agent SDK Integration - For Claude-powered agents
- LangChain.js Integration - For LangChain-based agents
Or learn more about:
- Trust Chain Verification - How verification works
- API Reference - Complete SDK documentation