技能 编程开发 Lindy AI集成部署指南

Lindy AI集成部署指南

v20260423
lindy-deploy-integration
这是一份关于部署与Lindy AI代理集成的应用层指南。它详细介绍了如何设置Webhook接收器、管理回调处理器以及确保应用在Docker、Vercel等多种平台上的生产环境就绪。内容涵盖了部署前置要求、分步部署流程、后部署验证脚本和回滚计划,适用于需要构建和维护复杂AI后端服务的场景。
获取技能
459 次下载
概览

Lindy Deploy Integration

Overview

Lindy agents run on Lindy's managed infrastructure. Deployment focuses on your integration layer: webhook receivers, callback handlers, and application code that Lindy agents interact with via HTTP Request actions and webhook triggers.

Prerequisites

  • Lindy agents configured and tested
  • Application with webhook receiver endpoints
  • Deployment platform (Vercel, Railway, Docker, AWS, GCP)
  • Lindy API key and webhook secrets

Instructions

Step 1: Prepare Application for Deployment

// src/server.ts — Production-ready Lindy webhook receiver
import express from 'express';
import helmet from 'helmet';

const app = express();
app.use(helmet());
app.use(express.json({ limit: '1mb' }));

// Health check for load balancer
app.get('/health', (req, res) => {
  res.json({
    status: 'ok',
    timestamp: new Date().toISOString(),
    version: process.env.APP_VERSION || 'unknown',
  });
});

// Lindy webhook receiver with auth verification
app.post('/lindy/callback', (req, res) => {
  const auth = req.headers.authorization;
  if (auth !== `Bearer ${process.env.LINDY_WEBHOOK_SECRET}`) {
    return res.status(401).json({ error: 'Unauthorized' });
  }

  // Respond immediately, process async
  res.json({ received: true });

  // Async processing
  processWebhook(req.body).catch(err => {
    console.error('Webhook processing error:', err);
  });
});

async function processWebhook(payload: any) {
  const { taskId, status, result } = payload;
  // Your business logic here
  console.log(`Task ${taskId}: ${status}`, result);
}

const PORT = process.env.PORT || 3000;
app.listen(PORT, () => console.log(`Listening on :${PORT}`));

Step 2: Docker Deployment

# Dockerfile
FROM node:20-alpine
WORKDIR /app
COPY package*.json ./
RUN npm ci --production
COPY dist/ ./dist/
EXPOSE 3000
ENV NODE_ENV=production
HEALTHCHECK --interval=30s --timeout=3s \
  CMD wget -qO- http://localhost:3000/health || exit 1
CMD ["node", "dist/server.js"]
# Build and run
docker build -t lindy-integration .
docker run -d \
  -p 3000:3000 \
  -e LINDY_API_KEY="$LINDY_API_KEY" \
  -e LINDY_WEBHOOK_SECRET="$LINDY_WEBHOOK_SECRET" \
  --name lindy-app \
  lindy-integration

Step 3: Vercel Deployment

# Install Vercel CLI
npm i -g vercel

# Set secrets
vercel secrets add lindy-api-key "$LINDY_API_KEY"
vercel secrets add lindy-webhook-secret "$LINDY_WEBHOOK_SECRET"

# Deploy
vercel --prod
// vercel.json
{
  "env": {
    "LINDY_API_KEY": "@lindy-api-key",
    "LINDY_WEBHOOK_SECRET": "@lindy-webhook-secret"
  }
}

Step 4: Update Lindy Agent Webhook URLs

After deployment, update all Lindy agents with production URLs:

  1. In Lindy dashboard, open each agent with a webhook trigger
  2. Navigate to the HTTP Request action (if agent calls your API)
  3. Update URL from dev/staging to production:
    OLD: https://abc123.ngrok.io/lindy/callback
    NEW: https://api.yourapp.com/lindy/callback
    
  4. For webhook triggers, callers need the Lindy-generated URL (unchanged)
  5. Test with a sample webhook to verify end-to-end

Step 5: Post-Deploy Verification

#!/bin/bash
echo "=== Post-Deploy Verification ==="

PROD_URL="https://api.yourapp.com"

# Health check
echo "[1/3] Health check..."
curl -sf "$PROD_URL/health" | jq .

# Webhook endpoint reachable
echo "[2/3] Webhook endpoint..."
STATUS=$(curl -s -o /dev/null -w "%{http_code}" \
  -X POST "$PROD_URL/lindy/callback" \
  -H "Authorization: Bearer $LINDY_WEBHOOK_SECRET" \
  -H "Content-Type: application/json" \
  -d '{"test": true}')
echo "Webhook endpoint: HTTP $STATUS (expect 200)"

# Trigger a test agent run
echo "[3/3] Agent trigger test..."
curl -s -X POST "https://public.lindy.ai/api/v1/webhooks/YOUR_ID" \
  -H "Authorization: Bearer $LINDY_WEBHOOK_SECRET" \
  -H "Content-Type: application/json" \
  -d '{"event": "deploy.verify", "env": "production"}'
echo "Agent triggered — check Tasks tab in Lindy dashboard"

Step 6: Rollback Plan

# If deployment fails, rollback:
# Vercel
vercel rollback

# Docker
docker stop lindy-app
docker run -d --name lindy-app-rollback \
  -e LINDY_API_KEY="$LINDY_API_KEY" \
  -e LINDY_WEBHOOK_SECRET="$LINDY_WEBHOOK_SECRET" \
  lindy-integration:previous-tag

# Update Lindy agents back to previous URLs if needed

Deployment Checklist

Step Verification
Build passes npm run build exits 0
Tests pass npm test all green
Secrets configured API key + webhook secret in platform
Health check responds GET /health returns 200
Webhook auth works POST with valid token returns 200
Webhook auth rejects POST without token returns 401
Lindy agent URLs updated HTTP Request actions point to prod
End-to-end test Trigger agent, receive callback

Error Handling

Issue Cause Solution
Webhook 502 App crashed/not running Check container logs, restart
Webhook timeout Slow processing Respond 200 immediately, process async
Wrong URL in Lindy Not updated post-deploy Update HTTP Request action URLs
SSL error Certificate issue Verify HTTPS cert is valid
Secret mismatch Dev secret in prod Verify production secrets match Lindy config

Resources

Next Steps

See lindy-webhooks-events for advanced webhook patterns.

信息
Category 编程开发
Name lindy-deploy-integration
版本 v20260423
大小 6.03KB
更新时间 2026-04-26
语言