openevidence-deploy-integration

name	openevidence-deploy-integration
description	Deploy OpenEvidence integrations to healthcare production environments. Use when deploying to production, setting up staging environments, or configuring cloud deployments for clinical AI applications. Trigger with phrases like "deploy openevidence", "openevidence staging", "openevidence production deploy", "release openevidence".
allowed-tools	Read, Write, Edit, Bash(npm:*), Grep
version	1.0.0
license	MIT
author	Jeremy Longshore <[e-mail]>

OpenEvidence Deploy Integration

Overview

Deploy OpenEvidence clinical AI integrations to production healthcare environments with proper security, compliance, and rollback capabilities.

Prerequisites

• CI pipeline configured (see openevidence-ci-integration)
• Production credentials from OpenEvidence
• Cloud platform configured (GCP, AWS, Azure)
• Signed BAA for production environment

Instructions

Step 1: Multi-Environment Configuration

// src/config/environments.ts
interface OpenEvidenceEnvConfig {
  baseUrl: string;
  timeout: number;
  retries: number;
  secretPath: string;
  features: {
    deepConsult: boolean;
    webhooks: boolean;
    caching: boolean;
  };
}

export const environments: Record<string, OpenEvidenceEnvConfig> = {
  development: {
    baseUrl: 'https://api.sandbox.openevidence.com',
    timeout: 60000,
    retries: 1,
    secretPath: 'local', // Uses .env
    features: {
      deepConsult: true,
      webhooks: false,
      caching: false,
    },
  },
  staging: {
    baseUrl: 'https://api.sandbox.openevidence.com',
    timeout: 45000,
    retries: 2,
    secretPath: 'projects/staging/secrets/openevidence',
    features: {
      deepConsult: ,
      : ,
      : ,
    },
  },
  : {
    : ,
    : ,
    : ,
    : ,
    : {
      : ,
      : ,
      : ,
    },
  },
};

  ():  {
   env = process.. || ;
   environments[env] || environments.;
}

Step 2: GitHub Actions Deployment Workflow

# .github/workflows/deploy.yml
name: Deploy OpenEvidence Integration

on:
  push:
    branches: [main]
    tags: ['v*']
  workflow_dispatch:
    inputs:
      environment:
        description: 'Deployment environment'
        required: true
        default: 'staging'
        type: choice
        options:
          - staging
          - production

env:
  PROJECT_ID: ${{ secrets.GCP_PROJECT_ID }}
  SERVICE_NAME: clinical-evidence-api
  REGION: us-central1

jobs:
  test:
    uses: ./.github/workflows/openevidence-ci.yml
    secrets: inherit

  build:
    needs: test
    runs-on: ubuntu-latest
    outputs:
      image: ${{ steps.build.outputs.image }}
    steps:
      -  

          
         
        
             

         
              

          
         
         
  
     
     
     
       
    
        

          
         
        
             

            
         
          
        
             
             
         
  
     [, ]
     
     
      
    
        

          
         
        
             

              
         
           
         
           
         
          
             
             
                  

           [    ] 
             
             
          

         
         
         
         
        
           

Step 3: Kubernetes Deployment

# k8s/openevidence-deployment.yaml
apiVersion: apps/v1
kind: Deployment
metadata:
  name: clinical-evidence-api
  labels:
    app: clinical-evidence-api
spec:
  replicas: 3
  selector:
    matchLabels:
      app: clinical-evidence-api
  template:
    metadata:
      labels:
        app: clinical-evidence-api
    spec:
      serviceAccountName: clinical-evidence-api
      containers:
        - name: api
          image: gcr.io/PROJECT_ID/clinical-evidence-api:VERSION
          ports:
            - containerPort: 8080
          env:
            - name: NODE_ENV
              value: production
            - name: OPENEVIDENCE_API_KEY
              valueFrom:
                secretKeyRef:
                  name: openevidence-secrets
                  key: api-key
            -  
              
                
                   
                   
          
            
               
               
            
               
               
          
            
               
               
             
             
          
            
               
               
             
             

 
 

   

  
     
  
      
       
   

 
 

   
  
     
     

  
     
         
       
  
      
      
        
            
             
            
              
                 
                
                   

Step 4: Health Check Endpoint

// src/routes/health.ts
import { Router } from 'express';
import { OpenEvidenceClient } from '@openevidence/sdk';
import { getConfig } from '../config/environments';

const router = Router();

router.get('/health', (req, res) => {
  res.json({
    status: 'healthy',
    timestamp: new Date().toISOString(),
    version: process.env.npm_package_version,
  });
});

router.get('/health/openevidence', async (req, res) => {
  const startTime = Date.now();

  try {
    const client = new OpenEvidenceClient({
      apiKey: process.env.OPENEVIDENCE_API_KEY!,
      orgId: process.env.OPENEVIDENCE_ORG_ID!,
      baseUrl: getConfig().baseUrl,
    });

    await client.health.check();

    res.({
      : ,
      : .() - startTime,
      : process..,
      : ().,
    });
  }  (: ) {
    res.().({
      : ,
      : .() - startTime,
      : error.,
    });
  }
});

  router;

Step 5: Rollback Procedure

#!/bin/bash
# scripts/rollback.sh

set -e

SERVICE_NAME="clinical-evidence-api"
REGION="us-central1"

echo "=== OpenEvidence Integration Rollback ==="

# List recent revisions
echo "Recent revisions:"
gcloud run revisions list --service $SERVICE_NAME --region $REGION --limit 5

# Get previous revision
CURRENT=$(gcloud run services describe $SERVICE_NAME --region $REGION \
  --format 'value(status.traffic[0].revisionName)')
PREVIOUS=$(gcloud run revisions list --service $SERVICE_NAME --region $REGION \
  --format 'value(metadata.name)' --limit 2 | tail -1)

echo "Current: $CURRENT"
echo "Rolling back to: $PREVIOUS"

read -p "Proceed with rollback? (y/n) " -n 1 -r
echo
if [[ $REPLY =~ ^[Yy]$ ]]; then
  gcloud run services update-traffic $SERVICE_NAME \
    --region $REGION \
    --to-revisions $PREVIOUS=100

  echo "Rollback complete. Monitoring..."
  sleep 30

  # Verify health
  gcloud run services describe $SERVICE_NAME --region $REGION \
    --format 'value(status.url)' | xargs -I {} curl -sf {}/health/openevidence

Output

• Multi-environment deployment configuration
• CI/CD pipeline with canary releases
• Kubernetes deployment manifests
• Health check endpoints
• Rollback procedures

Deployment Checklist

• Staging environment tested
• Production secrets configured
• Health checks passing
• Monitoring alerts configured
• Rollback procedure tested
• On-call notified of deployment
• Clinical team notified of changes

Error Handling

Deployment Issue	Detection	Resolution
Health check fails	Readiness probe	Rollback to previous revision
High error rate	Monitoring alert	Rollback, investigate logs
Secret missing	Container fails to start	Check secret manager
Rate limit hit	429 errors spike	Scale down, check quotas

Resources

• Cloud Run Documentation
• Kubernetes Deployments

Next Steps

For webhook configuration, see openevidence-webhooks-events.