Deployment

Deploying AI Builder use cases requires a structured approach to move your solutions from development to production. This guide outlines strategies for versioning, deploying, and managing applications across different environments to ensure reliability and performance.

Deployment Strategies

Best for: Basic applications with minimal complexity, such as simple prompting agentsIn this approach, you use a single workspace for both development and production:

Development: Create and test your application directly in the workspace
Production: Once tested, the same workspace serves as the production environment
Version Control: Use versioning to save stable points for potential rollback

Advantages:

Simplest deployment approach
No need to migrate between workspaces
Quick iteration and updates

Considerations:

Limited separation between development and production
Higher risk of disruptive changes affecting live users
Best suited for applications with low complexity and minimal regulatory constraints

Versioning Your Application

Configure Git Repository

Set up version control for your workspace:

Access workspace raw configuration by clicking “Edit source code” in the workspace action menu
Add repository configuration to the end of the file:

repositories:
  github:
    name: My GitHub Repository
    type: git
    mode: read-write
    config:
      url: https://github.com/YourUser/your-repository.git
      branch: main
      auth:
        user: 'your git user'
        password: 'your git password'

You can use secrets for sensitive information:

auth:
  password: '{{secret.gitPassword}}'

For SSH authentication, use:

auth:
  sshkey: |-
    YOUR SSH
    KEY
    HERE

Remember to use git@github.com:YourUser/your-repository.git format for SSH URLs.

Push to Repository

Save your workspace state:

From the workspace action menu, select “Versions”
Choose the repository you configured
Click “Push to repository”
Enter a commit message describing your changes
Submit to save the current state

This creates a snapshot of your workspace in the remote repository.

Pull from Repository

Retrieve saved workspace state:

From the workspace action menu, select “Versions”
Choose the repository you want to pull from
Click “Pull from repository”
Confirm the operation

This updates your workspace with the state stored in the repository.

Exclude Files from Import

Protect specific configurations during pulls:

repositories:
  github:
    name: My GitHub Repository
    type: git
    mode: read-write
    config:
      url: https://github.com/YourUser/your-repository.git
      branch: main
      auth:
        user: 'your git user'
        password: 'your git password'
    pull:
      exclude:
        - path: 'index' # Preserve workspace config
        - path: 'security' # Preserve security settings
        - path: 'pages/custom' # Preserve specific page

This prevents overwriting custom configurations when pulling changes.

Deployment Methods

Built-in Deployment

CI/CD Integration

Integrate with DevSecOps pipelines for automated deployment:Setup Steps:

Configure Git repositories for your workspaces
Create CI/CD pipeline in your preferred system (Jenkins, GitHub Actions, etc.)
Set up pipeline stages for testing, validation, and deployment
Configure deployment scripts using Prisme.ai API
Implement approval gates for production deployments

Pipeline Example:

# Example GitHub Actions workflow
name: Deploy Workspace

on:
  push:
    branches: [ main ]

jobs:
  deploy:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v2
      
      - name: Validate configuration
        run: ./validate-config.sh
        
      - name: Deploy to staging
        run: ./deploy.sh staging
        
      - name: Run tests
        run: ./run-tests.sh
        
      - name: Deploy to production
        if: success()
        run: ./deploy.sh production

CI/CD integration provides automated, consistent deployments with added security and validation.

API-Based Deployment

Create custom deployment processes using the Prisme.ai API:Key API Endpoints:

GET /workspaces/{id}/export - Export workspace
POST /workspaces/{id}/import - Import workspace
POST /workspaces/{id}/push - Push to repository
POST /workspaces/{id}/pull - Pull from repository

Example Script:

// Example Node.js deployment script
const axios = require('axios');

async function deployWorkspace(sourceId, targetId, apiKey) {
  // Export source workspace
  const exportResponse = await axios.get(`https://api.studio.prisme.ai/v2/workspaces/${sourceId}/export`, {
    headers: { 'Authorization': `Bearer ${apiKey}` },
    responseType: 'arraybuffer'
  });
  
  // Import to target workspace
  await axios.post(`https://api.studio.prisme.ai/v2/workspaces/${targetId}/import`, 
    exportResponse.data,
    { 
      headers: { 
        'Authorization': `Bearer ${apiKey}`,
        'Content-Type': 'application/octet-stream'
      }
    }
  );
  
  console.log('Deployment completed successfully');
}

deployWorkspace('source-workspace-id', 'target-workspace-id', 'your-api-key');

API-based deployment provides the most flexibility for custom deployment workflows.

Understanding What’s Versioned

Included in Versioning

Workspace components that are saved and deployed:

Pages and their configurations
Blocks and custom components
Automations and workflows
Security roles and permissions
App configurations
Workspace settings

Not Included in Versioning

Data that remains specific to each environment:

Events and their history
Collection data
Crawled documents
Uploaded files
User-specific settings
Runtime state

The versioned content is identical to what’s included in the workspace export archive.

Environment-Specific Configuration

Manage differences between environments:

Use Workspace Secrets

Store environment-specific values as secrets:

# In development workspace
secrets:
  apiEndpoint: https://dev-api.example.com
  maxRequests: 10

# In production workspace
secrets:
  apiEndpoint: https://api.example.com
  maxRequests: 100

Access these values in your application using:

{{secret.apiEndpoint}}

Conditional Logic

Implement environment-aware behavior in automations:

# Example automation with environment detection
slug: processData
do:
  - conditions: 
    "{{environment}} == 'production'":
      - callAPI:
          url: "{{secret.productionApi}}"
    default:
      - callAPI:
          url: "{{secret.developmentApi}}"

This allows your applications to adapt to different environments.

Handling Pull Results

After each import or repository pull, Prisme.ai emits a workspaces.imported event with details:

{
 "files": [
  "index",
  "security",
  "pages/test.yml",
   ...
 ],
 "deleted": [
  "automations/removedAutomation.yml"
 ],
 "version": {
  "name": "latest",
  "repository": {
   "id": "yourRepositoryId"
  }
 },
 "errors": [
  {
   "msg": "Could not rename workspace slug from {oldSlug} to {newSlug} as it is already used by workspaceId zlkpbRF",
   "err": "SlugAlreadyInUse",
   "conflictingWorkspaceId": "..."
  }
 ]
}

You can create automations that listen for this event to:

Log deployment activities
Notify team members of successful deployments
Track errors and conflicts
Trigger post-deployment tasks

Deployment Best Practices

Version Everything

Maintain complete history of your application:

Commit changes frequently with clear messages
Use branches for feature development
Tag important releases (e.g., v1.0.0)
Document significant version changes
Never work directly in production

Test Before Deployment

Validate thoroughly before moving to production:

Test in development environment first
Verify integrations with external systems
Test with realistic data sets
Include user acceptance testing
Conduct security testing

Controlled Deployment

Implement safeguards around deployment:

Use approval workflows for production changes
Deploy during low-traffic periods
Implement monitoring during deployment
Prepare rollback procedures
Document deployment steps

Environment Isolation

Maintain clear boundaries between environments:

Use separate API keys for each environment
Configure different external service endpoints
Apply appropriate security controls by environment
Use visual indicators to distinguish environments
Limit production access to necessary personnel

Troubleshooting Deployments

Import/Pull Failures

Common causes and solutions for import problems:Issue: Slug Conflicts

"Could not rename workspace slug from {oldSlug} to {newSlug} as it is already used"

Solution: Ensure unique slugs across workspaces or exclude index file from import.Issue: Missing Dependencies

"Cannot find referenced component {componentId}"

Solution: Ensure all dependencies are included in the import or available in the target workspace.Issue: Permission Errors

"User does not have permission to update {resource}"

Solution: Verify user has appropriate permissions in the target workspace.

Post-Deployment Issues

Rollback Procedures

How to revert to a previous state:Git-Based Rollback:

Identify the stable version tag or commit
Pull that specific version into your workspace:

repositories:
  github:
    config:
      branch: main
      tag: v1.0.0  # Or specific commit hash

Export/Import Rollback:

Maintain archives of known-good workspace states
Import the last stable archive if issues occur

Best Practice: Test rollback procedures regularly to ensure they work when needed.

Next Steps

RBAC

Configure role-based access control across environments

Integrations

Connect your applications to external systems

Testing & Debugging

Learn how to thoroughly test applications before deployment

Overview

AI SecureChat

AI Store

AI Knowledge

AI Builder

AI Governance

AI Collection (beta)

AI Insights (beta)

Deployment Strategies

Versioning Your Application

Deployment Methods

Understanding What’s Versioned

Included in Versioning

Not Included in Versioning

Environment-Specific Configuration

Handling Pull Results

Deployment Best Practices

Version Everything

Test Before Deployment

Controlled Deployment

Environment Isolation

Troubleshooting Deployments

Next Steps

RBAC

Integrations

Testing & Debugging

Overview

AI SecureChat

AI Store

AI Knowledge

AI Builder

AI Governance

AI Collection (beta)

AI Insights (beta)

​Deployment Strategies

​Versioning Your Application

​Deployment Methods

​Understanding What’s Versioned

Included in Versioning

Not Included in Versioning

​Environment-Specific Configuration

​Handling Pull Results

​Deployment Best Practices

Version Everything

Test Before Deployment

Controlled Deployment

Environment Isolation

​Troubleshooting Deployments

​Next Steps

RBAC

Integrations

Testing & Debugging

Deployment Strategies

Versioning Your Application

Deployment Methods

Understanding What’s Versioned

Environment-Specific Configuration

Handling Pull Results

Deployment Best Practices

Troubleshooting Deployments

Next Steps