ZBrain Documentation
  • ZBrain Documentation
    • Components of ZBrain
    • Getting started with ZBrain
    • 📚Knowledge base
      • How to create a knowledge base?
      • Knowledge source
      • Information schema
      • File summary
      • Automated reasoning
      • Retrieval testing
      • Knowledge base settings
    • 📱APP
      • How to create a new app?
      • How to set up and customize your app?
      • How to access the app reporting dashboard?
    • 🤖ZBrain AI agents
      • Get started with agents on ZBrain
      • Deploying pre-built agents
      • Creating custom AI agents
        • Agent setup
        • Define input sources
        • Define Flow
          • Key elements of a Flow
          • Flow Components
            • Activepieces Platform
            • AITable
            • Airtable
            • Amazon S3
            • Amazon SNS
            • Amazon SQS
            • Amazon Textract
            • Apify
            • Apollo
            • Approval
            • Asana
            • Azure Communication Services
            • Azure Cosmos DB
            • Azure Document Intelligence
            • Azure OpenAI
            • Azure Translation
            • Bannerbear
            • Baserow
            • Beamer
            • Bedrock Claude
            • Bettermode
            • Binance
            • Bing Search
            • Blackbaud
            • Bonjoro
            • Box
            • Brevo
            • Brilliant Directories
            • Bubble
            • CSV
            • Calendly
            • Certopus
            • Clearout
            • Clockodo
            • Code
            • Confluence
            • Connections
            • Constant Contact
            • Contiguity
            • Contentful
            • Customer.io
            • Crypto
            • Databricks
            • Data Mapper
            • Date Helper
            • DeepL
            • Delay
            • Discord
            • Discourse
            • Drip
            • Dropbox
            • Dust
            • Facebook Pages
            • Figma
            • Files Helper
            • Flowise
            • Flowlu
            • Formbricks
            • Frame
            • Freshdesk
            • Freshsales
            • GCloud Pub/Sub
            • GenerateBanners
            • GhostCMS
            • GitHub
            • GitLab
            • Gmail
            • Google Calendar
            • Google Contacts
            • Google Docs
            • Google Drive
            • Google Forms
            • Google Gemini
            • Google My Business
            • Google Search
            • Google Search Console
            • Google Sheets
            • Google Tasks
            • Groq
            • Hacker News
            • Heartbeat
            • HubSpot
            • HTTP
            • Image Helper
            • Inputs
            • Instagram for Business
            • Intercom
            • Invoice Ninja
            • Jira Cloud
            • Jotform
            • Kimai
            • Kizeo Forms
            • LeadConnector
            • Line Bot
            • Linear
            • LinkedIn
            • LinkedIn Actions
            • LLMRails
            • Lusha
            • MailerLite
            • Mailchimp
            • Mautic
            • Microsoft Dynamics 365 Business Central
            • Microsoft Dynamics CRM
            • Microsoft Excel 365
            • Microsoft OneDrive
            • Microsoft Outlook Calendar
            • Microsoft Teams
            • Mixpanel
            • MongoDB
            • Notion
            • Odoo
            • OpenAI
            • OpenRouter
            • Pastebin
            • PDF
            • Postgres
            • PostHog
            • Pushover
            • Qdrant
            • Queue
            • Razorpay
            • Router
            • Salesforce
            • SendGrid
            • ServiceNow
            • SFTP
            • SharePoint
            • Slack
            • SMTP
            • Snowflake
            • SOAP
            • Spotify
            • Stability AI
            • Stable Diffusion Web UI
            • Storage
            • Stripe
            • SurrealDB
            • SurveyMonkey
            • Taskade
            • Telegram Bot
            • Text Helper
            • Trello
            • Twilio
            • Twitter
            • Utilities
            • WhatsApp Business
            • WordPress
            • XML
            • YouTube
            • ZBrain
            • Zendesk
            • ZeroBounce
            • Zoho Books
            • Zoho CRM
            • Zoho Invoice
            • Zoom
          • How to Define a Flow?
          • How to Test Each Step in the Flow?
        • Configure Additional Settings
        • Test and Deploy Agents
        • How to access, monitor, and manage agent performance and tasks?
    • Settings
    • 📖API Tutorials
      • 📚Knowledge base
        • Automated Reasoning
      • 📱APP
      • 🤖Agents
Powered by GitBook
On this page
  • Step 1: Test the Agent
  • Step 2: Deploy the Agent
  • Post-Deployment Monitoring
  1. ZBrain Documentation
  2. ZBrain AI agents
  3. Creating custom AI agents

Test and Deploy Agents

PreviousConfigure Additional SettingsNextHow to access, monitor, and manage agent performance and tasks?

Last updated 4 months ago

This step ensures that your agent performs as expected before being fully deployed to handle tasks. This phase involves testing the agent with sample data, verifying its functionality, making necessary adjustments, and then deploying it for use. After deployment, you can monitor its performance to ensure it continues to operate effectively.

Step 1: Test the Agent

Testing is critical to ensure your agent functions as expected before going live. To begin, navigate to the Test Agent page and follow the steps outlined below to thoroughly test your agent's performance.

  1. Upload a Sample Document: Provide a sample document for the agent to process, simulating real-world conditions. This allows you to assess its ability to handle actual input data.

  2. Monitor Processing: Review the reports generated during the test to ensure the agent performs its tasks correctly and efficiently.

  1. Adjust Configurations if Needed:

    • If the agent’s performance isn’t as expected, click ‘Configure’ to return to the Additional Settings page.

    • Make any necessary adjustments to input sources, output sources, or workflows.

  1. Repeat Testing: Continue testing and adjusting configurations until the agent functions flawlessly, ensuring it meets your operational requirements.

Step 2: Deploy the Agent

Once the agent passes the testing phase, it is ready for deployment. To deploy the agent:

  1. Click ‘Deploy Agent’: At the bottom-right corner of the Test Agent page, click the ‘Deploy Agent’

  2. Redirect to Agent Dashboard Page: After deployment, you will be redirected to the Agent Dashboard Page, where you can monitor the agent’s performance.

Post-Deployment Monitoring

After deployment, continuous monitoring is crucial to ensure optimal performance:

  1. Access Uploaded Documents: Navigate to the Agent Dashboard to review any uploaded documents and verify that they are processed as expected.

  2. Review Activity Logs and Reports: Check the logs and generated reports to track the agent’s performance and detect any issues.

  1. Monitor the Agent’s Performance: Navigate to the Performance Page to track key metrics. Regularly review these details to ensure the agent is processing tasks efficiently and handling live data effectively.

By thoroughly testing and deploying the agent, you ensure its reliability and smooth integration into your workflow. Ongoing monitoring after deployment helps maintain performance and address issues proactively.

🤖