SSAVQO - Documentation Index

Table of Contents

• Release notes
  • What's new
    • Previous releases
      • Release notes for May 2026
      • Release notes for April 2026
      • Release notes for March 2026
      • Release notes for February 2026
      • Release notes for January 2026
      • Release notes for 2025
      • Release notes for 2024
      • Release notes for 2023
      • Release notes for 2022
  • Known issues and limitations
    • Control plane
    • Building agents
    • Connections and authentication
    • Building tools
    • Deploying and channels
    • Using Orchestrate Chat
    • Evaluation
    • Monitoring and analytics
    • Administration
  • Feature parity across deployments
  • Prebuilt agent feature parity across deployments
  • Deprecations
    • Deprecated and withdrawn functions
    • Deprecated and withdrawn models
  • Release notes for ADK
  • Release notes for On-premises
  • Regional availability and outbound IP addresses
  • Regulatory compliance
    • Accessibility features
    • Security and Privacy by Design (SPbD) overview
    • Notices
  • Licenses and entitlements
    • Licenses and entitlements on Amazon Web Services
    • Licenses and entitlements on Amazon Web Services GovCloud (US)
    • Licenses and entitlements on Groq
    • Licenses and entitlements on IBM Cloud
      • Managing service usage on IBM Cloud
    • Licenses and entitlements for third-party models
    • Understanding billing and usage tracking
• Getting started
  • Logging in to IBM watsonx Orchestrate
    • Understanding authentication in AWS and regulated environments
    • Using the IBM SaaS Console with accounts
      • Getting started with the IBM SaaS Console with accounts
      • Granting access through service IDs and API keys from the IBM SaaS Console
      • Email notifications
    • Using the IBM SaaS Console
      • Getting started with the IBM SaaS Console
    • Accessing the IBM watsonx Orchestrate trial version
  • Explore documentation sources
  • Getting help and support
    • Collecting diagnostic data for performance issues
  • Installing IBM watsonx Orchestrate On-premises
  • Tutorials to get started with the agents
    • Sales Contract Management agent
      • Connecting Microsoft for Sales Contract Management agent
      • Creating the Sales Contract Management agent
      • Monitoring the Sales Contract Management agent
• Designing
  • Discovering the catalog
    • Prebuilt agents
      • Agents for finance operations
      • Agents for HR operations
        • Connecting watsonx Employee Support agents to Slack
      • Agents for IT operations
      • Agents for legal operations
      • Agents for procurement operations
      • Agents for productivity
      • Agents for sales operations
        • IBM watsonx Sales Prospecting Manager
      • Agents for supply chain operations
      • Prebuilt agents videos
      • Visualizing artifact hierarchy
    • Prebuilt tools
      • Tools for finance operations
      • Tools for HR operations
      • Tools for IT operations
      • Tools for legal operations
      • Tools for procurement operations
      • Tools for productivity
      • Tools for sales operations
      • Tools for sales operations
    • Managing agent and tool versions
    • Categories
      • Finance
      • HR
      • IT
      • Legal
      • Procurement
      • Productivity
      • Sales
      • Supply Chain
    • Partner agents
      • Partner native agents
      • External partner agents
      • Partner A2A (Agent2Agent) agents
      • Purchasing partner agents
    • Apps
    • Catalog onboarding tutorials
    • Exploring MCP servers
    • Exploring OpenAPI tools
  • Preparing to build AI agents
    • Introduction to AI agents
    • Understanding the core components of an AI agent
    • Designing high-performance AI agents
• Building
  • Building agents
    • Creating and customizing agents
      • Working with agent and tool versions
      • Recommendations for agent descriptions
      • Choosing an AI model for the agent
        • Considerations for choosing an AI model
        • Available AI models
          • GPT-OSS model behavior and instruction guidelines
        • Adding custom AI models
      • Choosing a style for the agent
      • Customizing the welcome message and starter prompts
      • Configuring agent behavior
        • Adding instructions
        • Adding guidelines
        • Enabling document-based interactions in chat
        • Enabling scheduling for agents
      • Enhancing agents with rich responses
    • Orchestrating agents
      • Understanding agent orchestration
      • Context sharing with collaborator agents
      • Adding agents from catalog
      • Adding an agent from a local instance
      • Importing LangGraph agents
        • Creating a LangGraph package for import
        • Creating connections for LangGraph agents
        • Enabling state persistence for LangGraph agents
      • Importing agents from external platforms
        • Adding agents from third-party platforms
        • Adding watsonx Orchestrate assistants as collaborator agents
        • Adding AI assistants as collaborator agents
    • Overview of workspaces
      • Managing access with trusted profiles
      • Working with workspaces
      • Managing workspace artifacts
      • Workspace privacy and collaboration
      • Workspace naming and limits
    • Building agents using the ADK
    • Agents onboarding tutorials
  • Building tools
    • Agentic workflows
      • Creating agentic workflows in the workflow builder
        • Adding tools
        • Adding agents
        • Adding user activities
          • Configuring multiple languages for user interactions
        • Adding logic blocks
        • Adding generative prompts
        • Adding decision
        • Adding document classifiers
        • Adding document extractors
        • Adding text extractors
        • Inspecting agentic workflows
        • Getting started with document processing
        • Consuming intermediate context updates in agentic workflows
        • Scheduling agentic workflows in chat
        • Masking sensitive information in agentic workflows
      • Logic blocks
      • Videos
      • View and edit pro-code workflows in the workflow builder
      • Arranging a workflow layout automatically
    • OpenAPI tools
      • Importing from an OpenAPI
    • MCP servers
      • Importing tools from an MCP server
        • Adding MCP server tools to multiple agents
        • Adding MCP server tools to a single agent
        • Adding MCP server tools from the catalog to an agent
        • Adding MCP server tools from your local instance to an agent
        • Managing MCP servers
      • Configuring MCP server connections
    • Tools onboarding tutorials
    • Customizing input-output schemas of prebuilt tools
  • Building AI assistants
    • Accessing AI assistant builder
    • Building your AI assistant with actions
      • Understanding your users' questions or requests
      • Boost words
      • Adding AI assistant responses
      • Collecting information from your customers
      • Using variables to manage conversation information
      • Adding conditions to an action
      • Adding conditions to a step
      • Choosing what to do at the end of a step
      • Adding support for global audiences
      • Adding actions from web-based automation flows
    • Managing and organizing your actions
      • Duplicating an action
      • Copying an action
      • Uploading or downloading all actions
      • Filtering actions
      • Organizing actions in collections
      • Calling a custom extension
      • Allowing your customers to change the topic of the conversation
      • Providing options when a question or request cannot be answered
      • Connecting to a live agent
      • Detecting trigger words
      • Saving your actions
      • Reviewing and debugging your actions
      • Global settings for actions
      • Response modes
      • Autocorrecting user input
      • Algorithm version and training
    • Using watsonx.ai for generative AI capabilities
      • Information gathering
      • Hate, Abuse, and Profanity Filter
      • Conversational search
    • Overview: Customizing and developing
      • Configuring base LLM
      • Defining responses by using the JSON editor
      • Dynamic options
      • Guiding customers with journeys
      • Writing expressions
      • Extending your assistant with webhooks
        • Making a call before processing a message
          • Calling a service before processing a message for IBM Cloud Pak for Data
          • Calling a service before processing a message for IBM Cloud Pak for Data - classic experience
          • Calling a service before processing a message for watsonx Orchestrate
        • Making a call after processing a message
          • Calling a service after processing a message for IBM Cloud Pak for Data
          • Calling a service after processing a message for IBM Cloud Pak for Data - classic experience
          • Calling a service after processing a message for watsonx Orchestrate
        • Authentication method for webhooks
        • Defining an audio webhook
        • Recording a caller's utterance
      • Logging activity with a webhook
      • Building a custom extension
      • Adding an extension to your assistant
      • Understanding extension matching during action migration
      • Adding search
        • Elasticsearch search integration set up
        • Custom service integration setup
        • Milvus search integration setup
        • Search integration enhancement
      • AI assistant builder API overview
        • Generating an API key for AI assistant builder
        • API methods summary
        • Building a custom client by using the API
        • Processing input attachments
    • Overview: Previewing and publishing
      • Previewing and sharing your assistant
      • Publishing your content
      • Adding and using multiple environments
    • Deploying your AI assistant
      • Deploying your assistant
      • Web chat overview
        • How the web chat works
        • Web chat development overview
        • Web chat development tutorials
          • Tutorial: Customizing the size and position of the web chat
          • Tutorial: Setting context variables for actions from the web chat
          • Tutorial: Authenticating a user in the middle of a session
          • Tutorial: Rendering a custom response as a content carousel
          • Tutorial: Customizing the greeting action based on the current page
          • Tutorial: implementing custom option buttons in the web chat
          • Tutorial: Customizing text elements based on the current page
          • Tutorial: Providing a downloadable conversation transcript
          • Tutorial: Interacting with the host web page
          • Tutorial: Displaying a pre-chat or post-chat form
        • Embedding the web chat on your page
        • Web chat setup overview
        • Configuring style and appearance
        • Configuring the home screen
        • Configuring suggestions
        • Supporting global audiences in web chat
        • Controlling the web chat version
        • Adding contact center support
      • Adding integrations
        • Integrating with Genesys
        • Integrating with Salesforce
        • Integrating with Zendesk
        • Integrating with Facebook Messenger
        • Integrating with Genesys Bot Connector
        • Integrating with Genesys Audio Connector
        • Integrating with Microsoft Teams
        • Integrating with Slack
        • Integrating with WhatsApp
        • Integrating with a custom client app
      • Deploying for phone
        • Integrating with phone
        • Phone integration configuration
        • Integrating with phone and Genesys Cloud
        • Integrating with phone and Twilio Flex
        • Integrating with SMS
        • Handling phone interactions
        • Troubleshooting and Logs
        • Phone integration context variables
      • Adding more assistants
      • Deleting assistants
    • Use analytics to review your entire assistant at a glance
      • Conversational search analytics
      • Evaluating the assistant
      • Understand your most and least successful actions
      • Use unrecognized requests to get action recommendations
      • Review customer conversations
      • Sending events to Segment
    • Browser support
    • Securing your AI assistant
      • Securing your web chat
        • Enabling web chat security
        • Authenticating users in web chat
        • Encrypting sensitive data in web chat
        • Adding the web chat to your mobile application
        • Managing user identity information in web chat
  • Configuring
    • Adding knowledge to agents
      • Creating knowledge descriptions
      • Customizing knowledge settings
      • Building a knowledge source
      • Connecting to an Astra DB content repository
      • Connecting to an Elasticsearch content repository
        • Configuring custom filters for Elasticsearch
        • Configuring query body for Elasticsearch
      • Connecting to an OpenSearch content repository
      • Connecting to a Milvus content repository
        • Enabling granular access control for Milvus
      • Connecting to a custom content service
      • Uploading files as a knowledge source
        • Using Tiny Milvus for On-premises
    • Managing tools
      • Adding from a catalog
      • Adding from a local instance
    • Configuring voice
      • Configuring voice settings for agents
      • Connecting agents to Phone
        • Connecting to Phone with Genesys Audio Connector
        • Connecting to Phone with SIP trunk
          • Configuring Tenant ID for service providers
      • Enabling real-time voice agents
    • Connecting to channels
      • Connecting agents to Orchestrate Chat
      • Customizing Orchestrate Chat using agents
      • Channel-specific context variables
      • Connecting to Facebook Messenger
      • Connecting to Genesys Bot Connector
      • Connecting to Microsoft Teams
      • Connecting to Slack
        • Enabling form interaction within Slack
      • Connecting to SMS with Twilio
      • Connecting to WhatsApp with Twilio
    • Connecting to a live agent
      • Adding a custom service desk
    • Enabling Agent Assist in Genesys platform
      • Integration with the Genesys platform
      • Getting values from browser developer tools
      • Owning a conversation redirected by an agent
      • Accessing summaries
    • Managing app connections and credentials
      • Connection overview
      • Connection architecture and lifecycle
      • Authentication types overview
      • Credentials overview
      • Connections across environments and deployments
      • Creating and managing connections
        • Connection status: definition and connect behavior
        • Managing app credentials
          • Managing team credentials
          • Managing member credentials
      • Binding and deploying connections
      • Connecting apps for prebuilt agents
        • Adobe Workfront
        • Box
        • Coupa
        • Dun & Bradstreet for Sales
        • Dun & Bradstreet
        • Dropbox
        • Dynamics 365
        • GitLab
        • Gmail
        • Google Drive
        • HubSpot
        • IBM Cloud Object Storage
        • IBM Planning Analytics
        • IBM Targetprocess
        • Jenkins
        • Jira
        • Maximo IT
        • Microsoft Outlook
        • Microsoft SharePoint
        • Microsoft Teams
        • MySQL
        • Onit
        • Oracle Fusion
        • Oracle HCM
        • Red Hat Ansible
        • Salesforce
        • SAP Ariba
        • SAP S/4HANA
        • SAP SuccessFactors
        • Salesloft
        • Seismic
        • ServiceNow
        • Slack
        • Sterling Order Management System (OMS)
        • Supply chain IBM SIP
        • TestRail
        • Twilio
        • Workday HCM
        • Zendesk
        • ZoomInfo
      • Authenticating your users with single sign-on
        • OAuth2 On Behalf Of Flow
        • OAuth2 Token Exchange
        • OAuth2 Direct Access Token
      • Managing agent access using the security control center
  • Testing
    • Debugging your agent
    • Evaluating with simulated users
      • Recording and managing test cases
      • Running evaluations
      • Understanding evaluation metrics
  • Deploying
    • Environments
    • Deploying agents to live environment
    • Publishing to Catalog
      • Design guidelines for agent icons
    • Integrating agents with web applications
      • Embedding the agent to web applications
      • Enabling voice capabilities in the embedded agent
      • Securing the embedded chat
        • Configuring security for embedded chat
          • Configuring security with scripting
      • Embedded chat tutorials
        • Creating a language selector in the chat header
        • Loading an existing conversation thread
        • Sending silent messages to trigger agent responses
        • Triggering messages from URL parameters
        • Display custom content with writeable elements
• Using
  • Using Orchestrate Chat
    • Providing feedback on agent responses
    • Uploading files during a conversation
    • Scheduling tasks
    • Understanding agent memory
      • Agent memory retention policy
      • Using agent memory in conversations
      • Managing agent memory
      • Memory behavior and limits
      • Frequently asked questions
• Operating
  • Agentic control plane overview
    • Getting started with the control plane dashboard
    • Control plane agent
    • Managing operational alerts
    • Controls overview
      • Managing controls
    • Reviewing agent fleet health
  • Agent analytics and monitoring
    • Analytics dashboard
    • Agent-level metrics
    • Metrics overview
    • Understanding metrics
    • Monitoring agents (legacy experience)
      • Understanding trace details
      • Practical guidance for reading traces
• Administering
  • Administering your instance
    • Managing users
      • Roles on IBM watsonx Orchestrate
      • Managing users on AWS
      • Managing users on AWS (GovCloud, HIPAA, Premium)
      • Managing users on IBM Cloud
    • Managing account access
      • Configuring SSO for platform access
      • Managing your account
    • Managing instance settings
      • Managing language preferences
      • Managing credentials
      • Managing agent analytics settings
      • Managing model selection settings
      • Restricting builder access to security settings
    • Managing resilience and availability
      • Managing data retention
      • Configuring instances for multi-region active deployment
      • High availability, business continuity, backups and disaster recovery on AWS
      • High availability, business continuity, backups and disaster recovery on IBM Cloud
      • Data isolation on watsonx Orchestrate
      • Backing up and restoring AI assistant data
    • Securing your instance
      • Chat privacy and data protection
      • Creating private paths using AWS PrivateLink
      • Enhancing security for instances on IBM Cloud
        • Allowing specific IP addresses
        • Configuring TLS tunnel
      • Enhancing security for instances on AWS
        • Security, privacy, and data protection
        • Allowing specific IP addresses on AWS
      • Monitoring and auditing platform activity
        • Activity tracking events on IBM Cloud
        • Enabling external logging for AWS
        • Migration for existing users
        • List of events for activity tracking
      • Using private network endpoints
  • Migrating from watsonx Assistant
    • Migration overview
    • Planning your migration
    • Migrating to watsonx Orchestrate
    • Automatic migration
    • Modernizing AI assistants into AI Agents
• Learning resources
• Reference
  • AI assistant reference
    • Response types reference
    • Preview runtime errors
    • Built-in global variables
    • Security certificates reference
    • Integration variables
    • Expression language methods for actions
    • Phone integration context variables
    • SMS integration reference
    • CDR log event reference
    • Segment event reference
    • Web chat API reference
    • REST API reference
    • Filter query reference
  • IBM watsonx Orchestrate API
    • Getting started with the API
      • Getting the API endpoint
      • Generating an API key for AWS
      • Getting service instance URL
      • Generating a JWT for AWS
      • Generating an API key for IBM Cloud
      • Generating an access token for IBM Cloud
      • Generating an API key for on-premises
    • API reference
  • Using the skill-based experience
  • Glossary of terms