The Dialogflow Console is an UI that lets you intuitively create interactive agents that use generative playbooks, generative data stores, and deterministic flows.
Console compatibility
Agents previously built with the Dialogflow CX console or the Vertex AI Agent Builder console can be seamlessly accessed by the Conversational Agents console. Agents created with any of these three consoles use the same Dialogflow API.
Supported regions
The console supports all regions previously supported by the Dialogflow CX console or the Vertex AI Agent Builder console.
Pricing
Pricing for agents created with the Conversational Agents console is the same as it has been for agents created with the Dialogflow CX console. Agents that use generative features such as playbooks, data stores, generators, or generative fallback for a request are charged as generative requests.
Create an agent
Prebuilt agents can be used to accelerate development for agents in common scenarios.
To create an agent with prebuilt agents:
- Open the Conversational Agents console.
- Select or create a project.
- Click Create agent.
- Click Use a prebuilt agent.
- Select a prebuilt agent and follow the installation instructions.
To create an agent with playbooks or flows:
- Open the Conversational Agents console.
- Select or create a project.
- Select Create agent.
- Select Build your own.
- Provide an agent name, location, time zone, and default language.
- Select a Conversation start option.
- If you want your agent to begin the conversation using a generative playbook, select Playbook.
- If you want your agent to begin the conversation using a deterministic flow, select Flow.
- Click Create.
- The console loads the editor page for the default playbook.
To create an agent with data store handlers:
- Go to the Conversational Agents console.
- Select or create a project.
- Select Create agent.
- Select Create a Q&A Agent.
- You are redirected to the Builder console. Follow the instructions to create an agent.
Top row navigation
The horizontal row at the top of the interface contains the following controls:
| X | Item |
|---|---|
| Project | The selected project. |
| Agent | The selected agent. |
| Language | The selected language to be used when entering language-specific data, and the Manage Languages option which brings you to flow language settings. |
| Search | Used to search, filter, and access core resources within the agent. |
| Collaborators | Collaborators that have access to the agent. |
| Export/Import agent | Agent export and import. |
| Change history | Agent change history. |
| Settings | Agent settings. |
| Publish agent | A convenience button for setting up Dialogflow CX Messenger. |
| Simulator | The simulator used for testing the agent. |
| Help | Links to documentation and other help content. |
| Feedback | Used to provide product feedback. |
Left navigation
| X | Item |
|---|---|
| Playbooks | List, create, and edit generative playbooks. |
| Flows | Visualize, create, and edit deterministic flows. |
| Tools | List, create, and edit playbook tools. |
| Prebuilts | Import ready to use prebuilt agents for a variety of common use cases. |
| Test Cases | List and define test cases to regularly verify agent behavior. |
| Conversation History | View the conversation history. |
| Environments | Create and edit various deployment environments. |
| Versions | Snapshot flow and playbook versions. |
| Integrations | Integrate your agent with ready to use user interfaces and other conversation platforms. |
| Conversation Profiles | List, create, and edit conversational profiles for advanced telephony use cases. |
Simulator
The simulator is used to test your agent. You can have a conversation with your agent while monitoring changes in session state.
Simulator settings
When you first open the simulator, configure the following settings:
- Agent environment: Select an environment.
- Start resource: Select a playbook or flow to receive the first input.
- Model or page: Select the generative model if the start resource is a playbook, or select a page if the start resource is a flow.
When interacting with the simulator, you provide end-user input as text or audio from your microphone, then press Enter or tap the Send button.
In addition to plain text, you can choose alternate input types by typing @
and then selecting an option:
- Channel: Send a channel value as a query parameter.
- DTMF: Send dual-tone multi-frequency signaling (Touch-Tone) input for telephony interactions.
- End-user metadata: Provide a contextual hint in the form of a JSON object
for generative AI to better answer an end-user query. For example, for data
store fulfillment or tool input, you could use:
{"firstName": "John", "lastName": "Doe"}. - Event: Invoke an event.
- Cloud Storage uri: Cloud Storage URI for a file to upload. Use this file for images and documents that provide context to the generative AI.
- Parameter: Inject a parameter value. Provide new parameters or preset values for existing parameters.
- Parameter scope: Set the parameter scope.
Simulator controls
Once the conversation starts, use the following controls at the top of the simulator panel:
- Debug trace
- Original response
- Show latency
- Create test case
- Undo
- Replay conversation
- Reset conversation
- Options:
- Enable or disable webhook calls at any time with the webhook toggle button.
- Enable or disable partial responses.
- Enable or disable sentiment analysis.
- Save as example: See instructions.
Simulator history
When you use the simulator to test your agent, the conversation history appears in the simulator panel. Select each turn to expand more details. For agent responses, you can also preview the audio response by tapping the play button for the turn.