[[["易于理解","easyToUnderstand","thumb-up"],["解决了我的问题","solvedMyProblem","thumb-up"],["其他","otherUp","thumb-up"]],[["很难理解","hardToUnderstand","thumb-down"],["信息或示例代码不正确","incorrectInformationOrSampleCode","thumb-down"],["没有我需要的信息/示例","missingTheInformationSamplesINeed","thumb-down"],["翻译问题","translationIssue","thumb-down"],["其他","otherDown","thumb-down"]],["最后更新时间 (UTC):2025-08-18。"],[[["\u003cp\u003eThe Conversational Agents console is a new platform combining features from the Dialogflow CX and Agent Builder consoles, designed for creating interactive agents with both generative and deterministic features.\u003c/p\u003e\n"],["\u003cp\u003eUsers of the Agent Builder console are advised to migrate to the Conversational Agents console, and Dialogflow CX console users interested in generative features should also transition.\u003c/p\u003e\n"],["\u003cp\u003eAgents previously built in either the Dialogflow CX or Agent Builder consoles are fully compatible and accessible within the Conversational Agents console.\u003c/p\u003e\n"],["\u003cp\u003eThe Conversational Agents console offers functionalities such as generative playbooks, data stores, deterministic flows, prebuilt agents, and a simulator, maintaining the same pricing as Dialogflow CX while charging generative feature requests as such.\u003c/p\u003e\n"],["\u003cp\u003eThe Conversational Agents console is currently in a pre-GA stage, meaning it is "as is" with potentially limited support, and the documentation is still being refined to be specific to the console.\u003c/p\u003e\n"]]],[],null,["# Conversational Agents console overview\n\n| **Important:** This document is a work in progress. The features used in each of the consoles are conceptually the same, but the detailed steps for using these features may vary. Over the next few months, the feature documentation will be updated to be specific to the Conversational Agents console.\n\nThe Conversational Agents console is a new console that includes\nfeatures from both the Dialogflow CX console and the AI Applications console.\nAll users of the AI Applications console should start using the\nConversational Agents console.\nUsers of the Dialogflow CX console that want to incorporate generative features\ninto agents should start using the\nConversational Agents console.\n\nUsing the Conversational Agents console,\nyou can create interactive agents that use\n[generative playbooks](/dialogflow/cx/docs/concept/playbook),\n[generative data stores](/dialogflow/cx/docs/concept/data-store),\nand\n[deterministic flows](/dialogflow/cx/docs/basics).\n\nThe Dialogflow CX console will continue to be supported until the\nConversational Agents console is generally available (GA).\n\nConsole compatibility\n---------------------\n\nAgents previously built with the Dialogflow CX console or the AI Applications console\ncan be seamlessly accessed by the Conversational Agents console.\nAgents created with any of these three consoles\nuse the same Dialogflow API.\n\nSupported regions\n-----------------\n\nThis console supports all\n[regions](/dialogflow/cx/docs/concept/region)\nalready supported by the Dialogflow CX console or the AI Applications console.\n\nPricing\n-------\n\n[Pricing](/dialogflow/pricing)\nfor agents created with the Conversational Agents console\nis the same as it has been for agents created with the Dialogflow CX console.\nAgents that utilize generative features such as playbooks,\ndata stores, generators, or generative fallback\nfor a request are charged as generative requests.\n\nCreate an agent\n---------------\n\nPrebuilt agents can be used to accelerate development for new agents\nin common scenarios.\nTo create an agent with prebuilt agents:\n\n1. Open the [Conversational Agents console](https://conversational-agents.cloud.google.com).\n2. If you haven't already, select or create a project.\n3. Click **Create agent**.\n4. Click **Use a prebuilt agent**.\n5. Select a prebuilt agent and follow instructions for installation.\n\nTo create an agent with\n[playbooks](/dialogflow/cx/docs/concept/playbook)\nand/or\n[flows](/dialogflow/cx/docs/basics):\n\n1. Open the [Conversational Agents console](https://conversational-agents.cloud.google.com).\n2. If you haven't already, select or create a project.\n3. Click **Create agent**.\n4. Click **Build your own**.\n5. Provide an agent name, [location](/dialogflow/cx/docs/concept/region), time zone, and [default language](/dialogflow/cx/docs/concept/agent-multilingual).\n6. 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**.\n7. Click **Create**.\n8. The console loads the editor page for the [default playbook](/dialogflow/cx/docs/concept/playbook#default).\n\nTo create an agent with\n[data store handlers](/dialogflow/cx/docs/concept/data-store/handler):\n\n1. Open the [Conversational Agents console](https://conversational-agents.cloud.google.com).\n2. If you haven't already, select or create a project.\n3. Click **Create agent**.\n4. Click **Create a Q\\&A Agent**.\n5. You are redirected to the AI Applications console. Follow the instructions to [create an agent](/dialogflow/cx/docs/concept/data-store/handler#create-agent).\n\nTop row navigation\n------------------\n\nThe horizontal row at the top of the interface contains the following controls:\n\nLeft navigation\n---------------\n\nSimulator\n---------\n\nThe simulator is used to test your agent.\nYou can have a conversation with your agent\nwhile monitoring changes in session state.\n\n### Simulator settings\n\nWhen you first open the simulator,\nyou can configure the following settings:\n\n- [Agent environment](/dialogflow/cx/docs/concept/version)\n- Start resource, which can be a [playbook](/dialogflow/cx/docs/concept/playbook) or [flow](/dialogflow/cx/docs/concept/flow) which receives the first input.\n- The generative model (start resource is a playbook) or a page (start resource is a flow).\n\n### Simulator input\n\nWhen interacting with the simulator,\nyou provide end-user input as text or audio from your microphone,\nthen press enter or click the\nsend sendbutton.\n\nIn addition to plain text,\nyou can choose alternate input types by typing `@`,\nthen selecting the option:\n\n- **Channel** : Send a [channel](/dialogflow/cx/docs/concept/fulfillment#channel-specific) value as a query parameter.\n- **DTMF**: Send dual-tone multi-frequency signaling (Touch-Tone) input for telephony interactions.\n- **End user metadata** : Provide 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\"}`.\n- **Event** : Invoke an [event](/dialogflow/cx/docs/concept/handler#event).\n- **Cloud Storage uri**: Cloud Storage URI for a file to upload. You can use this file for images and documents that provide context to the generative AI.\n- **Parameter** : Inject a [parameter](/dialogflow/cx/docs/concept/parameter) value. You can provide new parameters or provide preset values for existing parameters.\n- **Parameter scope** : Set the [parameter scope](/dialogflow/cx/docs/concept/handler#scope).\n\n### Simulator controls\n\nOnce the conversation has started,\nthere are several controls available at the top of the simulator panel:\n\n- **Debug trace**\n- **Original response**\n- **Show latency**\n- **Create test case**\n- **Undo**\n- **Replay conversation**\n- **Reset conversation**\n- **Options** :\n - Enable or disable [webhook](/dialogflow/cx/docs/concept/webhook) calls at any time with the webhook toggle button.\n - Enable or disable [partial responses](/dialogflow/cx/docs/concept/fulfillment#partial-response).\n - Enable or disable [sentiment analysis](/dialogflow/cx/docs/concept/sentiment).\n- **Save as example** : [See instructions](/dialogflow/cx/docs/quick/build-agent-playbook#provide_examples)\n\n### Simulator history\n\nWhen using the simulator to have a conversation with your agent,\nthe conversation history is shown in the simulator panel.\nYou can click each conversational turn\nto expand more details about that turn.\nFor agent responses,\nyou can also preview the audio response by clicking the\nplay button for the conversational turn."]]