Stay organized with collections
Save and categorize content based on your preferences.
This document is a basic guide to Agent Assist resources. More details
about the resources described here can be found in the
reference documentation.
We recommend that you familiarize yourself with this guide before diving into
Agent Assist.
Conversations
A Conversation
is a resource that represents an exchange between two or more
participants. Every time a
conversation participant's utterance is passed to Agent Assist as part
of a Conversation resource, a message is created and stored for the
conversation.
Conversation profiles
Each conversation is created using a conversation profile. Each
conversation profile contains a set of parameters that control the
suggestions provided to the human agent during a conversation. You can also configure a conversation profile for use with a virtual agent, which autogenerates a reply
to the end user.
Dialogflow agents are virtual agents that you can build and configure for
conversations with your end-users. You can design your system so that end-users
initially interact with a
virtual agent before being escalated to a
human agent. You can use either a
Dialogflow ES or
Dialogflow CX agent.
Conversation stages
A conversation with an end-user has multiple stages:
Connection stage:
The end-user initiates a chat dialog.
Virtual agent stage:
The end-user is connected to a
Dialogflow virtual agent.
The virtual agent communicates with the end-user and attempts to resolve all
issues and requests. You can set up your system to bypass the virtual agent
and handoff stages.
Handoff stage:
If the virtual agent cannot resolve all end-user issues, the conversation is
handed off
to a human agent.
Agent Assist stage:
The end-user is connected to one or more human agents.
Agent Assist
supplies real-time document and response suggestions to the human agent that
are relevant to the conversation.
Termination stage:
The end-user is disconnected from the text chat.
Agent desktop
To implement Agent Assist for text-based conversations you must
integrate the Agent Assist API into your agent desktop.
[[["Easy to understand","easyToUnderstand","thumb-up"],["Solved my problem","solvedMyProblem","thumb-up"],["Other","otherUp","thumb-up"]],[["Hard to understand","hardToUnderstand","thumb-down"],["Incorrect information or sample code","incorrectInformationOrSampleCode","thumb-down"],["Missing the information/samples I need","missingTheInformationSamplesINeed","thumb-down"],["Other","otherDown","thumb-down"]],["Last updated 2025-08-25 UTC."],[[["\u003cp\u003eAgent Assist facilitates real-time suggestions for human agents during conversations, utilizing resources like Conversations, Conversation Profiles, and Participants.\u003c/p\u003e\n"],["\u003cp\u003eA Conversation represents an exchange between participants, with each participant's utterance creating a stored message, and each conversation is configured using a Conversation Profile, which defines suggestion parameters.\u003c/p\u003e\n"],["\u003cp\u003eConversations involve participants like \u003ccode\u003eEND_USER\u003c/code\u003e, \u003ccode\u003eAUTOMATED_AGENT\u003c/code\u003e, and \u003ccode\u003eHUMAN_AGENT\u003c/code\u003e, and can involve Dialogflow virtual agents.\u003c/p\u003e\n"],["\u003cp\u003eA typical conversation progresses through stages, including connection, virtual agent interaction, potential handoff to a human, Agent Assist support, and finally, termination.\u003c/p\u003e\n"],["\u003cp\u003eImplementing Agent Assist for text-based interactions requires integration of the Agent Assist API into the agent's desktop.\u003c/p\u003e\n"]]],[],null,["# Agent Assist basics\n\nThis document is a basic guide to Agent Assist resources. More details\nabout the resources described here can be found in the\n[reference documentation](/agent-assist/docs/reference-index).\nWe recommend that you familiarize yourself with this guide before diving into\nAgent Assist.\n\nConversations\n-------------\n\nA [Conversation](/dialogflow/es/docs/reference/rest/v2/projects.conversations)\nis a resource that represents an exchange between two or more\n[participants](/agent-assist/docs/basics#participants). Every time a\nconversation participant's utterance is passed to Agent Assist as part\nof a Conversation resource, a **message** is created and stored for the\nconversation.\n\nConversation profiles\n---------------------\n\nEach conversation is created using a **conversation profile** . Each\nconversation profile contains a set of parameters that control the\nsuggestions provided to the human agent during a conversation. You can also configure a conversation profile for use with a [virtual agent](/dialogflow/cx/docs), which autogenerates a reply\nto the end user.\n\nYou can create a conversation profile either by using the\n[Agent Assist Console](/agent-assist/docs/conversation-profile)\nor by [calling the API directly](/agent-assist/docs/how-to).\n\nParticipants\n------------\n\nEach conversation can have the following\n[participants](/dialogflow/es/docs/reference/rest/v2/projects.conversations.participants):\n\n- `END_USER`: Participant is an end-user. This is the person the agent is conversing with.\n- `AUTOMATED_AGENT`: Participant is an [automated agent](/dialogflow/cx/docs).\n- `HUMAN_AGENT`: Participant is a human agent.\n\nVirtual agents\n--------------\n\nDialogflow agents are virtual agents that you can build and configure for\nconversations with your end-users. You can design your system so that end-users\ninitially interact with a\n[virtual agent](/dialogflow/cx/docs) before being escalated to a\nhuman agent. You can use either a\n[Dialogflow ES](/dialogflow/es/docs/quick/build-agent) or\n[Dialogflow CX](/dialogflow/cx/docs/quick/build-agent) agent.\n\nConversation stages\n-------------------\n\nA conversation with an end-user has multiple stages:\n\n1. **Connection stage**: The end-user initiates a chat dialog.\n2. **Virtual agent stage** : The end-user is connected to a [Dialogflow virtual agent](/dialogflow/cx/docs). The virtual agent communicates with the end-user and attempts to resolve all issues and requests. You can set up your system to bypass the virtual agent and handoff stages.\n3. **Handoff stage** : If the virtual agent cannot resolve all end-user issues, the conversation is [handed off](/agent-assist/docs/handoff) to a human agent.\n4. **Agent Assist stage** : The end-user is connected to one or more human agents. [Agent Assist](/agent-assist/docs) supplies real-time document and response suggestions to the human agent that are relevant to the conversation.\n5. **Termination stage**: The end-user is disconnected from the text chat.\n\nAgent desktop\n-------------\n\nTo implement Agent Assist for text-based conversations you must\nintegrate the Agent Assist API into your agent desktop."]]
