Conversations Tutorial

This tutorial demonstrates how to have multi-turn conversations with Knowledge Discovery agents. Unlike single-shot questions, conversations maintain context across multiple questions and answers, allowing for more natural dialogue.

Prerequisites

Valid access token (see Authentication)
At least one configured agent with data sources
Discovery User role or higher

warning

If your HTTP client doesn't add the User-Agent header by default, make sure to include it in all requests below (e.g., User-Agent: PostmanRuntime/7.28.4) to avoid receiving a 403 Forbidden response.

Tutorial Overview

In this tutorial, you'll learn to:

Start a new conversation with an agent
Ask follow-up questions within a conversation
Use streaming endpoints to receive incremental responses via SSE
List and retrieve conversation messages
Provide feedback on conversation messages
Manage conversations (list, update metadata)

Response time

Conversation API calls are synchronous and may take up to 480 seconds for complex GraphRAG queries. Ensure your HTTP client is configured with a timeout of at least 480 seconds to avoid premature disconnects.

Part 1: Start a Conversation

To start a new conversation, send the first question using the conversation endpoint. The response returns both the new conversationId and the first message (question and answer).

HTTP
curl

POST {{baseUrl}}/qna/agents/{{agentId}}/conversations
Authorization: Bearer {{accessToken}}
Hxp-Environment: {{environmentKey}}
Hxp-App: {{appKey}}
Content-Type: application/json

{
  "question": "What is the status of the project?",
  "contextObjectIds": null,
  "dynamicFilter": null
}

AGENT_ID="project-agent-id"

curl -X POST "${DISCOVERY_API_BASE_URL}/qna/agents/${AGENT_ID}/conversations" \
  -H "Authorization: Bearer ${DISCOVERY_ACCESS_TOKEN}" \
  -H "Hxp-Environment: ${ENVIRONMENT_KEY}" \
  -H "Hxp-App: ${APP_KEY}" \
  -H "Content-Type: application/json" \
  -d '{
    "question": "What is the status of the project?",
    "contextObjectIds": null,
    "dynamicFilter": null
  }'

Start Conversation Request Parameters

Field	Required	Description
`question`	Yes	Initial natural language question for the conversation
`dynamicFilter`	No	Runtime filter expression to narrow search results
`enableCitations`	No	When `true`, the agent includes inline citation references in the answer text. Defaults to `false` when not provided.

Response Structure

{
  "conversation": {
    "id": "a1b2c3d4-e5f6-7890-abcd-ef1234567890",
    "lastModified": "2025-12-02T10:30:00Z"
  },
  "message": {
    "id": "f1e2d3c4-b5a6-7890-1234-567890abcdef",
    "answer": "Based on the documents, the project is currently in progress.",
    "documentReferences": [
      {
        "documentId": "doc001",
        "references": [
          {
            "referenceId": "ref001",
            "rankScore": 0.95,
            "rank": 1
          }
        ]
      }
    ],
    "question": "What is the status of the project?",
    "feedback": null,
    "staticFilter": null,
    "dynamicFilter": null,
    "dateCreated": "2025-12-02T10:30:00Z",
    "dateAnswered": "2025-12-02T10:30:05Z",
    "agentVersion": 1,
    "status": "Answered"
  }
}

Graph RAG vs Standard RAG Responses

The response structure shown above applies to standard RAG agents. If you're using a Graph RAG agent (see Managing Agents Tutorial - Graph RAG), the response will include graphReferences instead of (or in addition to) documentReferences. Graph references provide entity and relationship information from the knowledge graph domain.

Understanding the Response

Field	Type	Description
`conversation.id`	UUID	Unique identifier for the conversation
`conversation.lastModified`	datetime	Timestamp of the last update
`message.id`	UUID	Unique identifier for this message
`message.question`	string	The question text
`message.answer`	string	The generated answer
`message.documentReferences`	array	Source documents and chunks used for the answer
`message.dateCreated`	datetime	When the question was submitted
`message.dateAnswered`	datetime	When the answer was generated
`message.agentVersion`	integer	Version of the agent used
`message.status`	enum	Message status: `Unspecified`, `Submitted`, `Answered`, `Error`, `Blocked`

Starting a Conversation with Dynamic Filter

You can apply dynamic filters when starting a conversation to narrow the search scope for all messages in that conversation. This is useful when you want to focus the conversation on specific documents, date ranges, or categories.

HTTP
curl

POST {{baseUrl}}/qna/agents/{{agentId}}/conversations
Authorization: Bearer {{accessToken}}
Hxp-Environment: {{environmentKey}}
Hxp-App: {{appKey}}
Content-Type: application/json

{
  "question": "What are the latest security policies?",
  "dynamicFilter": {
    "logicalOperator": "And",
    "conditions": [
      {
        "field": "DocumentType",
        "type": "Text",
        "operator": "Equals",
        "value": "Policy"
      },
      {
        "field": "Department",
        "type": "Text",
        "operator": "Equals",
        "value": "Security"
      },
      {
        "field": "LastModified",
        "type": "Date",
        "operator": "GreaterThan",
        "value": "2024-01-01"
      }
    ]
  }
}

curl -X POST "${DISCOVERY_API_BASE_URL}/qna/agents/${AGENT_ID}/conversations" \
  -H "Authorization: Bearer ${DISCOVERY_ACCESS_TOKEN}" \
  -H "Hxp-Environment: ${ENVIRONMENT_KEY}" \
  -H "Hxp-App: ${APP_KEY}" \
  -H "Content-Type: application/json" \
  -d '{
    "question": "What are the latest security policies?",
    "dynamicFilter": {
      "logicalOperator": "And",
      "conditions": [
        {
          "field": "DocumentType",
          "type": "Text",
          "operator": "Equals",
          "value": "Policy"
        },
        {
          "field": "Department",
          "type": "Text",
          "operator": "Equals",
          "value": "Security"
        },
        {
          "field": "LastModified",
          "type": "Date",
          "operator": "GreaterThan",
          "value": "2024-01-01"
        }
      ]
    }
  }'

tip

Dynamic filters can also be provided when asking follow-up questions (see Part 2). Each message can have its own filter, or you can omit it to use the agent's default behavior. For more details on filter syntax, see the Single-Shot Questions Tutorial.

Dynamic Filter Template Matching

The dynamicFilter you provide at runtime must match the structure defined in the agent's dynamicFilterTemplate, which is configured during agent creation or update. The template defines which fields can be filtered and their types, while the runtime filter provides the actual values.

Example: Agent Configuration with Dynamic Filter Template

When creating or updating an agent, you define which fields users can filter by:

{
  "name": "Document Search Agent",
  "description": "Agent that allows users to filter by department, document type, and date",
  "modelName": "amazon.nova-micro-v1:0",
  "sourceIds": ["your-source-id"],
  "dynamicFilterTemplate": {
    "logicalOperator": "And",
    "conditions": [
      {
        "field": "Department",
        "type": "Text"
      },
      {
        "field": "DocumentType",
        "type": "Text"
      },
      {
        "field": "LastModified",
        "type": "Date"
      }
    ]
  }
}

Corresponding Runtime Dynamic Filter

When starting a conversation or asking questions, provide the operator and value for the fields defined in the template:

{
  "dynamicFilter": {
    "logicalOperator": "And",
    "conditions": [
      {
        "field": "Department",
        "type": "Text",
        "operator": "Equals",
        "value": "Security"
      },
      {
        "field": "DocumentType",
        "type": "Text",
        "operator": "Equals",
        "value": "Policy"
      },
      {
        "field": "LastModified",
        "type": "Date",
        "operator": "GreaterThan",
        "value": "2024-01-01"
      }
    ]
  }
}

note

The fields in your runtime dynamicFilter must match those defined in the agent's dynamicFilterTemplate. You can provide values for all or a subset of the template fields. For more information on configuring agents with dynamic filter templates, see the Managing Agents Tutorial.

Starting a Conversation with Citations

Set enableCitations to true to receive inline citation markers in the answer text, indicating which source documents support each part of the response.

HTTP
curl

POST {{baseUrl}}/qna/agents/{{agentId}}/conversations
Authorization: Bearer {{accessToken}}
Hxp-Environment: {{environmentKey}}
Hxp-App: {{appKey}}
Content-Type: application/json

{
  "question": "What is the company's leave policy?",
  "enableCitations": true
}

curl -X POST "${DISCOVERY_API_BASE_URL}/qna/agents/${AGENT_ID}/conversations" \
  -H "Authorization: Bearer ${DISCOVERY_ACCESS_TOKEN}" \
  -H "Hxp-Environment: ${ENVIRONMENT_KEY}" \
  -H "Hxp-App: ${APP_KEY}" \
  -H "Content-Type: application/json" \
  -d '{
    "question": "What is the company'\''s leave policy?",
    "enableCitations": true
  }'

enableCitations can also be combined with dynamicFilter and contextObjectIds in the same request. When omitted or set to false, citation markers are not included in the answer.

Part 2: Ask Follow-Up Questions

Once you have a conversationId, you can continue the conversation by sending follow-up questions. The agent will use previous messages in the conversation as context.

HTTP
curl

POST {{baseUrl}}/qna/agents/{{agentId}}/conversations/{{conversationId}}/messages
Authorization: Bearer {{accessToken}}
Hxp-Environment: {{environmentKey}}
Hxp-App: {{appKey}}
Content-Type: application/json

{
  "question": "Can you provide more details about the timeline?",
  "contextObjectIds": null,
  "dynamicFilter": null
}

AGENT_ID="project-agent-id"
CONVERSATION_ID="a1b2c3d4-e5f6-7890-abcd-ef1234567890"

curl -X POST "${DISCOVERY_API_BASE_URL}/qna/agents/${AGENT_ID}/conversations/${CONVERSATION_ID}/messages" \
  -H "Authorization: Bearer ${DISCOVERY_ACCESS_TOKEN}" \
  -H "Hxp-Environment: ${ENVIRONMENT_KEY}" \
  -H "Hxp-App: ${APP_KEY}" \
  -H "Content-Type: application/json" \
  -d '{
    "question": "Can you provide more details about the timeline?",
    "contextObjectIds": null,
    "dynamicFilter": null
  }'

Response

The response contains a ConversationMessageDto with the new question and answer, similar to the initial message:

{
  "id": "b2c3d4e5-f6a7-8901-2345-678901abcdef",
  "answer": "The latest update shows the project timeline is on track with an expected completion date of Q2 2026.",
  "documentReferences": [
    {
      "documentId": "doc002",
      "references": [
        {
          "referenceId": "ref002",
          "rankScore": 0.92,
          "rank": 1
        }
      ]
    }
  ],
  "question": "Can you provide more details about the timeline?",
  "feedback": null,
  "staticFilter": null,
  "dynamicFilter": null,
  "dateCreated": "2025-12-02T10:35:00Z",
  "dateAnswered": "2025-12-02T10:35:03Z",
  "agentVersion": 1,
  "status": "Answered"
}

Context Awareness

The agent understands that "the timeline" refers to the project mentioned in the first question, demonstrating context retention across the conversation.

note

Follow-up messages also accept enableCitations (optional boolean). Set it to true on individual messages to request citations for that specific response, independently of whether citations were enabled for the conversation's first message.

Part 3: Streaming Conversations

Instead of waiting for a complete answer, you can observe the agents progress as a Server-Sent Events (SSE) stream. This reduces perceived latency and allows you to follow the models reasoning.

Start a Conversation with Streaming

HTTP
curl

POST {{baseUrl}}/qna/agents/{{agentId}}/conversations/stream
Authorization: Bearer {{accessToken}}
Hxp-Environment: {{environmentKey}}
Hxp-App: {{appKey}}
Content-Type: application/json
Accept: text/event-stream

{
  "question": "What is the status of the project?"
}

AGENT_ID="project-agent-id"

curl -X POST "${DISCOVERY_API_BASE_URL}/qna/agents/${AGENT_ID}/conversations/stream" \
  -H "Authorization: Bearer ${DISCOVERY_ACCESS_TOKEN}" \
  -H "Hxp-Environment: ${ENVIRONMENT_KEY}" \
  -H "Hxp-App: ${APP_KEY}" \
  -H "Content-Type: application/json" \
  -H "Accept: text/event-stream" \
  --no-buffer \
  -d '{"question": "What is the status of the project?"}'

Continue a Conversation with Streaming

HTTP
curl

POST {{baseUrl}}/qna/agents/{{agentId}}/conversations/{{conversationId}}/messages/stream
Authorization: Bearer {{accessToken}}
Hxp-Environment: {{environmentKey}}
Hxp-App: {{appKey}}
Content-Type: application/json
Accept: text/event-stream

{
  "question": "When is the new deadline?"
}

AGENT_ID="project-agent-id"
CONVERSATION_ID="a1b2c3d4-e5f6-7890-abcd-ef1234567890"

curl -X POST "${DISCOVERY_API_BASE_URL}/qna/agents/${AGENT_ID}/conversations/${CONVERSATION_ID}/messages/stream" \
  -H "Authorization: Bearer ${DISCOVERY_ACCESS_TOKEN}" \
  -H "Hxp-Environment: ${ENVIRONMENT_KEY}" \
  -H "Hxp-App: ${APP_KEY}" \
  -H "Content-Type: application/json" \
  -H "Accept: text/event-stream" \
  --no-buffer \
  -d '{"question": "When is the new deadline?"}'

SSE Event Reference

Each line of the stream is an SSE event with the format data: <json>\n\n. The event field in the JSON payload identifies the event type:

Event type	Description
`stream.started`	Emitted once at the start of the stream. Contains `conversationId` and `messageId`.
`stream.delta`	Consecutive steps of Chain of Thought. This can be either Model reasoning or tool call results.
`stream.tool_call`	Emitted when the agent invokes an internal tool (e.g., search). Contains `name` and optional `arguments`.
`stream.completed`	Final event when the full answer is ready. Contains the complete `message` object and, for new conversations, a `conversation` object.
`response.incomplete`	Emitted instead of `stream.completed` when the answer is blocked by a guardrail (content filter). Contains the `message` object with `status: "Blocked"` and a `null` answer, and, for new conversations, a `conversation` object.
`stream.error`	Emitted if an error occurs. Contains `error` (description) and optional `code`.

Example stream output:

data: {"event":"stream.started","conversationId":"a1b2c3d4-e5f6-7890-abcd-ef1234567890","messageId":"f1e2d3c4-b5a6-7890-1234-567890abcdef"}

data: {"name":"search_vector_db","arguments":"{\u0022natural_language_question\u0022: \u0022What are the procedures and steps to request or obtain PTO (Paid Time Off)?\u0022}","event":"stream.tool_call"}

data: {"delta":"{\n  \u0022results\u0022: [],\n  \u0022_total_results\u0022: 0\n}","event":"stream.delta"}

data: {"event":"stream.completed","message":{"id":"f1e2d3c4-b5a6-7890-1234-567890abcdef","answer":"Based on the documents, the project is currently in progress.","documentReferences":[...],"question":"What is the status of the project?","status":"Answered",...},"conversation":{"id":"a1b2c3d4-e5f6-7890-abcd-ef1234567890","lastModified":"2025-12-02T10:30:05Z"}}

tip

Use the stream.completed event to retrieve the full structured response including documentReferences. The conversation field in stream.completed is only populated for the first message (start conversation); it is null for follow-up messages.

If the agent's guardrail blocks the response, you will receive response.incomplete instead of stream.completed. In that case, message.status is "Blocked" and message.answer is null.

Part 4: List Conversation Messages

You can retrieve the history of messages within a conversation using cursor-based pagination.

HTTP
curl

GET {{baseUrl}}/qna/agents/{{agentId}}/conversations/{{conversationId}}/messages?pageSize=50&maxContentLength=500
Authorization: Bearer {{accessToken}}
Hxp-Environment: {{environmentKey}}
Hxp-App: {{appKey}}

AGENT_ID="project-agent-id"
CONVERSATION_ID="a1b2c3d4-e5f6-7890-abcd-ef1234567890"

curl -X GET "${DISCOVERY_API_BASE_URL}/qna/agents/${AGENT_ID}/conversations/${CONVERSATION_ID}/messages?pageSize=50&maxContentLength=500" \
  -H "Authorization: Bearer ${DISCOVERY_ACCESS_TOKEN}" \
  -H "Hxp-Environment: ${ENVIRONMENT_KEY}" \
  -H "Hxp-App: ${APP_KEY}"

Query Parameters

Parameter	Required	Default	Description
`cursor`	No	null	Cursor token for fetching the next page
`pageSize`	No	100	Number of items per page (1-100)
`maxContentLength`	No	null	Maximum length of question/answer content (10-1000)
`fields`	No	null	Comma-separated list of fields to include in response

Response Structure

{
  "data": [
    {
      "id": "f1e2d3c4-b5a6-7890-1234-567890abcdef",
      "answer": "Based on the documents, the project is currently in progress.",
      "documentReferences": [
        {
          "documentId": "doc001",
          "references": [
            {
              "referenceId": "ref001",
              "rankScore": 0.95,
              "rank": 1
            }
          ]
        }
      ],
      "question": "What is the status of the project?",
      "feedback": "Good",
      "staticFilter": null,
      "dynamicFilter": null,
      "dateCreated": "2025-12-02T10:30:00Z",
      "dateAnswered": "2025-12-02T10:30:05Z",
      "agentVersion": 1,
      "status": "Answered"
    },
    {
      "id": "b2c3d4e5-f6a7-8901-2345-678901abcdef",
      "answer": "The latest update shows the project timeline is on track...",
      "documentReferences": [
        {
          "documentId": "doc002",
          "references": [
            {
              "referenceId": "ref002",
              "rankScore": 0.92,
              "rank": 1
            }
          ]
        }
      ],
      "question": "Can you provide more details about the timeline?",
      "feedback": null,
      "staticFilter": null,
      "dynamicFilter": null,
      "dateCreated": "2025-12-02T10:35:00Z",
      "dateAnswered": "2025-12-02T10:35:03Z",
      "agentVersion": 1,
      "status": "Answered"
    }
  ],
  "pagination": {
    "nextCursor": null,
    "hasMore": false
  }
}

Understanding Cursor Pagination

Unlike page-based pagination (used in single-shot question history), conversation messages use cursor-based pagination:

Field	Type	Description
`pagination.nextCursor`	string (nullable)	Token to fetch the next page; null if no more pages
`pagination.hasMore`	boolean	Indicates whether more messages are available

To fetch the next page, pass the nextCursor value as the cursor query parameter.

Part 5: Get a Single Message

To retrieve a specific message (for example, when a user opens a detailed view), use the message ID.

HTTP
curl

GET {{baseUrl}}/qna/agents/{{agentId}}/conversations/{{conversationId}}/messages/{{messageId}}
Authorization: Bearer {{accessToken}}
Hxp-Environment: {{environmentKey}}
Hxp-App: {{appKey}}

AGENT_ID="project-agent-id"
CONVERSATION_ID="a1b2c3d4-e5f6-7890-abcd-ef1234567890"
MESSAGE_ID="f1e2d3c4-b5a6-7890-1234-567890abcdef"

curl -X GET "${DISCOVERY_API_BASE_URL}/qna/agents/${AGENT_ID}/conversations/${CONVERSATION_ID}/messages/${MESSAGE_ID}" \
  -H "Authorization: Bearer ${DISCOVERY_ACCESS_TOKEN}" \
  -H "Hxp-Environment: ${ENVIRONMENT_KEY}" \
  -H "Hxp-App: ${APP_KEY}"

The response is a single ConversationMessageDto object with full details about the message.

Part 6: Provide Feedback on Messages

Feedback on conversation answers is recorded per message, similar to single-shot questions.

HTTP
curl

POST {{baseUrl}}/qna/agents/{{agentId}}/conversations/{{conversationId}}/messages/{{messageId}}/feedback
Authorization: Bearer {{accessToken}}
Hxp-Environment: {{environmentKey}}
Hxp-App: {{appKey}}
Content-Type: application/json

{
  "feedback": "Good"
}

AGENT_ID="project-agent-id"
CONVERSATION_ID="a1b2c3d4-e5f6-7890-abcd-ef1234567890"
MESSAGE_ID="f1e2d3c4-b5a6-7890-1234-567890abcdef"

curl -X POST "${DISCOVERY_API_BASE_URL}/qna/agents/${AGENT_ID}/conversations/${CONVERSATION_ID}/messages/${MESSAGE_ID}/feedback" \
  -H "Authorization: Bearer ${DISCOVERY_ACCESS_TOKEN}" \
  -H "Hxp-Environment: ${ENVIRONMENT_KEY}" \
  -H "Hxp-App: ${APP_KEY}" \
  -H "Content-Type: application/json" \
  -d '{"feedback": "Good"}'

`feedback` (enum)

Value	Meaning
`Good`	Answer was helpful and accurate
`Bad`	Answer was not helpful or contained errors
`Retry`	Request to regenerate the answer

Part 7: Manage Conversations

List All Conversations for an Agent

Retrieve all conversations associated with an agent.

HTTP
curl

GET {{baseUrl}}/qna/agents/{{agentId}}/conversations?pageSize=20
Authorization: Bearer {{accessToken}}
Hxp-Environment: {{environmentKey}}
Hxp-App: {{appKey}}

AGENT_ID="project-agent-id"

curl -X GET "${DISCOVERY_API_BASE_URL}/qna/agents/${AGENT_ID}/conversations?pageSize=20" \
  -H "Authorization: Bearer ${DISCOVERY_ACCESS_TOKEN}" \
  -H "Hxp-Environment: ${ENVIRONMENT_KEY}" \
  -H "Hxp-App: ${APP_KEY}"

Query Parameters

Parameter	Required	Default	Description
`cursor`	No	null	Cursor token for fetching the next page
`pageSize`	No	20	Number of items per page (1-100)

Response

{
  "data": [
    {
      "id": "a1b2c3d4-e5f6-7890-abcd-ef1234567890",
      "name": "Project Status Discussion",
      "description": "Conversation about current project status",
      "lastModified": "2025-12-02T10:35:00Z"
    },
    {
      "id": "b2c3d4e5-f6a7-8901-bcde-f12345678901",
      "name": null,
      "description": null,
      "lastModified": "2025-12-01T15:20:00Z"
    }
  ],
  "pagination": {
    "nextCursor": "eyJsYXN0SWQiOiJiMmMzZDRlNS1mNmE3LTg5MDEtYmNkZS1mMTIzNDU2Nzg5MDEiLCJsYXN0TW9kaWZpZWQiOiIyMDI1LTEyLTAxVDE1OjIwOjAwWiJ9",
    "hasMore": true
  }
}

Get Conversation Details

Retrieve metadata for a specific conversation.

HTTP
curl

GET {{baseUrl}}/qna/agents/{{agentId}}/conversations/{{conversationId}}
Authorization: Bearer {{accessToken}}
Hxp-Environment: {{environmentKey}}
Hxp-App: {{appKey}}

AGENT_ID="project-agent-id"
CONVERSATION_ID="a1b2c3d4-e5f6-7890-abcd-ef1234567890"

curl -X GET "${DISCOVERY_API_BASE_URL}/qna/agents/${AGENT_ID}/conversations/${CONVERSATION_ID}" \
  -H "Authorization: Bearer ${DISCOVERY_ACCESS_TOKEN}" \
  -H "Hxp-Environment: ${ENVIRONMENT_KEY}" \
  -H "Hxp-App: ${APP_KEY}"

Update Conversation Metadata

Update the name and description of a conversation for better organization.

HTTP
curl

PUT {{baseUrl}}/qna/agents/{{agentId}}/conversations/{{conversationId}}
Authorization: Bearer {{accessToken}}
Hxp-Environment: {{environmentKey}}
Hxp-App: {{appKey}}
Content-Type: application/json

{
  "name": "Project Status Q2 2025",
  "description": "Discussion about project timeline and deliverables for Q2"
}

AGENT_ID="project-agent-id"
CONVERSATION_ID="a1b2c3d4-e5f6-7890-abcd-ef1234567890"

curl -X PUT "${DISCOVERY_API_BASE_URL}/qna/agents/${AGENT_ID}/conversations/${CONVERSATION_ID}" \
  -H "Authorization: Bearer ${DISCOVERY_ACCESS_TOKEN}" \
  -H "Hxp-Environment: ${ENVIRONMENT_KEY}" \
  -H "Hxp-App: ${APP_KEY}" \
  -H "Content-Type: application/json" \
  -d '{
    "name": "Project Status Q2 2025",
    "description": "Discussion about project timeline and deliverables for Q2"
  }'

Update Request Parameters

Field	Required	Description
`name`	Yes	Conversation name (for display/organization)
`description`	No	Optional longer description of the conversation purpose

Summary

In this tutorial, you learned how to:

✅ Start multi-turn conversations with agents
✅ Ask follow-up questions that use previous context
✅ Receive incremental responses using SSE streaming endpoints
✅ List and retrieve conversation messages
✅ Provide feedback on conversation messages
✅ Manage conversation metadata (list, view, update)

Next Steps

Learn about Single-Shot Questions for one-off queries
Review the API Reference for complete endpoint documentation

Prerequisites​

Tutorial Overview​

Part 1: Start a Conversation​

Start Conversation Request Parameters​

Response Structure​

Understanding the Response​

Starting a Conversation with Dynamic Filter​

Dynamic Filter Template Matching​

Starting a Conversation with Citations​

Part 2: Ask Follow-Up Questions​

Response​

Part 3: Streaming Conversations​

Start a Conversation with Streaming​

Continue a Conversation with Streaming​

SSE Event Reference​

Part 4: List Conversation Messages​

Query Parameters​

Response Structure​

Understanding Cursor Pagination​

Part 5: Get a Single Message​

Part 6: Provide Feedback on Messages​

feedback (enum)​

Part 7: Manage Conversations​

List All Conversations for an Agent​

Query Parameters​

Response​

Get Conversation Details​

Update Conversation Metadata​

Update Request Parameters​

Summary​

Next Steps​

Prerequisites

Tutorial Overview

Part 1: Start a Conversation

Start Conversation Request Parameters

Response Structure

Understanding the Response

Starting a Conversation with Dynamic Filter

Dynamic Filter Template Matching

Starting a Conversation with Citations

Part 2: Ask Follow-Up Questions

Response

Part 3: Streaming Conversations

Start a Conversation with Streaming

Continue a Conversation with Streaming

SSE Event Reference

Part 4: List Conversation Messages

Query Parameters

Response Structure

Understanding Cursor Pagination

Part 5: Get a Single Message

Part 6: Provide Feedback on Messages

`feedback` (enum)

Part 7: Manage Conversations

List All Conversations for an Agent

Query Parameters

Response

Get Conversation Details

Update Conversation Metadata

Update Request Parameters

Summary

Next Steps