Save & Contribute Insights

Insights are a durable knowledge artifacts that was captured from a conversation.

Insight are durable knowledge artifacts captured from conversations and related sources. Each insight (also referred to as a note) contains summarized understanding, provenance, and links to underlying messages and document references.

Create an Insight from a Message

Often, outputs of Interact with Bot are highly valuable as input for further conversations. Saving these pieces of knowledge as insights in your system makes them easily reusable.

Here we need to provide a specific messageId of the response of the LLM that we want so store. This messageId is being sent in the Interact with Bot step.

Generate Chat Note

post

Authorizations

HTTPBearer

AuthorizationstringRequired

Bearer authentication header of the form Bearer <token>.

Query parameters

convo_idstringRequired

latest_message_idstringRequired

insightsGraphTypeany ofOptional

string · enumOptionalPossible values:

or

nullOptional

is_pinany ofOptional

booleanOptional

or

nullOptional

parent_pathstringOptionalDefault: root

is_include_chat_historyany ofOptionalDefault: true

booleanOptional

or

nullOptional

Responses

200

Successful Response

application/json

422

Validation Error

application/json

post

/cortex/notes/generate

HTTPcURLJavaScriptPython

POST /cortex/notes/generate?convo_id=text&latest_message_id=text HTTP/1.1
Host: 
Authorization: Bearer YOUR_SECRET_TOKEN
Accept: */*

{
  "_id": "text",
  "createdAt": "2025-12-09T10:42:34.307Z",
  "modifiedAt": "2025-12-09T10:42:34.307Z",
  "summary": "text",
  "userId": "text",
  "fromUserId": "text",
  "latestMessageCreated": "2025-12-09T10:42:34.307Z",
  "isEdited": false,
  "messages": [
    {
      "_id": "text",
      "createdAt": "2025-12-09T10:42:34.307Z",
      "modifiedAt": "2025-12-09T10:42:34.307Z",
      "botId": "text",
      "userId": "text",
      "convoId": "text",
      "content": "text",
      "role": "text",
      "model": "gpt-3.5-turbo",
      "messageType": "assistant-greeting",
      "status": "activate",
      "langfuseUrl": "",
      "currentContent": "text",
      "uploadedFiles": [
        {
          "_id": "text",
          "createdAt": "2025-12-09T10:42:34.307Z",
          "modifiedAt": "2025-12-09T10:42:34.307Z",
          "name": "text",
          "key": "text",
          "tokens": 0,
          "enabled": true,
          "errorMessage": "text",
          "status": "SUCCESS",
          "calculatedStatus": "SUCCESS",
          "convoId": "",
          "fileType": "TEXT",
          "fileSize": 0,
          "imageDescription": "",
          "imageWidth": 1,
          "imageHeight": 1,
          "userId": "text",
          "imageResizeKey": "text",
          "imageThumbKey": "text",
          "zipAttachmentId": "text"
        }
      ],
      "isSaved": false,
      "noteId": "text",
      "insightsGraphType": "STATUS_QUO",
      "doc_ref": [
        {
          "doc_id": "text",
          "doc_name": "text",
          "text_id": [
            {
              "content": "text",
              "text_id": "text",
              "score": 0,
              "isRelatedChunk": false,
              "doc_type": "vector-db"
            }
          ],
          "originalUrl": "text",
          "score": 0,
          "doc_type": "vector-db",
          "isTemplate": false,
          "imageResizeKey": "text",
          "imageThumbKey": "text"
        }
      ],
      "webSearchQuery": "text",
      "webRef": [
        {
          "query": "text",
          "answer": [
            {
              "id": "text",
              "url": "text",
              "content": "text",
              "title": "text"
            }
          ]
        }
      ],
      "webProRef": [
        {
          "question": "text",
          "search_queries": [
            {
              "query": "text",
              "answer": [
                {
                  "id": "text",
                  "url": "text",
                  "content": "text",
                  "title": "text"
                }
              ]
            }
          ]
        }
      ],
      "isWebProR": false,
      "webSearchConfig": {
        "webSearchProvider": "linkup_normal_web_search",
        "isLinkupDeepSearch": false
      },
      "standaloneQuestion": "text",
      "relatedChunkIds": [
        "text"
      ],
      "targetText": "text",
      "targetMessageId": "text",
      "targetFullContent": "text",
      "messageReplies": [
        "text"
      ],
      "isUnliked": true,
      "feedbackAt": "2025-12-09T10:42:34.307Z",
      "feedbackContent": "text",
      "selectedWorkflowId": "",
      "workflow": {
        "runTaskId": "",
        "workflowId": "",
        "workflowName": "",
        "stepId": "",
        "stepName": ""
      },
      "selectedAgentTask": {
        "_id": "text",
        "createdAt": "2025-12-09T10:42:34.307Z",
        "modifiedAt": "2025-12-09T10:42:34.307Z",
        "category_id": "text",
        "name": "text",
        "type": "custom",
        "description": "text",
        "category": {
          "_id": "text",
          "createdAt": "2025-12-09T10:42:34.307Z",
          "modifiedAt": "2025-12-09T10:42:34.307Z",
          "icon": "text",
          "name": "text",
          "owner": "text",
          "count": 1,
          "isPublic": false,
          "isTemplate": false
        },
        "isFavorite": false,
        "isSelected": false,
        "isActive": false,
        "categoryName": "text"
      },
      "selectedStrategy": "",
      "selectedAction": "",
      "strategies": [
        {
          "name": "text",
          "slug": "",
          "description": "",
          "icon": "",
          "actions": [
            {
              "name": "text",
              "slug": "",
              "prompt": "",
              "description": "",
              "icon": "",
              "isDefault": true
            }
          ],
          "scopes": [
            "scope-user"
          ],
          "isDefault": true
        }
      ],
      "targetSelectedStrategy": "",
      "targetSelectedAction": "",
      "targetAgentTask": {
        "_id": "text",
        "createdAt": "2025-12-09T10:42:34.307Z",
        "modifiedAt": "2025-12-09T10:42:34.307Z",
        "category_id": "text",
        "name": "text",
        "type": "custom",
        "description": "text",
        "category": {
          "_id": "text",
          "createdAt": "2025-12-09T10:42:34.307Z",
          "modifiedAt": "2025-12-09T10:42:34.307Z",
          "icon": "text",
          "name": "text",
          "owner": "text",
          "count": 1,
          "isPublic": false,
          "isTemplate": false
        },
        "isFavorite": false,
        "isSelected": false,
        "isActive": false,
        "categoryName": "text"
      },
      "promptMessageId": "",
      "messageQueryId": "text",
      "mediaUrls": [
        {
          "contentType": "text",
          "url": "text"
        }
      ]
    }
  ],
  "title": "New Note",
  "tokenCount": 0,
  "insightsGraphType": "STATUS_QUO",
  "model": "gpt-3.5-turbo",
  "isContributed": false,
  "vectorData": [
    1
  ],
  "messageId": "",
  "doc_ref": [
    {
      "doc_id": "text",
      "doc_name": "text",
      "text_id": [
        {
          "content": "text",
          "text_id": "text",
          "score": 0,
          "isRelatedChunk": false,
          "doc_type": "vector-db"
        }
      ],
      "originalUrl": "text",
      "score": 0,
      "doc_type": "vector-db",
      "isTemplate": false,
      "imageResizeKey": "text",
      "imageThumbKey": "text"
    }
  ],
  "web_ref": [
    {
      "query": "text",
      "answer": [
        {
          "id": "text",
          "url": "text",
          "content": "text",
          "title": "text"
        }
      ]
    }
  ],
  "web_pro_ref": [
    {
      "question": "text",
      "search_queries": [
        {
          "query": "text",
          "answer": [
            {
              "id": "text",
              "url": "text",
              "content": "text",
              "title": "text"
            }
          ]
        }
      ]
    }
  ],
  "webSearchQuery": "text",
  "is_web_pro_r": false,
  "webSearchConfig": {
    "webSearchProvider": "linkup_normal_web_search",
    "isLinkupDeepSearch": false
  },
  "updateType": "manual",
  "path": "text",
  "lastGeneratedContentAt": "2025-12-09T10:42:34.307Z",
  "isIncludeChatHistory": true,
  "mediaUrls": [
    {
      "contentType": "text",
      "url": "text"
    }
  ]
}

Create a new Insight

Creating fully new Insights from knowledge retrieved programmatically is another use case that can be handled by API. In this case, a new insight is added based on a title and summary in a given path, where root, again, is the base folder.

Add Chat Note Manual

post

Authorizations

HTTPBearer

AuthorizationstringRequired

Bearer authentication header of the form Bearer <token>.

Body

application/json

titleany ofOptional

stringOptional

or

nullOptional

summaryany ofOptional

stringOptional

or

nullOptional

parent_pathany ofOptionalDefault: root

stringOptional

or

nullOptional

Responses

200

Successful Response

application/json

422

Validation Error

application/json

post

/cortex/notes/add-note

HTTPcURLJavaScriptPython

POST /cortex/notes/add-note HTTP/1.1
Host: 
Authorization: Bearer YOUR_SECRET_TOKEN
Content-Type: application/json
Accept: */*
Content-Length: 54

{
  "title": "text",
  "summary": "text",
  "parent_path": "root"
}

{
  "_id": "text",
  "createdAt": "2025-12-09T10:42:34.307Z",
  "modifiedAt": "2025-12-09T10:42:34.307Z",
  "title": "text",
  "isEdited": false
}

Contribute Insights to a Knowledgebase

Going one step further, insights can also be contributed and therefore stored in Knowledgebases. The large difference here is that, upon using as part of a Create a Dataroom, Insights are pulled into the context in full, while Knowledgebases are used for RAG and only relevant chunks of data are being added to the context.

Save Insights To Knowledge Base

post

Authorizations

HTTPBearer

AuthorizationstringRequired

Bearer authentication header of the form Bearer <token>.

Body

application/json

time_zoneany ofOptionalDefault: 2

numberOptional

or

nullOptional

Responses

200

Successful Response

application/json

Responseany

422

Validation Error

application/json

post

/cortex/notes/save-insights-to-knowledge-base

HTTPcURLJavaScriptPython

POST /cortex/notes/save-insights-to-knowledge-base HTTP/1.1
Host: 
Authorization: Bearer YOUR_SECRET_TOKEN
Content-Type: application/json
Accept: */*
Content-Length: 138

{
  "knowledge_folders_destination": [
    {
      "noteIds": [
        "text"
      ],
      "folderNoteIds": [
        "text"
      ],
      "knowledgeBase": "text",
      "folderPath": "root"
    }
  ],
  "time_zone": 2
}

No content