How to Get AppId

Open the app configuration page and copy the appId query parameter from the browser address bar.

Start a Session

Authenticate with an API key. When calling chat/completions, pass appId in the request body.
For OpenAI SDK compatibility, you can also use Authorization: Bearer <apiKey>-<appId> and omit body.appId.
FastGPT resolves the app ID in this order: body.appId, the <apiKey>-<appId> suffix, then the app ID associated with a legacy API key.
In FastGPT >= v4.15.0, the team owner can enable authProxy when creating or editing a key to act as a team member. That member must have permission to access the target app and chat.
Some packages require adding v1 to the BaseUrl. If you get a 404 error, try adding v1 and retry.

Start Chat

The v1 chat API supports the OpenAI chat-completion format. Configure your client with the FastGPT BaseUrl and Authorization header, and follow these rules:

Parameters like model and temperature are ignored. These values are determined by your workflow configuration.
The top-level usage fields contain placeholder token counts. Set detail=true and total the tokens values in responseData to calculate actual usage.

Request

curl --location --request POST 'http://localhost:3000/api/v1/chat/completions' \
--header 'Authorization: Bearer fastgpt-xxxxxx' \
--header 'Content-Type: application/json' \
--data-raw '{
"appId": "your_app_id",
"chatId": "my_chatId",
"stream": false,
"detail": false,
"responseChatItemId": "my_responseChatItemId",
"variables": {
    "uid": "asdfadsfasfd2323",
    "name": "Zhang San"
},
"messages": [
    {
        "role": "user",
        "content": "Who is the director?"
    }
]
}'

Only messages differs slightly; other parameters are the same.
Direct file uploads are not supported. Upload files to your object storage and provide the URL.

curl --location --request POST 'http://localhost:3000/api/v1/chat/completions' \
--header 'Authorization: Bearer fastgpt-xxxxxx' \
--header 'Content-Type: application/json' \
--data-raw '{
    "appId": "your_app_id",
    "chatId": "abcd",
    "stream": false,
    "messages": [
        {
            "role": "user",
            "content": [
                {
                    "type": "text",
                    "text": "Who is the director?"
                },
                {
                    "type": "image_url",
                    "image_url": {
                        "url": "Image URL"
                    }
                },
                {
                    "type": "file_url",
                    "name": "File name",
                    "url": "Document URL; supported formats: txt md html word pdf ppt csv excel"
                }
            ]
        }
    ]
}'

headers.Authorization: Bearer [apikey]
chatId: string | undefined.
- Empty or omitted: FastGPT context is not used, and context is built entirely from messages.
- Non-empty string: uses chatId for the chat, automatically reads messages from the FastGPT session, and uses only the last item in messages as the user question. Other messages are ignored. Make sure chatId is unique and shorter than 250 characters. You can use the conversation ID from your own system.
messages: Same structure as GPT chat messages .
responseChatItemId: string | undefined. If provided, FastGPT uses it as the response message ID and stores it in the database. Make sure it is unique under the current chatId.
detail: Whether to return intermediate values, such as node status and complete results. In stream mode, they are separated by event; in non-stream mode, they are stored in responseData.
variables: Node variables. This object replaces [key] placeholders in input fields.

Response

Token counts, timings, and scores in the examples below are illustrative.

{
  "id": "adsfasf",
  "model": "",
  "usage": {
    "prompt_tokens": 1,
    "completion_tokens": 1,
    "total_tokens": 1
  },
  "choices": [
    {
      "message": {
        "role": "assistant",
        "content": "Suzume was directed by Makoto Shinkai."
      },
      "finish_reason": "stop",
      "index": 0
    }
  ]
}

data: {"id":"","object":"","created":0,"choices":[{"delta":{"content":""},"index":0,"finish_reason":null}]}

data: {"id":"","object":"","created":0,"choices":[{"delta":{"content":"The "},"index":0,"finish_reason":null}]}

data: {"id":"","object":"","created":0,"choices":[{"delta":{"content":"film "},"index":0,"finish_reason":null}]}

data: {"id":"","object":"","created":0,"choices":[{"delta":{"content":"Suzume"},"index":0,"finish_reason":null}]}

{
  "responseData": [
    // Response values vary by node and version. Log the response to inspect the latest structure.
    {
      "moduleName": "Dataset Search",
      "price": 1.2000000000000002,
      "model": "Embedding-2",
      "tokens": 6,
      "similarity": 0.61,
      "limit": 3
    },
    {
      "moduleName": "AI Chat",
      "price": 454.5,
      "model": "FastAI-4k",
      "tokens": 303,
      "question": "Who is the director?",
      "answer": "Suzume was directed by Makoto Shinkai.",
      "maxToken": 2050,
      "quoteList": [
        {
          "dataset_id": "646627f4f7b896cfd8910e38",
          "id": "8099",
          "q": "Who is the protagonist?",
          "a": "The protagonist is a girl named Suzume.",
          "source": "Manually edited"
        },
        {
          "dataset_id": "646627f4f7b896cfd8910e38",
          "id": "8686",
          "q": "Who is the male lead in Suzume?",
          "a": "The male lead in Suzume is Souta Munakata, voiced by Hokuto Matsumura.",
          "source": ""
        },
        {
          "dataset_id": "646627f4f7b896cfd8910e38",
          "id": "19339",
          "q": "Who directed Suzume?22",
          "a": "Suzume was directed by Makoto Shinkai.",
          "source": "Manually edited"
        }
      ],
      "completeMessages": [
        {
          "obj": "System",
          "value": "Dataset content:\n1. [Who is the protagonist?\nThe protagonist is a girl named Suzume.]\n2. [Who is the male lead in Suzume?\nThe male lead in Suzume is Souta Munakata, voiced by Hokuto Matsumura.]\n3. [Who directed Suzume?22\nSuzume was directed by Makoto Shinkai.]\n"
        },
        {
          "obj": "System",
          "value": "1. Remember that you are Baidu's next-generation knowledge-enhanced language model. Answer using only the supplied Dataset content.\n\n2. Disregard knowledge unrelated to the film Suzume."
        },
        {
          "obj": "System",
          "value": "Answer only questions about Suzume. For any other question, reply: I do not know."
        },
        {
          "obj": "Human",
          "value": "Who is the director?"
        },
        {
          "obj": "AI",
          "value": "Suzume was directed by Makoto Shinkai."
        }
      ]
    }
  ],
  "id": "",
  "model": "",
  "usage": {
    "prompt_tokens": 1,
    "completion_tokens": 1,
    "total_tokens": 1
  },
  "choices": [
    {
      "message": {
        "role": "assistant",
        "content": "Suzume was directed by Makoto Shinkai."
      },
      "finish_reason": "stop",
      "index": 0
    }
  ]
}

event: flowNodeStatus
data: {"status":"running","name":"Dataset search"}

event: flowNodeStatus
data: {"status":"running","name":"AI Chat"}

event: answer
data: {"id":"","object":"","created":0,"model":"","choices":[{"delta":{"content":"Suzume"},"index":0,"finish_reason":null}]}

event: answer
data: {"id":"","object":"","created":0,"model":"","choices":[{"delta":{"content":" was "},"index":0,"finish_reason":null}]}

event: answer
data: {"id":"","object":"","created":0,"model":"","choices":[{"delta":{"content":"directed "},"index":0,"finish_reason":null}]}

event: answer
data: {"id":"","object":"","created":0,"model":"","choices":[{"delta":{"content":"by Makoto "},"index":0,"finish_reason":null}]}

event: answer
data: {"id":"","object":"","created":0,"model":"","choices":[{"delta":{"content":"Shinkai."},"index":0,"finish_reason":null}]}

event: answer
data: {"id":"","object":"","created":0,"model":"","choices":[{"delta":{},"index":0,"finish_reason":"stop"}]}

event: answer
data: [DONE]

event: flowResponses
data: [{"moduleName":"Dataset search","moduleType":"datasetSearchNode","runningTime":1.78},{"question":"Who is the director?","quoteList":[{"id":"654f2e49b64caef1d9431e8b","q":"Who directed Suzume?","a":"Suzume was directed by Makoto Shinkai!","indexes":[{"type":"qa","dataId":"3515487","text":"Who directed Suzume?","_id":"654f2e49b64caef1d9431e8c","defaultIndex":true}],"datasetId":"646627f4f7b896cfd8910e38","collectionId":"653279b16cd42ab509e766e8","sourceName":"data (81).csv","sourceId":"64fd3b6423aa1307b65896f6","score":0.8935586214065552},{"id":"6552e14c50f4a2a8e632af11","q":"Who is the director?","a":"Suzume was directed by Makoto Shinkai.","indexes":[{"defaultIndex":true,"type":"qa","dataId":"3644565","text":"Who is the director?\nSuzume was directed by Makoto Shinkai.","_id":"6552e14dde5cc7ba3954e417"}],"datasetId":"646627f4f7b896cfd8910e38","collectionId":"653279b16cd42ab509e766e8","sourceName":"data (81).csv","sourceId":"64fd3b6423aa1307b65896f6","score":0.8890955448150635},{"id":"654f34a0b64caef1d946337e","q":"Who is the protagonist?","a":"The protagonist is a girl named Suzume.","indexes":[{"type":"qa","dataId":"3515541","text":"Who is the protagonist?","_id":"654f34a0b64caef1d946337f","defaultIndex":true}],"datasetId":"646627f4f7b896cfd8910e38","collectionId":"653279b16cd42ab509e766e8","sourceName":"data (81).csv","sourceId":"64fd3b6423aa1307b65896f6","score":0.8738770484924316},{"id":"654f3002b64caef1d944207a","q":"Who is the male lead in Suzume?","a":"The male lead in Suzume is Souta Munakata, voiced by Hokuto Matsumura.","indexes":[{"type":"qa","dataId":"3515538","text":"Who is the male lead in Suzume?","_id":"654f3002b64caef1d944207b","defaultIndex":true}],"datasetId":"646627f4f7b896cfd8910e38","collectionId":"653279b16cd42ab509e766e8","sourceName":"data (81).csv","sourceId":"64fd3b6423aa1307b65896f6","score":0.8607980012893677},{"id":"654f2fc8b64caef1d943fd46","q":"Who wrote Suzume?","a":"Makoto Shinkai wrote the screenplay.","indexes":[{"defaultIndex":true,"type":"qa","dataId":"3515550","text":"Who wrote Suzume?22","_id":"654f2fc8b64caef1d943fd47"}],"datasetId":"646627f4f7b896cfd8910e38","collectionId":"653279b16cd42ab509e766e8","sourceName":"data (81).csv","sourceId":"64fd3b6423aa1307b65896f6","score":0.8468944430351257}],"moduleName":"AI Chat","moduleType":"chatNode","runningTime":1.86}]

Event values:

answer: Text returned to the client (counts as the final answer)
chatTitle: Chat title generated from the current user question
fastAnswer: Preset reply text returned to the client (counts as the final answer)
toolCall: Tool execution
toolParams: Tool parameters
toolResponse: Tool response
flowNodeStatus: Current workflow step status
flowResponses: Complete workflow step responses
updateVariables: Updated variables
interactive: Interactive config
error: Error

Response

Interactive node responses

If your workflow contains interactive nodes, still call this API with detail=true:

stream=true: read the interactive config from event=interactive in data.interactive.
stream=false: read the element that contains the interactive field from choices[].message.content.

The interactive payload returned to external callers is display config only. It contains only type and params; internal runtime fields such as entryNodeIds, memoryEdges, nodeOutputs, and nodeResponseId are not returned. If the workflow internally hits a children / loop / tool wrapper interaction, the API returns the deepest user-facing interaction.

When calling a workflow with interactive steps, if an interaction is encountered, it returns immediately. The examples below show the element inside choices[].message.content[] when stream=false; when stream=true, event=interactive returns { "interactive": ... } as its data:

{
  "interactive": {
    "type": "userSelect",
    "params": {
      "description": "Test",
      "userSelectOptions": [
        {
          "value": "Confirm",
          "key": "option1"
        },
        {
          "value": "Cancel",
          "key": "option2"
        }
      ]
    }
  }
}

{
  "interactive": {
    "type": "userInput",
    "params": {
      "description": "Test",
      "inputForm": [
        {
          "type": "input",
          "key": "测试 1",
          "label": "Test 1",
          "description": "",
          "value": "",
          "defaultValue": "",
          "valueType": "string",
          "required": false,
          "list": [
            {
              "label": "",
              "value": ""
            }
          ]
        },
        {
          "type": "numberInput",
          "key": "测试 2",
          "label": "Test 2",
          "description": "",
          "value": "",
          "defaultValue": "",
          "valueType": "number",
          "required": false,
          "list": [
            {
              "label": "",
              "value": ""
            }
          ]
        }
      ]
    }
  }
}

Continue Interaction

After receiving interactive info, render your UI to guide user input or selection. Then call this API again to continue the workflow. Use this format:

For user selection, simply pass the selected value to messages.

curl --location --request POST 'http://localhost:3000/api/v1/chat/completions' \
--header 'Authorization: Bearer fastgpt-xxx' \
--header 'Content-Type: application/json' \
--data-raw '{
    "appId": "your_app_id",
    "stream": true,
    "detail": true,
    "chatId":"22222231",
    "messages": [
        {
            "role": "user",
            "content": "Confirm"
        }
    ]
}'

For form input, serialize an object as a JSON string in messages[].content. Use the exact form keys from the interaction response and the values entered by the user. Keep the same chatId when continuing the workflow.

curl --location --request POST 'http://localhost:3000/api/v1/chat/completions' \
--header 'Authorization: Bearer fastgpt-xxxx' \
--header 'Content-Type: application/json' \
--data-raw '{
    "appId": "your_app_id",
    "stream": true,
    "detail": true,
    "chatId":"22231",
    "messages": [
        {
            "role": "user",
            "content": "{\"测试 1\":\"Text entered in the input field\",\"测试 2\":666}"
        }
    ]
}'

Request Plugin

The request example below has one plugin input parameter, query.

Plugin API is identical to chat API, with slight parameter differences:

When calling a Plugin-type App, the endpoint uses detail mode by default.
No need to pass chatId since plugins run only once.
No need to pass messages.
Pass variables to represent plugin inputs.
Get plugin outputs from pluginData.

Request

curl --location --request POST 'http://localhost:3000/api/v1/chat/completions' \
--header 'Authorization: Bearer test-xxxxx' \
--header 'Content-Type: application/json' \
--data-raw '{
    "appId": "your_app_id",
    "stream": false,
    "chatId": "test",
    "variables": {
        "query":"Hello"
    }
}'

Response

Find plugin output by locating moduleType=pluginOutput in responseData. Its pluginOutput contains the output.
Stream output is still available via choices.

{
  "responseData": [
    {
      "nodeId": "fdDgXQ6SYn8v",
      "moduleName": "AI Chat",
      "moduleType": "chatNode",
      "totalPoints": 0.685,
      "model": "FastAI-3.5",
      "tokens": 685,
      "query": "Hello",
      "maxToken": 2000,
      "historyPreview": [
        {
          "obj": "Human",
          "value": "Hello"
        },
        {
          "obj": "AI",
          "value": "Hello! How can I help you? Ask me a question."
        }
      ],
      "contextTotalLen": 14,
      "runningTime": 1.73
    },
    {
      "nodeId": "pluginOutput",
      "moduleName": "Plugin output",
      "moduleType": "pluginOutput",
      "totalPoints": 0,
      "pluginOutput": {
        "result": "Hello! How can I help you? Ask me a question."
      },
      "runningTime": 0
    }
  ],
  "newVariables": {
    "query": "Hello"
  },
  "id": "safsafsa",
  "model": "",
  "usage": {
    "prompt_tokens": 1,
    "completion_tokens": 1,
    "total_tokens": 1
  },
  "choices": [
    {
      "message": {
        "role": "assistant",
        "content": "Hello! How can I help you? Ask me a question."
      },
      "finish_reason": "stop",
      "index": 0
    }
  ]
}

Get plugin output by deserializing the event=flowResponses string into an array. Find moduleType=pluginOutput element; its pluginOutput contains the output.
Stream output works the same as chat API.

event: flowNodeStatus
data: {"status":"running","name":"AI Chat"}

event: answer
data: {"id":"","object":"","created":0,"model":"","choices":[{"delta":{"role":"assistant","content":""},"index":0,"finish_reason":null}]}

event: answer
data: {"id":"","object":"","created":0,"model":"","choices":[{"delta":{"role":"assistant","content":"Hel"},"index":0,"finish_reason":null}]}

event: answer
data: {"id":"","object":"","created":0,"model":"","choices":[{"delta":{"role":"assistant","content":"lo"},"index":0,"finish_reason":null}]}

event: answer
data: {"id":"","object":"","created":0,"model":"","choices":[{"delta":{"role":"assistant","content":"!"},"index":0,"finish_reason":null}]}

event: answer
data: {"id":"","object":"","created":0,"model":"","choices":[{"delta":{"role":"assistant","content":" "},"index":0,"finish_reason":null}]}

event: answer
data: {"id":"","object":"","created":0,"model":"","choices":[{"delta":{"role":"assistant","content":"How"},"index":0,"finish_reason":null}]}

event: answer
data: {"id":"","object":"","created":0,"model":"","choices":[{"delta":{"role":"assistant","content":" "},"index":0,"finish_reason":null}]}

event: answer
data: {"id":"","object":"","created":0,"model":"","choices":[{"delta":{"role":"assistant","content":"can"},"index":0,"finish_reason":null}]}

event: answer
data: {"id":"","object":"","created":0,"model":"","choices":[{"delta":{"role":"assistant","content":" "},"index":0,"finish_reason":null}]}

event: answer
data: {"id":"","object":"","created":0,"model":"","choices":[{"delta":{"role":"assistant","content":"I"},"index":0,"finish_reason":null}]}

event: answer
data: {"id":"","object":"","created":0,"model":"","choices":[{"delta":{"role":"assistant","content":" "},"index":0,"finish_reason":null}]}

event: answer
data: {"id":"","object":"","created":0,"model":"","choices":[{"delta":{"role":"assistant","content":"help"},"index":0,"finish_reason":null}]}

event: answer
data: {"id":"","object":"","created":0,"model":"","choices":[{"delta":{"role":"assistant","content":" you"},"index":0,"finish_reason":null}]}

event: answer
data: {"id":"","object":"","created":0,"model":"","choices":[{"delta":{"role":"assistant","content":"?"},"index":0,"finish_reason":null}]}

event: answer
data: {"id":"","object":"","created":0,"model":"","choices":[{"delta":{"role":"assistant","content":""},"index":0,"finish_reason":null}]}

event: answer
data: {"id":"","object":"","created":0,"model":"","choices":[{"delta":{},"index":0,"finish_reason":"stop"}]}

event: answer
data: [DONE]

event: flowResponses
data: [{"nodeId":"fdDgXQ6SYn8v","moduleName":"AI Chat","moduleType":"chatNode","totalPoints":0.033,"model":"FastAI-3.5","tokens":33,"query":"Hello","maxToken":2000,"historyPreview":[{"obj":"Human","value":"Hello"},{"obj":"AI","value":"Hello! How can I help you?"}],"contextTotalLen":2,"runningTime":1.42},{"nodeId":"pluginOutput","moduleName":"Plugin output","moduleType":"pluginOutput","totalPoints":0,"pluginOutput":{"result":"Hello! How can I help you?"},"runningTime":0}]

Event values:

answer: Text returned to the client and counted as the final answer
fastAnswer: Specified Reply text returned to the client and counted as the final answer
toolCall: Run a tool
toolParams: Tool parameters
toolResponse: Tool response
flowNodeStatus: Status of the current node
flowResponses: Complete node responses
updateVariables: Updated variables
error: Error

Chat CRUD

Important Fields

appId - The app ID
chatId - The ID of a session under an app
dataId - The ID of a message under a session

Session Management

Get Session List

curl --location --request POST 'http://localhost:3000/api/core/chat/history/getHistories' \
--header 'Authorization: Bearer [apikey]' \
--header 'Content-Type: application/json' \
--data-raw '{
    "appId": "appId",
    "offset": 0,
    "pageSize": 20,
    "source": "api"
}'

appId - App ID
offset - Offset (starting position)
pageSize - Number of items
source - Chat source. source=api means get API-created sessions only (excludes web UI sessions)

{
  "code": 200,
  "statusText": "",
  "message": "",
  "data": {
    "list": [
      {
        "chatId": "usdAP1GbzSGu",
        "updateTime": "2024-10-13T03:29:05.779Z",
        "appId": "66e29b870b24ce35330c0f08",
        "customTitle": "",
        "title": "Hello",
        "top": false
      },
      {
        "chatId": "lC0uTAsyNBlZ",
        "updateTime": "2024-10-13T03:22:19.950Z",
        "appId": "66e29b870b24ce35330c0f08",
        "customTitle": "",
        "title": "Test",
        "top": false
      }
    ],
    "total": 2
  }
}

Update Session Title

curl --location --request PUT 'http://localhost:3000/api/core/chat/history/updateHistory' \
--header 'Authorization: Bearer [apikey]' \
--header 'Content-Type: application/json' \
--data-raw '{
    "appId": "appId",
    "chatId": "chatId",
    "customTitle": "Custom title"
}'

appId - App ID
chatId - Session ID
customTitle - Custom session title

{
  "code": 200,
  "statusText": "",
  "message": "",
  "data": null
}

Update Session Pin Status

curl --location --request PUT 'http://localhost:3000/api/core/chat/history/updateHistory' \
--header 'Authorization: Bearer [apikey]' \
--header 'Content-Type: application/json' \
--data-raw '{
    "appId": "appId",
    "chatId": "chatId",
    "top": true
}'

appId - App ID
chatId - Session ID
top - Whether to pin. true = pin, false = unpin

{
  "code": 200,
  "statusText": "",
  "message": "",
  "data": null
}

Delete a Session

curl --location --request DELETE 'http://localhost:3000/api/core/chat/history/delHistory?chatId=[chatId]&appId=[appId]' \
--header 'Authorization: Bearer [apikey]'

appId - App ID
chatId - Session ID

{
  "code": 200,
  "statusText": "",
  "message": "",
  "data": null
}

Clear App Sessions

This endpoint clears sessions created through an API key. Sessions from the web UI, shared links, and other sources are preserved.

curl --location --request DELETE 'http://localhost:3000/api/core/chat/history/clearHistories?appId=[appId]' \
--header 'Authorization: Bearer [apikey]'

appId - App ID

{
  "code": 200,
  "statusText": "",
  "message": "",
  "data": null
}

Message Management

Operations on messages under a specific session.

Get Session Basic Info

curl --location --request GET 'http://localhost:3000/api/core/chat/init?appId=[appId]&chatId=[chatId]' \
--header 'Authorization: Bearer [apikey]'

appId - App ID
chatId - Session ID

{
  "code": 200,
  "statusText": "",
  "message": "",
  "data": {
    "chatId": "sPVOuEohjo3w",
    "appId": "66e29b870b24ce35330c0f08",
    "variables": {},
    "app": {
      "chatConfig": {
        "questionGuide": true,
        "ttsConfig": {
          "type": "web"
        },
        "whisperConfig": {
          "open": false,
          "autoSend": false,
          "autoTTSResponse": false
        },
        "chatInputGuide": {
          "open": false,
          "textList": [],
          "customUrl": ""
        },
        "instruction": "",
        "variables": [],
        "fileSelectConfig": {
          "canSelectFile": true,
          "canSelectImg": true,
          "maxFiles": 10
        },
        "_id": "66f1139aaab9ddaf1b5c596d",
        "welcomeText": ""
      },
      "chatModels": ["GPT-4o-mini"],
      "name": "Test",
      "avatar": "/imgs/app/avatar/workflow.svg",
      "intro": "",
      "type": "advanced",
      "pluginInputs": []
    }
  }
}

Get Message List

curl --location --request POST 'http://localhost:3000/api/core/chat/record/getPaginationRecords' \
--header 'Authorization: Bearer [apikey]' \
--header 'Content-Type: application/json' \
--data-raw '{
    "appId": "appId",
    "chatId": "chatId",
    "offset": 0,
    "pageSize": 10,
    "loadCustomFeedbacks": true
}'

appId - App ID
chatId - Session ID
offset - Offset
pageSize - Number of items
loadCustomFeedbacks - Whether to load custom feedbacks (optional)

{
  "code": 200,
  "statusText": "",
  "message": "",
  "data": {
    "list": [
      {
        "_id": "670b84e6796057dda04b0fd2",
        "dataId": "jzqdV4Ap1u004rhd2WW8yGLn",
        "obj": "Human",
        "value": [
          {
            "text": {
              "content": "Hello"
            }
          }
        ],
        "customFeedbacks": []
      },
      {
        "_id": "670b84e6796057dda04b0fd3",
        "dataId": "x9KQWcK9MApGdDQH7z7bocw1",
        "obj": "AI",
        "value": [
          {
            "text": {
              "content": "Hello! How can I help you?"
            }
          }
        ],
        "customFeedbacks": [],
        "totalQuoteList": [],
        "totalRunningTime": 2.42,
        "useAgentSandbox": false
      }
    ],
    "total": 2
  }
}

Get Message Run Details

curl --location --request GET 'http://localhost:3000/api/core/chat/record/getResData?appId=[appId]&chatId=[chatId]&dataId=[dataId]' \
--header 'Authorization: Bearer [apikey]'

appId - App ID
chatId - Session ID
dataId - Message ID

{
  "code": 200,
  "statusText": "",
  "message": "",
  "data": [
    {
      "id": "mVlxkz8NfyfU",
      "nodeId": "448745",
      "moduleName": "common:core.module.template.work_start",
      "moduleType": "workflowStart",
      "runningTime": 0
    },
    {
      "id": "b3FndAdHSobY",
      "nodeId": "z04w8JXSYjl3",
      "moduleName": "AI Chat",
      "moduleType": "chatNode",
      "runningTime": 1.22,
      "totalPoints": 0.02475,
      "model": "GPT-4o-mini",
      "tokens": 75,
      "query": "Test",
      "maxToken": 2000,
      "historyPreview": [
        {
          "obj": "Human",
          "value": "Hello"
        },
        {
          "obj": "AI",
          "value": "Hello! How can I help you?"
        },
        {
          "obj": "Human",
          "value": "Test"
        },
        {
          "obj": "AI",
          "value": "The test passed! What question or topic would you like to discuss?"
        }
      ],
      "contextTotalLen": 4
    }
  ]
}

Delete Message

curl --location --request DELETE 'http://localhost:3000/api/core/chat/record/delete?contentId=[contentId]&chatId=[chatId]&appId=[appId]' \
--header 'Authorization: Bearer [apikey]'

appId - App ID
chatId - Session ID
contentId - Message ID

{
  "code": 200,
  "statusText": "",
  "message": "",
  "data": null
}

Update Feedback (Like / Dislike)

Like / unlike:

curl --location --request POST 'http://localhost:3000/api/core/chat/feedback/updateUserFeedback' \
--header 'Authorization: Bearer [apikey]' \
--header 'Content-Type: application/json' \
--data-raw '{
    "appId": "appId",
    "chatId": "chatId",
    "dataId": "dataId",
    "userGoodFeedback": "yes"
}'

Dislike / remove dislike:

curl --location --request POST 'http://localhost:3000/api/core/chat/feedback/updateUserFeedback' \
--header 'Authorization: Bearer [apikey]' \
--header 'Content-Type: application/json' \
--data-raw '{
    "appId": "appId",
    "chatId": "chatId",
    "dataId": "dataId",
    "userBadFeedback": "yes"
}'

appId - App ID
chatId - Session ID
dataId - Message ID
userGoodFeedback - User feedback when liking (optional). Omit to unlike.
userBadFeedback - User feedback when disliking (optional). Omit to remove dislike.

{
  "code": 200,
  "statusText": "",
  "message": "",
  "data": null
}

Question Suggestions

New API for releases after 4.8.16

The suggested questions feature requires both appId and chatId. It automatically fetches the last 6 message turns from the session as context.

curl --location --request POST 'http://localhost:3000/api/core/ai/agent/v2/createQuestionGuide' \
--header 'Authorization: Bearer [apikey]' \
--header 'Content-Type: application/json' \
--data-raw '{
    "appId": "appId",
    "chatId": "chatId",
    "questionGuide": {
        "open": true,
        "model": "GPT-4o-mini",
        "customPrompt": "You are an intelligent assistant. Generate suggested follow-up questions based on the question from the user."
    }
}'

Parameter	Type	Required	Description
appId	string	✅	App ID
chatId	string	✅	Session ID
questionGuide	object		Custom configuration. If omitted, uses the latest published configuration for the App ID.

type CreateQuestionGuideParams = OutLinkChatAuthProps & {
  appId: string;
  chatId: string;
  questionGuide?: {
    open: boolean;
    model?: string;
    customPrompt?: string;
  };
};

{
  "code": 200,
  "statusText": "",
  "message": "",
  "data": ["What do you think about AI?", "Would you like to learn about AI applications?", "What would you like AI to do?"]
}

Chat API

How to Get AppId

Start a Session

Start Chat

Request

Response

Response

Continue Interaction

Request Plugin

Request

Response

Chat CRUD

Session Management

Get Session List

Update Session Title

Update Session Pin Status

Delete a Session

Clear App Sessions

Message Management

Get Session Basic Info

Get Message List

Get Message Run Details

Delete Message

Update Feedback (Like / Dislike)

Question Suggestions

On this page