AI

The AI API provides direct access to Egnyte AI capabilities such as:

• Document Q&A
• Summarization
• Knowledge Base queries
• Hybrid search

This API is primarily synchronous (request → response).

When to Use

Use the AI API when:

• You need quick answers on documents
• You want low-latency responses
• You do not need agent-level workflows

Use Agents API if you need orchestration or multi-step reasoning.

Base URL

https://{domain}.egnyte.com/pubapi/v1/ai

Authentication

All requests require an OAuth 2.0 Bearer token in the Authorization header:

Authorization: Bearer {access_token}

See Authentication for details on obtaining a token.

Rate Limits

AI API endpoints have stricter rate limits than standard Egnyte APIs. These limits apply per access token:

If your production application requires higher limits, contact Egnyte to discuss custom arrangements.

Core Capabilities

Ask a Question

Ask a question about a specific document and receive an AI-generated answer based on the document's content.

Request

POST /pubapi/v1/ai/document/{entry-id}/ask

Path Parameters

Request Body

Example Request

curl -i -X POST "https://{domain}.egnyte.com/pubapi/v1/ai/document/{entryId}/ask" \
     -H "Content-Type: application/json" \
     -H "Authorization: Bearer YOUR_ACCESS_TOKEN" \
     -d '{
  "question": "What is the date of Christmas Day in the US in 2025?",
  "chatHistory": {
    "messages": []
  },
  "includeCitations": true
}'

Response

200 OK

Example Response

{
  "response": {
    "text": "Christmas Day in the US in 2025 is on December 25th, which falls on a Thursday."
  },
  "citations": [
    {
      "filename": "Policy.pdf",
      "entryId": "535083b1-383a-4352-b993-77900bf5d98a",
      "chunks": [
        {
          "chunkId": "0",
          "sourceText": "Policy Renewal Migration Portability"
        }
      ]
    }
  ]
}

Summarize a Document

Generate an AI-powered summary of a document.

Request

POST /pubapi/v1/ai/document/{entry-id}/summary

Path Parameters

Request Body

Example Request

curl -i -X POST "https://{domain}.egnyte.com/pubapi/v1/ai/document/{entryId}/summary" \
     -H "Content-Type: application/json" \
     -H "Authorization: Bearer YOUR_ACCESS_TOKEN" \
     -d '{
  "chatHistory": {
    "messages": []
  }
}'

Response

200 OK

Example Response

{
  "response": {
    "text": "Summary of the document is here"
  }
}

AI Assistant

Ask questions to AI Assistant, which provides AI-driven answers based on content across your Egnyte domain. This is an asynchronous API — the ask returns an execution ID immediately, and you poll a separate status endpoint until the answer is ready.

Key Behavior

• Works across multiple documents

• Supports citations, tool calls, and web search

• Can be scoped via selectedItems

• Async: poll /{executionId}/status for the result

Request

POST /pubapi/v1/ai/assistant/ask

Request Body

Example Request

curl -i -X POST "https://{domain}.egnyte.com/pubapi/v1/ai/assistant/ask" \
     -H "Content-Type: application/json" \
     -H "Authorization: Bearer YOUR_ACCESS_TOKEN" \
     -d '{
  "question": "Describe the file",
  "chatHistory": {
    "messages": []
  },
  "selectedItems": {
    "folders": [
      {
        "id": "{folderId}"
      }
    ],
    "files": [
      {
        "entryId": "{entryId}"
      }
    ]
  },
  "includeCitations": true
}'

Response

200 OK

Example Response

{
  "conversationId": "2862a232-20ac-4244-8c1e-871b7a539ae2",
  "executionId": "10cd770f-8b8f-4a76-bf0a-7418dd910fb7",
  "executionStatus": "IN_PROGRESS",
  "truncated": false,
  "actions": []
}

Get Execution Status

Poll this endpoint after calling /ask to retrieve the answer and execution details.

GET /pubapi/v1/ai/assistant/{executionId}/status

Path Parameters

Query Parameters

Execution Status Response

200 OK

Example Response

{
  "status": "COMPLETED",
  "responseText": "The provided document shows several premium examples",
  "thoughts": [],
  "citations": [
    {
      "filename": "Policy.pdf",
      "entryId": "535083b1-383a-4352-b993-77900bf5d98a",
      "objectId": "52.1e0bffbc-500d-4669-82f9-c984a33813e5",
      "previewUrl": "/navigate/file/3bf2374e-49ac-45b2-b3b3-b44e0da88597",
      "chunks": [
        {
          "chunkId": "c590a032-9ac6-467a-aa65-9b58e30bff5b",
          "sourceText": "Text"
        }
      ]
    }
  ],
  "intent": "qna",
  "lastUpdated": 1780660757000,
  "numToolCalls": 2,
  "pendingActions": [],
  "toolCalls": {
    "call_abc123": {
      "name": "search_documents",
      "sourceDisplayName": "Egnyte",
      "toolDisplayName": "Document Search",
      "thought": "Searching for relevant documents",
      "input": { "query": "premium examples" },
      "output": "Found 3 relevant documents",
      "status": "success",
      "timestamp": "2025-06-05T10:00:00Z"
    }
  }
}

Copilot

Ask questions to Egnyte Copilot, which provides AI-driven answers based on content across your Egnyte domain. Optionally specify files or folders to limit the search context.

Key Behavior

• Works across multiple documents

• Supports citations

• Can be scoped via selectedItems

Request

POST /pubapi/v1/ai/copilot/ask

Request Body

Example Request

curl -i -X POST "https://{domain}.egnyte.com/pubapi/v1/ai/copilot/ask" \
     -H "Content-Type: application/json" \
     -H "Authorization: Bearer YOUR_ACCESS_TOKEN" \
     -d '{
  "question": "Describe the file",
  "chatHistory": {
    "messages": []
  },
  "selectedItems": {
    "folders": [
      {
        "id": "{folderId}"
      }
    ],
    "files": [
      {
        "entryId": "{entryId}"
      }
    ]
  },
  "includeCitations": true
}'

Response

200 OK

Deprecation Response Headers

Example Response

{
  "response": {
    "text": "The provided document shows several premium examples"
  },
  "citations": [
    {
      "filename": "Policy.pdf",
      "entryId": "535083b1-383a-4352-b993-77900bf5d98a",
      "objectId": "52.1e0bffbc-500d-4669-82f9-c984a33813e5",
      "previewUrl": "/navigate/file/3bf2374e-49ac-45b2-b3b3-b44e0da88597",
      "chunks": [
        {
          "chunkId": "c590a032-9ac6-467a-aa65-9b58e30bff5b",
          "sourceText": "Text"
        }
      ]
    }
  ],
  "conversationId": "c655a318-268c-47d2-b6e0-d53dedd117ef",
  "deprecationNotice": "This endpoint is deprecated and will be removed on September 30, 2026. Use POST /pubapi/v1/ai/assistant/ask instead. Note - the replacement endpoint is asynchronous: POST /assistant/ask returns an executionId; poll GET /assistant/{executionId}/status to retrieve the response."
}

List Knowledge Bases

Retrieve a list of Knowledge Bases (KBs) available in your Egnyte domain.

When to Use

• Structured enterprise knowledge

• Pre-indexed document sets

Request

POST /pubapi/v1/ai/kb/list

Request Body

Example Request

curl -i -X POST "https://{domain}.egnyte.com/pubapi/v1/ai/kb/list" \
     -H "Content-Type: application/json" \
     -H "Authorization: Bearer YOUR_ACCESS_TOKEN" \
     -d '{
  "sortBy": [
    "name"
  ],
  "sortDirection": [
    "ASC"
  ],
  "status": [
    "ACTIVE"
  ],
  "includePlaceholderData": false,
  "includeProcessingStatistics": false,
  "includePrompts": false
}'

Response

200 OK

Example Response

{
  "content": [
    {
      "id": "a691dcbf-2d97-4420-8b08-720c88809a9e",
      "name": "KB",
      "description": "KB description",
      "paths": [
        {
          "id": "8804124f-5fb7-476c-846d-10d6c94f7c33",
          "folderId": "0f53980d-7a3b-44d7-ae15-42048bacf993",
          "path": "/Shared/Documents",
          "permission": "Owner",
          "status": "ACTIVE"
        }
      ],
      "status": "ACTIVE",
      "type": "KBA",
      "createdBy": "Ankesh Katiyar",
      "createdByUser": {
        "firstName": "Ankesh",
        "lastName": "Katiyar",
        "userName": "akatiyar@akatiyar",
        "userId": 1,
        "avatarEosObjectId": ""
      },
      "createdOn": 1738570114257,
      "noResponseMessage": "",
      "iconName": "add",
      "subType": "USER_DEFINED",
      "progress": 100,
      "lastProcessedAt": 1738570336155,
      "pathCount": 1,
      "prompts": [],
      "subStatus": "WITHIN_LIMIT"
    }
  ],
  "number": 0,
  "size": 200,
  "first": true,
  "last": true,
  "empty": false,
  "numberOfElements": 1,
  "totalElements": 1,
  "totalPages": 1,
  "fileLimit": 10000
}

Ask Knowledge Base

Ask a question to a specific Knowledge Base and receive an AI-generated answer based on the KB's content.

Request

POST /pubapi/v1/ai/kb/{kb-id}/ask

Path Parameters

Request Body

Example Request

curl -i -X POST "https://{domain}.egnyte.com/pubapi/v1/ai/kb/{KbId}/ask" \
     -H "Content-Type: application/json" \
     -H "Authorization: Bearer YOUR_ACCESS_TOKEN" \
     -d '{
  "question": "Explain all the organizational HR policies",
  "chatHistory": {
    "messages": [
      {
        "role": "user",
        "content": "Show me the policies"
      },
      {
        "role": "assistant",
        "content": "I am sorry, but I do not know the answer to this question."
      }
    ]
  },
  "includeCitations": true
}'

Response

200 OK

Example Response

{
  "response": {
    "text": "The provided document shows the various HR policies."
  },
  "citations": [
    {
      "filename": "Policy.pdf",
      "entryId": "535083b1-383a-4352-b993-77900bf5d98a",
      "objectId": "52.1e0bffbc-500d-4669-82f9-c984a33813e5",
      "previewUrl": "/navigate/file/3bf2374e-49ac-45b2-b3b3-b44e0da88597",
      "chunks": [
        {
          "chunkId": "c590a032-9ac6-467a-aa65-9b58e30bff5b",
          "sourceText": "Policy Renewal Migration Portability"
        }
      ]
    }
  ],
  "conversationId": "c655a318-268c-47d2-b6e0-d53dedd117ef"
}

Hybrid Search

Perform a hybrid search that combines traditional keyword-based search with semantic (vector) search to deliver more relevant results.

Key Use Case

• Retrieve relevant chunks before sending to LLM

Important Notes

• Responses may include structured citations

• Tables and rich content are returned as text chunks

• No polling required (synchronous)

Request

POST /pubapi/v1/hybrid-search

Request Body

Example Request

curl -i -X POST "https://{domain}.egnyte.com/pubapi/v1/hybrid-search" \
     -H "Content-Type: application/json" \
     -H "Authorization: Bearer YOUR_ACCESS_TOKEN" \
     -d '{
  "semanticWeight": 0.5,
  "query": "quarterly financial report",
  "folderPath": "/Shared/Finance",
  "createdBy": "john.doe",
  "createdAfter": 1609459200000,
  "limit": 50,
  "preferredFolderPath": "/Shared/Finance/2023"
}'

Response

200 OK

Example Response

{
  "results": [
    {
      "filename": "Financial_Report_2023.pdf",
      "entryId": "a1b2c3d4-e5f6-7890-abcd-ef1234567890",
      "objectId": "obj123456",
      "chunks": [
        {
          "chunkId": "chunk-123",
          "chunkText": "The quarterly financial report shows a 15% increase in revenue compared to last year.",
          "type": "TEXT",
          "score": 0.92
        },
        {
          "chunkId": "chunk-124",
          "chunkText": "Operating expenses decreased by 5% due to cost optimization initiatives.",
          "type": "TEXT",
          "score": 0.85
        }
      ]
    },
    {
      "filename": "Budget_Forecast.xlsx",
      "entryId": "b2c3d4e5-f6g7-8901-bcde-fg2345678901",
      "objectId": "obj234567",
      "chunks": [
        {
          "chunkId": "chunk-456",
          "chunkText": "Q1 2023 | Q2 2023 | Q3 2023 | Q4 2023 (Projected)\n$1.2M | $1.4M | $1.5M | $1.7M",
          "type": "TABLE",
          "score": 0.85
        }
      ]
    }
  ]
}

Error Codes

Code Examples

curl -i -X POST "https://{domain}.egnyte.com/pubapi/v1/ai/document/13c5b4d4-a324-4141-b2a3-365625969d1e/ask" \
     -H "Content-Type: application/json" \
     -H "Authorization: Bearer YOUR_ACCESS_TOKEN" \
     -d '{
  "question": "What is the date of Christmas Day in the US in 2025?",
  "chatHistory": {
    "messages": []
  },
  "includeCitations": true
}'

Related Resources

• Authentication — How to obtain and refresh OAuth tokens
• Best Practices — Rate limiting, pagination, and error handling strategies
• File System API — Manage files and folders in Egnyte