This guide covers the core process for extending FastGPT's dataset functionality by adding custom third-party document library server type definitions.
Locate Type Definition File
All third-party document library server type definitions are managed in a dedicated core file within the FastGPT project. Navigate to FastGPT\packages\global\core\dataset\apiDataset.d.ts to access this file. This is the official location for defining authentication and configuration fields for external document library integrations.
Define Custom Server Type
Use TypeScript to define the required fields for your third-party document library. For example, the Yuque Dataset requires authentication and configuration fields, which can be implemented with the following code:
export type YuqueServer = {
userId: string;
token?: string;
basePath?: string;
};Below is a breakdown of each supported field:
| Field Name | Type | Requirement | Purpose |
|---|---|---|---|
userId | string | Required | Authenticate requests to the third-party document library |
token | string | Optional | Add secure supplementary authentication, if supported by the library |
basePath | string | Optional | Enable root directory selection for the connected document library |
Root Directory Configuration Requirements
🤖 Success: If your third-party document library includes a root directory selection feature, you must include the
basePathfield in your server type definition. For full details on integrating the root directory configuration form, refer to the root directory feature documentation.
Validate Implementation
After defining your custom server type, you can reference the provided example visual aid at /imgs/thirddataset-1.png to confirm your code aligns with FastGPT's expected third-party dataset system structure. This image illustrates the standard implementation pattern for custom external document library integrations.
Source: FastGPT official source
Applicability and version scope
Use this page for the documented Integrations scenario. Confirm the FastGPT, dependency, API, and deployment versions in the official source before applying a change.
Safety guardrails
Use [REDACTED_CREDENTIAL] for credentials and private data. Confirm the documented environment and version before review.
Rollback guidance
Restore the prior technical-content authority snapshot. Restore saved configuration and data snapshots, then repeat the smallest verification scenario.