Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.evergreens.ai/llms.txt

Use this file to discover all available pages before exploring further.

AuthorAuthor: Adnan Ali
Tedi KnowledgebaseEntity defines the structure and properties of a knowledge base entity used in Tedi’s Knowledge Base API.
Rate Limit: The Knowledge Base Search API is currently rate limited to 30 requests every 60 seconds to ensure optimal performance and reliability. If you exceed this limit, you may receive a 429 Too Many Requests response. We recommend implementing retry logic with exponential backoff in your application to handle rate limit responses gracefully.

​
Definition

Below is the definition for the KnowledgebaseEntity:
KnowledgebaseEntity
interface KnowledgebaseEntity {
  cluster: {
    id: string;
    name: string;
    type: 'file-upload' | 'website' | 'sitemap' | string;
    metadata?: {
      url?: string;
      title?: string;
      pullType?: 'url-only' | 'entire-website' | string;
      excludeFooter?: boolean;
      excludeHeader?: boolean;
    };
    created_at: string;
    updated_at: string;
  };

  searchChunks: {
    clusterId: string;
    fileId: string;
    filename: string;
    name: string;
    text: string; // always markdown format
  };
}
PropertyTypeDescription
clusterobjectAn object representing the cluster to which the knowledge base entity belongs, including fields such as:
- id: The unique identifier of the cluster.
- name: The name of the cluster.
- type: The type of the cluster (e.g., “website”, “sitemap”, etc.).
- metadata: An optional object containing additional metadata about the cluster,
which may include fields such as url, title, pullType, excludeFooter, and excludeHeader.
searchChunksarrayAn array of search chunks associated with the knowledge base entity, where each chunk includes fields such as:
- id: The unique identifier of the chunk.
- score: The relevance score of the chunk, indicating how closely it matches a search query.
- metadata: An object containing additional metadata about the chunk, including
fields such as clusterId, fileId, filename, name, and text (which is always in markdown format).