Quickstart: Your First API Call

Query Trove’s GraphQL API to search your knowledge base, save documents, and manage connectors.

Prerequisites

• A Trove account (Trove is currently in private beta. Contact the team for an invitation.)
• A Bearer token (see Authentication)

The GraphQL Endpoint

All API requests go to a single endpoint:

POST https://api.ontrove.sh/graphql

Every request requires two headers:

Header	Value
Authorization	Bearer <token>
Content-Type	application/json

Search Your Knowledge Base

Find documents by meaning using semantic search:

curl -X POST https://api.ontrove.sh/graphql \
  -H "Authorization: Bearer YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "query": "{ search(query: \"machine learning\", limit: 5) { results { document { id title author contentType } snippet relevanceScore } totalMatches queryTimeMs } }"
  }'

Response:

{
  "data": {
    "search": {
      "results": [
        {
          "document": {
            "id": "a1b2c3d4e5f6g7h8",
            "title": "Attention Is All You Need",
            "author": "Vaswani et al.",
            "contentType": "text"
          },
          "snippet": "The dominant sequence transduction models are based on complex recurrent or convolutional neural networks...",
          "relevanceScore": 0.92
        }
      ],
      "totalMatches": 23,
      "queryTimeMs": 45
    }
  }
}

Save a Document

Store text, notes, or any content in your knowledge base:

curl -X POST https://api.ontrove.sh/graphql \
  -H "Authorization: Bearer YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "query": "mutation { saveDocument(input: { text: \"Important note about project architecture...\", title: \"Architecture Notes\", tags: [\"architecture\", \"notes\"] }) { id title wordCount tags } }"
  }'

Saved documents are automatically indexed for semantic search.

List Your Connectors

See all configured data sources and their sync status:

curl -X POST https://api.ontrove.sh/graphql \
  -H "Authorization: Bearer YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "query": "{ connectors { id name connectorType status documentCount lastSyncedAt } }"
  }'

Response Structure

Every response follows the standard GraphQL format:

• data contains your query results on success.
• errors contains an array of error objects if something went wrong.

{
  "errors": [
    {
      "message": "Not authenticated",
      "extensions": { "code": "UNAUTHENTICATED" }
    }
  ]
}

A response can contain both data and errors if some fields resolved successfully while others failed.

Next Steps

• Authentication: How to obtain and manage tokens
• API Reference: Full GraphQL schema documentation
• Queries: All available GraphQL queries
• Mutations: All available GraphQL mutations