Search API documentation.

Everything you need to add instant search to your app or AI agent. Pick your integration path below.

MCP Server for AI Agents

Connect Claude, Cursor, Windsurf, or any MCP client to Sprigr in one config block. Your agent gets 16 tools for search, storage, and index management.

Install

npx -y @sprigr/mcp-server

Configuration

Add to your MCP client config (Claude Desktop, Cursor, etc.):

claude_desktop_config.json

{
  "mcpServers": {
    "sprigr": {
      "command": "npx",
      "args": ["-y", "@sprigr/mcp-server"],
      "env": {
        "SPRIGR_API_KEY": "your-api-key"
      }
    }
  }
}

Environment variables

SPRIGR_API_KEY  - Required. Your API key from signup.
SPRIGR_API_BASE - Optional. Defaults to https://search.sprigr.com

Available tools (16)

signuplist_indexescreate_indexupdate_settingsimport_objectssearchmulti_searchget_objectexport_objectsget_statsdelete_objectsdelete_indexclear_indexset_aliasdelete_aliaslist_aliases

How it works

The MCP server runs locally on your machine via npx. Searches execute locally in the browser (~5ms) against a binary index that auto-downloads from the server. Writes go to the cloud API and the local index refreshes automatically on the next search. No server to host, no ports to open.

Learn more about MCP →

REST API

Use the REST API directly from any language. Algolia-compatible endpoints for indexing, search, and management.

Base URL

https://search.sprigr.com

Authentication

Include your API key in the request header:

X-Sprigr-API-Key: your-api-key

API keys support three permission levels (admin, write, search) and optional index-level ACLs. Restrict a key to specific indexes by setting allowed_indexes when creating the key via the admin panel.

Create an index

PUT /1/indexes/products/settings
Content-Type: application/json

{
  "searchable_attributes": ["title", "description"],
  "attributes_for_faceting": ["category", "brand"]
}

Import objects

POST /1/indexes/products/batch
Content-Type: application/json

{
  "products": [
    {
      "objectID": "prod-1",
      "title": "Ergonomic Keyboard",
      "description": "Split mechanical keyboard",
      "category": "Tech",
      "price": 149.99
    }
  ]
}

Search

POST /1/indexes/products/query
Content-Type: application/json

{
  "query": "keyboard",
  "filters": "category:Tech",
  "hits_per_page": 10,
  "page": 0
}

All endpoints

GET/1/indexesList indexes
PUT/1/indexes/{index}/settingsCreate/update settings
GET/1/indexes/{index}/settingsGet settings
POST/1/indexes/{index}/batchImport objects
POST/1/indexes/{index}/querySearch
POST/1/indexes/*/queriesMulti-index search
GET/1/indexes/{index}/objects/{id}Get object
GET/1/indexes/{index}/objectsExport all objects
POST/1/indexes/{index}/deleteObjectsDelete objects
POST/1/indexes/{index}/clearClear index
DEL/1/indexes/{index}Delete index
PUT/1/aliases/{alias}Set alias
DEL/1/aliases/{alias}Delete alias
GET/1/aliasesList aliases
GET/1/statsAccount stats
POST/1/embedGenerate embedding

Data modeling tips

Searchable types

Strings, arrays of strings, numbers, and booleans are all indexed for full-text search. Nested objects are not. If you store {"tags": ["red", "blue"]}, searching for "red" will match.

Multi-value facets

Arrays work as multi-value facets. If tags is a faceting attribute and an object has "tags": ["red", "large"], filtering by tags:red or tags:large will both match.

Semantic search with _keywords

Include a _keywords string field with comma-separated tags, synonyms, and related concepts. This makes objects findable by meaning, not just exact text. AI agents should generate these automatically at import time.

{
  "objectID": "prod-1",
  "title": "Ergonomic Laptop Stand",
  "description": "Adjustable aluminum stand",
  "_keywords": "desk accessory, workstation, posture,
               home office, monitor riser, portable"
}

Always include _keywords in your searchable_attributes:

"searchable_attributes": ["title", "description", "_keywords"]

Hybrid semantic search

Sprigr supports hybrid search that combines traditional keyword matching with AI-powered vector search. When enabled, every query returns results ranked by both exact keyword relevance and semantic meaning, merged using Reciprocal Rank Fusion (RRF).

What is hybrid semantic search?

Standard full-text search matches keywords literally. A query for "frustrated conversation" only finds records containing those exact words. Hybrid semantic search adds a vector layer: Sprigr generates embeddings for your data and compares the meaning of queries against the meaning of your records. The two result sets are merged with RRF so you get the best of both worlds, precise keyword hits and semantically related matches.

Enabling semantic search

Enable it per index by setting semantic_search to true in your index settings:

REST API

PUT /1/indexes/my-index/settings
Content-Type: application/json

{
  "searchable_attributes": ["title", "content"],
  "attributes_for_faceting": ["category"],
  "semantic_search": true
}

How it works

When semantic_search is enabled on an index:

On import: Sprigr generates vector embeddings for each record automatically. Embeddings are stored alongside the full-text index.
On query: The search query is embedded in real time. Sprigr runs both a keyword search and a vector similarity search in parallel.
Merging: Results from both searches are combined using Reciprocal Rank Fusion (RRF). This produces a single ranked list that surfaces records matching by keywords, by meaning, or both.

API usage

Use the POST /1/indexes/{index}/query endpoint with the semantic parameter:

REST API - hybrid search query

POST /1/indexes/docs/query
Content-Type: application/json

{
  "query": "frustrated conversation",
  "semantic": true,
  "hits_per_page": 10
}

You can also generate an embedding directly without searching, using the POST /1/embed endpoint:

REST API - generate embedding

POST /1/embed
Content-Type: application/json

{
  "text": "frustrated conversation about billing"
}

This returns an embedding vector (384 dimensions) that you can use for custom similarity logic or external vector stores.

Notes

Semantic search is opt-in per index. Indexes without semantic_search: true continue to use keyword-only search with no changes.
Embeddings are generated automatically on import. There is no separate embedding step required.
Hybrid search is available on all plans, including the Free tier.
The semantic query parameter can be omitted to fall back to keyword-only search even on indexes with semantic search enabled.

Ready to get started?

Free for up to 1,000 objects. No credit card required.

Start free See pricing