canonical: https://jentic.com/apis/vtex.local/intelligent-search-api

# Vtex Intelligent Search API

>ℹ️ Onboarding guide > > Check the new [Search onboarding guide](https://developers.vtex.com/docs/guides/search-overview). We created this guide to improve the onboarding experience for developers at VTEX. It assembles all documentation on our Developer Portal about Search and is organized by focusing on the developer's journey. [VTEX Intelligent Search](https://help.vtex.com/en/tracks/vtex-intell. The API exposes 7 endpoints.

## For AI agents

Programmatically get list of the 10 most searched terms, get list of suggested terms and attributes similar to the search term. Covers 7 operations.

## Scope

Does not handle payments, communications, or crm - use for hr and recruiting only.

## Capabilities

- Get list of the 10 most searched terms
- Manage hr recruiting data programmatically
- Integrate Intelligent Search API into automated workflows
- Query and filter Intelligent Search API records by parameters
- Monitor Intelligent Search API operational status and events

## Use cases

### HR and Recruiting Operations

Use the Intelligent Search API to perform hr recruiting operations programmatically. The API provides 7 endpoints covering core functionality including get list of the 10 most searched terms, get list of suggested terms and attributes similar to the search term, get attempt of correction of a misspelled term.

Example prompt: Call GET /top_searches to get list of the 10 most searched terms

### Automated Autocomplete Management

Automate autocomplete operations by combining multiple Intelligent Search API endpoints. Agents can get list of suggested terms and attributes similar to the search term and then get attempt of correction of a misspelled term in a single workflow.

Example prompt: Call GET /autocomplete_suggestions to get list of suggested terms and attributes similar to the search term, then verify the result

### AI Agent Integration via Jentic

AI agents discover and call Intelligent Search API endpoints through Jentic without managing credentials directly. An agent searches for the required operation by intent, receives the matching endpoint schema, and executes the call with Jentic-managed authentication. This eliminates the need to read API documentation or handle none tokens manually.

Example prompt: Search Jentic for 'get list of the 10 most searched terms', load the operation schema, and execute with Jentic-managed credentials

## Key endpoints

| Method | Path | Description |
| --- | --- | --- |
| GET | `/top_searches` | Get list of the 10 most searched terms |
| GET | `/autocomplete_suggestions` | Get list of suggested terms and attributes similar to the search term |
| GET | `/correction_search` | Get attempt of correction of a misspelled term |
| GET | `/banners/{facets}` | Get list of banners registered for query |
| GET | `/search_suggestions` | Get list of suggested terms similar to the search term |
| GET | `/product_search/{facets}` | Get list of products for a query |
| GET | `/facets/{facets}` | Get list of the possible facets for a given query |

## Key resources

- **Autocomplete** — Operations related to Autocomplete
- **Product List Page** — Operations related to Product List Page

## Why Jentic

- **Setup:** Wiring the VTEX Intelligent Search API by hand means templating the {accountName} and {environment} host with the intelligent-search path prefix and shaping each facet query yourself. Through Jentic you install once, import the Intelligent Search API from the API Directory, and your agent calls it.
- **Permission scoping:** The Intelligent Search API targets each query through request parameters rather than a resource id in the path, so limit the agent to the operations it needs, such as autocomplete suggestions or product search. Every operation here reads search data, so no writes are possible.
- **Credential handling:** This Intelligent Search API endpoint set is public and needs no credential, so there is no secret for your agent to hold. Any configuration your Jentic One instance stores is kept encrypted and injected at execution time, never entering the agent's prompt, logs, or context.
- **Discovery method:** Agents search Jentic by intent such as 'get the top searched terms' or 'fetch autocomplete suggestions', and Jentic returns the matching Intelligent Search API operation with its input schema so the agent calls the right endpoint without browsing the reference docs.

## Related APIs

- **Greenhouse** — Alternative hr recruiting API
- **Lever** — Alternative hr recruiting API
- **Workday** — Complementary hr recruiting API

## FAQ

### What authentication does the Intelligent Search API use?

The Intelligent Search API uses no authentication. Through Jentic, these credentials are stored encrypted in your Jentic One instance and injected at execution time, so raw secrets never enter the agent context.

### Can I get list of the 10 most searched terms with the Intelligent Search API?

Yes. Use the GET /top_searches endpoint. The API returns structured JSON responses that agents can parse and act on directly.

### What are the rate limits for the Intelligent Search API?

Rate limits are not specified in the OpenAPI spec. Check the vendor documentation for current limits. Through Jentic, rate limiting is handled automatically with retry logic built into the execution layer.

### How do I get list of the 10 most searched terms through Jentic?

Install the Jentic SDK with pip install jentic, authenticate through Jentic One, the self-hosted execution layer, then search for 'get list of the 10 most searched terms'. Jentic returns the matching Intelligent Search API operation with its input schema. Load the schema and execute the call - credentials are injected automatically.

### How many endpoints does the Intelligent Search API have?

The Intelligent Search API exposes 7 endpoints covering autocomplete, product list page operations.

### Can I limit what my agent is allowed to do with the Intelligent Search API?

Yes. Because you run Jentic One yourself, your own rules decide which of this API's operations the agent may call, so you can allow only what a task needs, such as GET /autocomplete_suggestions and GET /search_suggestions, while withholding GET `/product_search/{facets}` or GET /top_searches. Every operation here reads search data and targets each query through request parameters rather than a resource id in the path, so the agent can only retrieve results and can never write or delete. This lets the operator keep an agent scoped to exactly the search reads it is meant to perform.