Claix is the controlled data retrieval and security layer for AI agents: it connects to your relational database and lets each agent retrieve only the tables and columns you allow and only execute the actions allowed following your rules and limits, without moving your data. Available via REST API, MCP, and A2A.

How do I get started?

Connect any SQL database with a read-only role (Claix drafts a Data Map for you to review), define routes and agents with your rules (no YAML or code), write Guard rules in plain English for the changes an agent may propose, then call one endpoint so your agent gets live typed answers and a verdict before any change, inside the perimeter you set.

Does Claix copy my data into a vector store?

No. Claix does not copy your data or build a vector index over it. Agents query your database where it lives, within the tables and columns you allow. Every query is validated before it runs and traced afterward.

Does Claix write to my database?

No. Claix never writes to your database. Retrieve connects with a read-only role. When an agent proposes an INSERT, UPDATE, or DELETE, Claix Guard answers allow or deny, with the reason, and your code executes. Loading documents into structured fields is a separate product, Claix Ingest, which you choose to use.

Which databases does Claix support?

Any SQL database, including Postgres, MySQL, and MariaDB, plus Postgres-compatible setups such as Supabase. You connect with a read-only role. Claix never executes writes: Guard only answers whether a proposed change is allowed, and your code runs it.

What are routes and agents?

A route is a retrieval capability: the tables and columns allowed plus a short guide with your rules and vocabulary. An agent is the identity that uses them—its own endpoint, business prompt, attached routes, and response mode (raw results or reasoned typed answers). Create a route once and attach it to as many agents as you need.

How do I call a Retrieve agent?

POST to /retrieve/{agent_name} with a JSON body of questions (each with an optional format such as string, int, boolean, json, or array:string). You get typed answers in order plus a log_id. The same agent is also reachable over MCP (claix.retrieve.ask) and A2A.

What answer formats can I request?

In reasoned mode you can ask for string, int, boolean, timestamp, timestamptz, date, json, array:string, or array:int. Typed values mean your workflow can branch on booleans, use numbers and lists directly, and skip parsing free text.

How does pricing work?

Every account gets 30 successful calls free each month, with no card. After that, Claix Managed AI is €0.10 per question. With your own LLM key (BYOK) from OpenAI, Google, Anthropic, or xAI, the Claix fee drops to €0.02 per question and you pay your provider’s tokens directly. Results with your own key depend on the model you choose—test in the Sandbox before going live. Claix Guard is €0.01 per call: each check of an INSERT, UPDATE, or DELETE. There is no model and no managed AI on that check.

Is agentic RAG the same as vector RAG?

No. They sit on different axes. “Agentic” is about who triggers retrieval (a fixed pipeline vs an agent calling a tool). “Vector” is about how results are matched (embeddings). An agent can call SQL, BM25, or embeddings; a fixed pipeline can use hybrid search. For relational business data, embeddings are often the wrong default.

Is SQL retrieval RAG?

Yes. RAG means retrieved information feeds the model. It does not require embeddings, and the retrieved information does not have to be text chunks. A SQL query that returns live records for the model is still retrieval.

When should I still use a vector database?

When you need semantic discovery over a large permanent text collection, heavy repeated search traffic, or a product where search is the product. For exact values, counts, dates, and live rows already in your SQL database, query the database where it lives with scoped access instead of copying it into an index.

Can I use Claix from LangChain, Cursor, or n8n?

Yes. Claix is not an agent framework—it is a retrieval layer you call from the stack you already use. Call REST from any HTTP client (including n8n), expose the agent as an MCP tool for Cursor and compatible clients, or connect over A2A so other agents can delegate data questions.

What happens when the answer is not in the allowed tables?

The field comes back empty instead of a guess. Queries only run inside the routes you attached to the agent; everything outside that perimeter does not exist for them. Traces show what was asked, which route answered, and what came back.

How is Claix different from giving an agent a database MCP server?

A generic database MCP connection is quick to wire up, but the access model depends on the server and the role you attach. Claix adds routes, per-agent allowlists, validation before queries run, typed multi-question answers, Claix Guard — which checks every proposed change against exact rules before your code runs — and a reviewable trace, so you define the boundary once and enforce it on every call.

Is MCP a security layer?

No. MCP defines how an agent calls a tool, not what it is allowed to do. A database MCP server with a broad role gives the agent that role's power.

Is Claix a competitor to MCP?

No. Claix Retrieve is an MCP server (the tool is claix.retrieve.ask), and Claix Guard checks statements headed for any MCP database tool before execute_sql.

When is Retrieve better than a raw MCP database server?

When an agent acts in production on data that matters and you want scoped tables and columns, typed answers, and no SQL written by a model. A raw MCP server is still the better choice for development, exploration, and ad hoc analysis.

When should I use Claix Guard?

When your agent proposes INSERT, UPDATE, or DELETE and you want a deterministic yes or no before anything runs. Claix never executes the statement. Your code does, and only after an explicit allow.

Do structured outputs make text-to-SQL safe?

No. Structured outputs make the format of the model's answer follow a schema. They do not verify that the values are correct, that the data exists, or that a statement is safe to execute.

How do I add a check before execute_sql on an MCP server?

Give the agent a guarded function that sends the statement to Claix Guard, returns the reason on deny, and calls execute_sql only on an explicit allow. Do not expose the raw tool. If the check fails, times out, or returns anything other than allow, do not run the statement.

What happens if Claix Guard is unavailable?

Design the wrapper to fail closed: if the check cannot be completed, the statement does not run.

Schemas · Create

Create a schema via API

Endpoint

POSThttps://claix.dev/api/create-schema

Creates a schema on the API key account, ready for extraction or Agent mode endpoints. It matches dashboard creation, with type and definition validation.

Always call https://claix.dev/api/create-schema.

Intended for server-to-server integrations. This call is free: it is not billed and does not consume your extraction quota.

1. Authentication

Every request must include your API key. It is a personal server credential, distinct from any session token.

Option A — Dedicated header (recommended):

x-api-key: <YOUR_API_KEY>

Option B — Standard Authorization header:

Authorization: Bearer <YOUR_API_KEY>

Either one is sufficient. If you send both, x-api-key takes priority.

The system checks that the key exists, is active, and the account is not suspended. Failure returns 401.

2. Request format

Method: POST · Content-Type: application/json

Field	Type	Required	Description
name	string	Yes	Visible schema name (max 200).
type	string	Yes	excel-json, json-excel, pdf-json, doc-json, img-json, txt-json, or audio-json.
schema_definition	object	Yes	Fields to extract. Each key is the output JSON name.
is_agent_mode	boolean	No	Enables Agent mode. If omitted but agent_definition is sent, it is assumed true.
agent_definition	object	If agent mode	Reasoning parameters. Required when is_agent_mode is true.
resumen_agent	string	No	Extra instruction for the agent phase (max 500).
window_context	boolean	No	Enables Save Data. When true, JSON extractions can insert the mapped row into the connected database (connection and mapping rules required). Defaults to false.
cita_por_campo	boolean	No	Enables source verification: each extracted field is bound to its location in the original document and checked against the source content. Defaults to false.

Each schema_definition field must look like:

{
  "nombre_dni": {
    "type": "string",
    "description": "full name printed on the ID card"
  }
}

Extraction types: string, integer, number, boolean. description is required.

Each agent_definition parameter uses boolean, string, closed, or integer. If type is closed, options must be a non-empty array:

{
  "calidad_foto": {
    "type": "closed",
    "options": ["buena calidad", "mala calidad", "ilegible", "perfecta"],
    "description": "Image quality and data legibility"
  }
}

The json-excel type does not support Agent mode.

3. How to build the call

Have your API key ready.
Send a JSON POST to https://claix.dev/api/create-schema.
Include name, type, and schema_definition.
For Agent mode, add is_agent_mode: true and agent_definition.
Check the HTTP status: 201 means it was created.

4. Request examples

Use the panel on the right for cURL, JavaScript, Node.js, Python, PHP, and n8n examples.

5. Successful response

201 Created · Content-Type: application/json

{
  "success": true,
  "schema": {
    "id": "b980cfe7-61ef-4a5a-9724-881c8a5541e2",
    "name": "DNI cliente",
    "type": "img-json",
    "schema_definition": {
      "nombre_dni": {
        "type": "string",
        "description": "nombre de la persona del dni"
      },
      "numero_dni": {
        "type": "string",
        "description": "numero de dni"
      },
      "fecha_caducidad": {
        "type": "string",
        "description": "fecha de caducidad del dni"
      }
    },
    "is_agent_mode": true,
    "agent_definition": {
      "edad": {
        "type": "integer",
        "description": "Edad exacta de la persona pero restale 3"
      },
      "calidad_foto": {
        "type": "closed",
        "options": ["buena calidad", "mala calidad", "ilegible", "perfecta"],
        "description": "La calidad de la imagen y la legibilidad de sus datos"
      },
      "es_mayor_edad": {
        "type": "boolean",
        "description": "Es mayor de edad actualmente la persona del dni?"
      },
      "fecha_vencimiento": {
        "type": "string",
        "description": "Cuado le vence el dni y cuanto tiempo queda para ello"
      }
    },
    "resumen_agent": null,
    "window_context": false,
    "window_time": null,
    "cita_por_campo": false,
    "created_at": "2026-08-17T13:10:00.000Z"
  }
}

Field	Type	Description
success	boolean	Always true when HTTP is 201.
schema	object	Persisted schema, including its UUID.
schema.id	uuid	Use later as schema_id for extraction.
schema.schema_definition	object	Normalized definition.
schema.agent_definition	object \| null	Null when is_agent_mode is false.
schema.window_context	boolean	Whether Save Data is on: JSON extractions insert the mapped row into the connected database.
schema.cita_por_campo	boolean	Whether source verification is enabled on the schema.

6. Error codes

{
  "error": "Descripción legible del problema.",
  "detalle": "Información técnica adicional (solo presente en algunos casos).",
  "log_id": "7c2e1a90-4b3d-4f8a-9e21-6d5c8b0a1f34"
}

400 — Invalid payload: missing name/type/definition, unknown type, empty description, closed without options, or json-excel with Agent mode.

401 — Authentication failed: missing, unknown, or disabled key, or suspended account.

405 — Method other than POST. · 500 — Internal error.

7. Status code summary

Code	Category	Retry?
201	Created	—
400	Client error (malformed data)	No — fix the request first
401	Authentication error	No — fix credentials first
405	Incorrect HTTP method	No — fix the method first
500	Internal server error	Yes, with caution

8. Best practices

Use https://claix.dev/api/create-schema.
type must match the extraction endpoint you will call later.
Clear descriptions improve AI accuracy for extraction and agent reasoning.
Never put your API key in frontend code or public repos.