claix

Endpoint

Hosted endpoint · transport

Connect

This is a remote MCP server — point an MCP client at the endpoint URL. Clients with native remote support (Claude, Cursor, VS Code, …):

For stdio-only clients (older Claude Desktop, etc.), bridge with mcp-remote:

If the server requires authentication, add the provider's token/header (e.g. an X-API-Key) per its docs. A quick reachability check:

Tools

• claix.schemas.list — List the Claix schemas available for the authenticated account. Returns id, name, type, schema_definition, is_agent_mode, and agent_definition. Call this first whenever you need a schema_id before claix.extract.* or claix.agent.* tools.
• claix.schemas.create — Create a Claix schema (POST /api/create-schema). Requires name, type (excel-json, json-excel, pdf-json, doc-json, img-json, txt-json) and schema_definition. Optional: is_agent_mode, agent_definition, resumen_agent, window_context, window_time.
• claix.schemas.delete — Delete a Claix schema owned by the API key (POST /api/delete-schema). Requires schema_id.
• claix.extract.excel — Extract structured JSON from an Excel (.xlsx) or CSV file using a Claix excel-json schema. Processes the first sheet only. Returns typed JSON matching the schema. Use when the user attaches a spreadsheet or CSV.
• claix.extract.pdf — Extract structured JSON from a PDF (text or scanned, max 15 MB) using a Claix pdf-json schema. Returns a single JSON object in data[]. Use for invoices, contracts, forms, or reports.
• claix.extract.doc — Extract structured JSON from a Word document (.docx, .txt, .md, .rtf; max 10 MB) using a Claix doc-json schema. Legacy .doc is not supported.
• claix.extract.image — Extract structured JSON from a visible image (JPEG, PNG, WebP, HEIC/HEIF; max 15 MB) using a Claix img-json schema. Use for photos of documents, tickets, or labels.
• claix.extract.text — Extract structured JSON from already processed plain text, HTML, or XML using a Claix txt-json schema. Send the payload in the content field (no file). Max 300,000 characters. POST /api/txt-json.
• claix.convert.json_to_excel — Convert JSON to an Excel .xlsx file using a Claix json-excel schema. Provide json_data inline or file_base64 with a .json file. Returns base64-encoded xlsx on success.
• claix.agent.excel — Run Excel/CSV extraction plus Agent mode semantic reasoning. Requires is_agent_mode on the schema. Returns data[] and agent_data with typed answers from agent_definition.
• claix.agent.pdf — Run PDF extraction plus Agent mode reasoning. Requires is_agent_mode. Returns structured data[] and agent_data for business rules, booleans, or summaries defined in the schema.
• claix.agent.doc — Run document extraction plus Agent mode reasoning on .docx/.txt/.md/.rtf. Requires is_agent_mode. Returns data[] and agent_data.
• claix.agent.image — Run image extraction plus Agent mode reasoning. Requires is_agent_mode. Returns data[] and agent_data from visible content.
• claix.agent.text — Run text/HTML/XML extraction plus Agent mode reasoning. Requires is_agent_mode. Send content (no file). Returns data[] and agent_data. POST /agent/txt-json.
• claix.window_context.get — Retrieve raw content and metadata from a persisted document (GET /window-context/{document_id}). Requires document_id from an extraction with window_context. Returns success, document_id, file_name, schema_id, processed_at, and content. Free call.
• claix.window_context.ask — Ask up to 5 questions about a persisted document (POST /window-context/{document_id}). Requires document_id from an extraction with window_context. Each question max 400 characters. Returns user_ask and ia_response (null if not in the document).
• claix.document.delete — Delete a persisted document by document_id (DELETE /delete-document/{document_id}). No body or query params. Returns { document_id } on success. Irreversible. Free call. 404 if the document does not exist, is foreign, or was already purged.

About MCP

The Model Context Protocol (MCP) is an open protocol Anthropic introduced for connecting LLM-based agents to external tools and data sources. Providers publish MCP servers that expose their API surface as structured, discoverable tools — an MCP-compatible client (Claude Desktop, Cursor, Cline, Continue, etc.) can connect to the server and call its tools without any per-provider integration code.

Browse every MCP server on the APIs.io network or compare with the broader Agent Skill surfaces of the same providers.

Work with this as data

Every MCP server here is available over the APIs.io API and to AI agents over MCP.

curl "https://apis.io/api/v1/mcp/claix"

curl "https://apis.io/api/v1/mcp?limit=25"