MCP Server
Claix as an MCP server
Endpoint
https://www.claix.dev/mcpClaix exposes a Model Context Protocol (MCP) server so AI assistants and automation tools can run extraction, schema listing, and Agent mode without hand-written REST integrations.
The public server URL is https://www.claix.dev/mcp. Authentication uses your Claix API key in the x-api-key header (or the optional api_key argument per tool when headers are not supported).
Tools use dot notation naming (claix.schemas.list, claix.extract.pdf, …) and return structuredContent with success, data, and error.
1. Compatible clients
Any MCP-capable program can connect to Claix. Common options in production and development:
- Cursor
- Claude Desktop
- Claude Code
- Windsurf
- Cline
- Continue
- Zed
- Smithery
- Lovable
- Replit Agent
- ChatGPT (with MCP connectors)
- n8n
- Make
- Zapier (via HTTP/MCP)
- LangGraph / LangChain
- Any MCP client with Streamable HTTP or SSE
IDEs and coding agents (Cursor, Windsurf, Cline…) typically use JSON config with the server URL and headers. Smithery and Streamable HTTP clients send JSON-RPC directly to POST /mcp. Claude Desktop usually uses SSE via mcp-remote (GET /mcp + POST /mcp/message).
2. Authentication
Recommended — HTTP header on every request:
x-api-key: <YOUR_API_KEY>
In Smithery, configure the header with x-from: { "header": "x-api-key" } so users enter their key when connecting. The API key is optional in the config schema when the client already sends x-api-key.
Per-tool fallback: many tools accept api_key in arguments when the MCP client cannot send custom headers.
3. Transport modes
| Mode | Typical use | How to connect |
|---|---|---|
| Streamable HTTP | Smithery, Cursor, modern clients | POST /mcp with JSON-RPC (initialize, tools/list, tools/call). Header Accept: application/json, text/event-stream |
| Legacy SSE | Claude Desktop, mcp-remote | GET /mcp (SSE stream) + POST /mcp/message?sessionId=… for messages |
Both modes are active on the same base URL. The server auto-detects the request type.
Static metadata for scanning (Smithery, clients): https://www.claix.dev/.well-known/mcp/server-card.json
4. Recommended workflow
- Call
claix.schemas.listto getschema_id, type, and whether Agent mode is enabled. - Use
claix.extract.*for structured extraction orclaix.agent.*whenis_agent_mode=true. - Send the file as
file_base64(Base64 or data URL). The server rebuilds multipart requests to the Claix REST API. - Read the response in
structuredContent.data(typed JSON) or the text incontent.
5. Available tools
After connecting, call tools/list for the full catalog with inputSchema, outputSchema, and annotations. Main tools:
| Tool | Description |
|---|---|
| claix.schemas.list | Lists account schemas (id, name, type, is_agent_mode, agent_definition). Call first to discover schema_id. |
| claix.extract.excel | Excel/CSV → typed JSON (POST /api/excel-json). First sheet only. |
| claix.extract.pdf | PDF → typed JSON (POST /api/pdf-json). Text or scanned, max 15 MB. |
| claix.extract.doc | Document (.docx, .txt, .md, .rtf) → JSON (POST /api/doc-json). |
| claix.extract.image | Image (JPEG, PNG, WebP, HEIC) → JSON (POST /api/img-json). |
| claix.convert.json_to_excel | JSON → Excel .xlsx (POST /api/json-excel). Returns file_base64. |
| claix.agent.excel | Excel/CSV + Agent mode → data[] and agent_data (POST /agent/excel-json). |
| claix.agent.pdf | PDF + Agent mode → data[] and agent_data (POST /agent/pdf-json). |
| claix.agent.doc | Document + Agent mode (POST /agent/doc-json). |
| claix.agent.image | Image + Agent mode (POST /agent/img-json). |
6. Workflow prompts
The server exposes reusable prompts via prompts/list:
| Prompt | Description |
|---|---|
| workflow.discover-and-extract | Discover schemas via claix.schemas.list and extract JSON with claix.extract.*. |
| workflow.agent-document-analysis | Run claix.agent.* when the schema has is_agent_mode enabled. |
| workflow.invoice-pdf | Invoice PDF workflow using claix.extract.pdf. |
7. Resources
Documentation resources available via resources/list:
| URI | Description |
|---|---|
| claix://docs/mcp | MCP connection guide and tool catalog. |
| claix://docs/openapi | Claix OpenAPI specification URL. |
| claix://docs/tools | JSON catalog of all MCP tools. |
8. Smithery example
One-click install:
npx -y @smithery/cli run info-f4xz/claix
Manual configuration:
{
"mcpUrl": "https://www.claix.dev/mcp",
"headers": {
"x-api-key": "YOUR_API_KEY"
}
}Directory listing: smithery.ai/server/info-f4xz/claix
9. Cursor example
Add to Cursor MCP settings (Settings → MCP):
{
"mcpServers": {
"claix": {
"url": "https://www.claix.dev/mcp",
"headers": {
"x-api-key": "YOUR_API_KEY"
}
}
}
}10. Claude Desktop example
{
"mcpServers": {
"claix": {
"command": "npx",
"args": [
"-y",
"mcp-remote",
"https://www.claix.dev/mcp",
"--header",
"x-api-key:YOUR_API_KEY"
]
}
}
}11. n8n and automation
In n8n, use an HTTP Request node with POST https://www.claix.dev/mcp, JSON-RPC headers, and body {"jsonrpc":"2.0","id":1,"method":"tools/call","params":{"name":"claix.extract.pdf","arguments":{"schema_id":"…","file_base64":"…"}}}. Community MCP nodes also work — point the base URL to Claix and pass x-api-key.