Vinkius
Sensible logo
Vinkius
Vinkius runs on Claude Code

How to Use the Sensible MCP in Claude Code

Run high-volume document parsing pipelines directly from your terminal using Claude Code and Sensible.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

Sensible MCP on Cursor AI Code Editor MCP Client Sensible MCP on Claude Desktop App MCP Integration Sensible MCP on OpenAI Agents SDK MCP Compatible Sensible MCP on Visual Studio Code MCP Extension Client Sensible MCP on GitHub Copilot AI Agent MCP Integration Sensible MCP on Google Gemini AI MCP Integration Sensible MCP on Lovable AI Development MCP Client Sensible MCP on Mistral AI Agents MCP Compatible Sensible MCP on Amazon AWS Bedrock MCP Support
MCP Servers — Included with Plan
Vinkius runs on Claude Code

Connect Sensible MCP to Claude Code

Create your Vinkius account to connect Sensible to Claude Code — we handle the hosting, security, and runtime updates so you don't have to. No server setup required.

GDPR Included with Plan

Key Capabilities

Headless document parsing pipelines via Claude Code MCP Server

The `extract_from_url` tool kicks off asynchronous data extraction for documents hosted online. Claude Code triggers this directly from your terminal, letting you process thousands of PDFs via simple shell scripts or cron jobs. To keep tabs on background tasks, Claude Code queries `list_extractions` with custom filters. You can pipe this output directly into grep or jq to monitor your queue without leaving the terminal.

Automated configuration versioning and deployments

The `list_configuration_versions` tool retrieves all draft and published versions of a parser configuration. Claude Code uses this to verify which layout rules are active before running terminal-based extraction tests. If you need to roll back a broken parser layout, Claude Code executes `delete_configuration_version` to unpublish the bad version. This gives you full control over your production configurations directly from your bash prompt.

Reference document management for golden datasets

The `list_goldens` tool retrieves all reference documents associated with a specific document type. Claude Code uses this MCP tool to audit your training data and ensure your parsers are calibrated against correct layouts. If a layout shifts, Claude Code calls `unassociate_golden` to remove outdated reference files. It then uploads fresh samples using `update_golden` to keep your parsing engine aligned with incoming document variations.

Setup guide

Set up Sensible MCP in Claude Code

Prerequisites

  • Claude Code CLI installed (npm install -g @anthropic-ai/claude-code)
  • Active Vinkius subscription with a valid endpoint token
  1. 1

    Run the add command

    Open your terminal and run the command shown on the right. Replace [YOUR_TOKEN_HERE] with your endpoint token from cloud.vinkius.com. Use --scope user to make it available across all projects.

  2. 2

    Verify the connection

    Start a Claude Code session and type /mcp to list connected servers. You should see sensible-mcp with a green status indicator.

  3. 3

    Start using tools

    Ask Claude Code something like "Check my latest Sensible transactions." It will automatically discover and invoke the available Sensible tools.

Terminal
claude mcp add --transport http sensible-mcp https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp

Why Choose Vinkius

Vinkius connects your tools to AI with real-time monitoring and automatic cost savings — all from one dashboard.

Real-time monitoring

Live

visibility into every interaction

Connect your favorite tools to your AI and see exactly what's happening — every request, every response, in real time.

Built-in savings

60%

lower AI costs

Vinkius compresses data between your apps and your AI automatically. Lower bills every month — no configuration required.

Single dashboard

One

place for every integration

Every tool your AI connects to, managed from a single screen. One account, complete control.

Common questions about Sensible MCP in Claude Code

Yes, you can pipe URLs into Claude Code and let it trigger `extract_from_url` for each file. It handles the asynchronous queue and outputs clean JSON that you can redirect to local files or database scripts.
You pass the failing document's ID to `get_document` to inspect the raw extraction errors. Claude Code displays the JSON output directly in your terminal so you can pinpoint layout mismatches instantly.
Yes, Claude Code can modify your extraction schemas using `update_document_type` and `update_configuration`. This lets you adjust target fields and layout rules without opening a web browser.
You run `generate_excel` through Claude Code to compile your JSON extraction outputs into a spreadsheet. The server writes the Excel file directly to your working directory, ready for immediate distribution.
All document text, metadata, and coordinates processed by this MCP Server are transmitted via encrypted HTTPS directly to the Sensible API. Vinkius runs the server in an ephemeral sandbox, leaving zero footprint of your files on disk after the command finishes.

Start using the Sensible MCP today

We host it, we monitor it, we maintain it. You just paste one token.

Built & Managed by Vinkius 30s setup 37 tools

We've already built the connector for Sensible. Just plug in your AI agents and start using Vinkius.

No hosting. No infrastructure. No complex setup.
All 37 tools are live and waiting. You're up and running in seconds.

Vinkius runs on Claude Claude
Vinkius runs on ChatGPT ChatGPT
Vinkius runs on Cursor Cursor
Vinkius runs on Gemini Gemini
Vinkius runs on Windsurf Windsurf
Vinkius runs on VS Code VS Code
Vinkius runs on JetBrains JetBrains
Vinkius runs on Vercel Vercel
+ other MCP clients

Vinkius gives your AI agents access to the full catalog of app connectors, all fully managed, secure, and enterprise-ready. One subscription, every tool you need.

Zero hosting required Full MCP catalog included Enterprise-grade security Auto-updated by Vinkius

Built, hosted, and secured by Vinkius. You just connect and go.