Redix AnyToAny MCP Server
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@Redix AnyToAny MCP ServerConvert this X12 837 claim to FHIR R4 bundle."
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Redix AnyToAny MCP Server
A developer sandbox for the Redix AnyToAny healthcare data conversion engine, exposed to AI clients (Claude Desktop, Claude Code, Cursor, and any MCP-compatible agent) over the Model Context Protocol.
Point your AI client at this server and ask it, in plain language, to validate an X12 claim, turn an 837 into a FHIR R4 Bundle, render a CMS-1500 PDF, or map an HL7 v2 message - and watch the Redix engine do it, with validation gates that block bad conversions instead of silently passing them.
What this is (and is not)
This server is an interactive evaluation and debugging surface for the Redix conversion engine. It is the fastest way to see what the engine does to your data before you write a line of integration code.
It is not a production pipeline. Healthcare conversion at volume runs through the deterministic Redix REST API (the same engine this server wraps), under a Redix license. When you have seen enough here, that is where you go next:
REST API and tools: https://demo.redix.com
Production licensing: ppc@redix.com
Related MCP server: Medplum MCP Server
Tools (11)
Tool | Description | Validation gate |
| Validate HIPAA X12 against 5010 implementation-guide rules | is the gate |
| X12 -> FHIR R4 Bundle | Gate 1 (input) |
| X12 -> RMap v5 intermediate | Gate 1 (input) |
| RMap v5 -> X12 | Gate 5 (output) |
| X12 -> relational DB tables | Gate 1 (input) |
| DB -> X12 | Gate 5 (output) |
| HL7 v2 -> FHIR R4 | - |
| CDA / C-CDA -> FHIR R4 | - |
| FHIR -> RMap v5 (inspect the intermediate mapping) | - |
| X12 837 -> PDF (CMS-1500 / UB-04 / ADA J400) | Gate 1 (input) |
| Capability discovery (no key needed) | - |
Reverse FHIR-to-X12 (278) is intentionally not exposed in this sandbox. External, provider-generated PAS bundles do not yet ingest cleanly end to end, so it is unfit for unattended use where a failure would look like an engine defect. It remains available through the REST API under a short private technical review - email ppc@redix.com.
Quick start - hosted (no install)
Point your MCP client at the hosted demo server. It runs with a sample-only demo key, so you can try every tool against Redix sample files immediately.
Streamable HTTP (recommended)
{
"mcpServers": {
"redix-anytoany": {
"url": "https://demo.redix.com/mcp",
"env": { "REDIX_API_KEY": "demo-key-12345" }
}
}
}SSE
{
"mcpServers": {
"redix-anytoany": {
"url": "https://demo.redix.com/sse",
"env": { "REDIX_API_KEY": "demo-key-12345" }
}
}
}Evaluating with your own data
The published demo-key-12345 is sample-only: it converts Redix sample files so you can see each tool work, but it will not process your own files. This protects healthcare data and keeps the demo a controlled technical evaluation.
When you send your own data, a tool returns a clear EVAL_KEY_REQUIRED response telling you exactly where to request a free, tool-scoped evaluation key - for example, the X12 validator points you to https://demo.redix.com/tools/validate. Submit the short "request an evaluation key" form on that page (or email ppc@redix.com); Redix issues a scoped key by email, usually for a 7-day evaluation. Then set REDIX_API_KEY to that key and rerun.
Use test or de-identified data for evaluation. Do not upload production PHI to the demo server.
Setup - local (self-hosted)
pip install -r requirements.txtEnvironment variables
Variable | Default | Description |
|
| Redix API base URL |
|
| API authentication key |
|
| HTTP timeout in seconds |
|
| Python logging level |
Running locally
# STDIO mode (Claude Desktop / Claude Code)
python3.11 server.py
# Streamable HTTP mode (multi-client)
fastmcp run server.py --transport streamable-http --port 8001
# SSE mode (multi-client)
fastmcp run server.py --transport sse --port 8000
# MCP Inspector (browser UI)
fastmcp dev server.pyClaude Code integration
Copy .mcp.json to your project root, or add to Claude Code settings:
{
"mcpServers": {
"redix-anytoany": {
"command": "python3.11",
"args": ["server.py"],
"cwd": "/opt/MCP/redix-mcp-server",
"env": {
"REDIX_API_BASE": "http://localhost:8080",
"REDIX_API_KEY": "demo-key-12345"
}
}
}
}Claude Desktop integration
Add to ~/.claude/claude_desktop_config.json:
{
"mcpServers": {
"redix-anytoany": {
"command": "python3.11",
"args": ["/opt/MCP/redix-mcp-server/server.py"],
"env": {
"REDIX_API_BASE": "http://localhost:8080",
"REDIX_API_KEY": "demo-key-12345"
}
}
}
}How it works
Each tool is a thin wrapper over the Redix AnyToAny REST API. Conversions are bracketed by validation gates:
Gate 1 (input) validates X12 against 5010 rules before converting. A malformed claim is refused, not silently mangled.
Gate 5 (output) re-validates X12 that a tool generates, so you never get back EDI that would bounce at a payer.
Transaction types are auto-detected from content when you do not specify them.
Testing
python3.11 tests/test_tools.pyRequires the Redix API running at REDIX_API_BASE. Note: against the hosted demo (sample-only key, 3-second rate limit) many cases return EVAL_KEY_REQUIRED or rate-limit errors by design; run the suite against a local API with an evaluation key for full coverage.
This server cannot be deployed
Maintenance
Related MCP Connectors
OCR, transcription, file extraction, and image generation for AI agents via MCP.
Convert files between 110+ document, image, audio, video, archive and ebook formats from AI agents.
Connect AI agents to Replynodes over the Model Context Protocol.
Connect AI clients to biomedical data and tools.
Related MCP Servers
- AlicenseNot gradedqualityDmaintenanceIntegrates Taiwan-specific medical data including ICD-10 codes, FDA drug databases, and nutrition information into the Model Context Protocol. It enables AI models to query clinical guidelines, verify medical codes, and convert health data into FHIR R4 standardized formats.MIT

Medplum MCP Serverofficial
FlicenseNot gradedqualityFmaintenanceEnables interaction with Medplum's FHIR data through the Model Context Protocol, allowing LLMs to access and manage healthcare data.5-- AlicenseNot gradedqualityCmaintenanceFile conversion built for AI agents. CLI, REST API, and MCP server — all sharing one engine.5 npm1MIT
- AlicenseNot gradedqualityBmaintenanceEnables standardized PACS-AI integration through the Model Context Protocol, providing medical imaging workflows with DICOMweb transport, hierarchical PHI de-identification, and AI model orchestration.Apache 2.0