Aframe MCP Connector
Enables Claude to create and manage transactions in Aframe via the Aframe Open API, supporting transaction creation and note addition.
README
Aframe MCP Connector
A Node.js MCP server that bridges Claude sessions to the Aframe Open API. Deployed on Railway, exposed to Claude.ai as a custom remote MCP connector.
Status
v0.2.0 — POC validated, expanded to core transaction CRUD + custom fields.
Tools
| Tool | Description |
|---|---|
create_transaction |
Create a new transaction in Aframe |
get_transaction |
Fetch an existing transaction by ID (full payload incl. custom fields) |
update_transaction |
PATCH transaction fields (partial update via JSON Patch under the hood) |
update_custom_field |
PATCH a custom/merge field by Merge Field Code (e.g. f_EarnestMoney) |
add_transaction_note |
Add an activity note to an existing transaction |
Architecture
Claude.ai session
↓ (custom MCP connector — Streamable HTTP)
Railway service (this repo)
↓ (HTTPS with X-AFrame-API-Key header)
Aframe Open API
Local development
npm install- Copy
.env.exampleto.envand fill inAFRAME_API_KEY npm start- Verify health:
curl http://localhost:3000/→{"status":"ok",...}
Deployment (Railway)
Auto-deploys on push to main. Required environment variable:
AFRAME_API_KEY— your Aframe API key
Generate a public domain under Settings → Networking after the first deploy.
Connecting to Claude.ai
Settings → Connectors → Add custom connector → paste https://<railway-url>/mcp → Add.
Tools become available in any new conversation with the connector enabled.
Implementation notes
Aframe response envelope: { payload, error }. The error block can appear alongside payload on 2xx success responses (as validation warnings, e.g. "defaulted to current AppUser"). HTTP status is the authoritative success/failure signal — error on 2xx is surfaced as warnings, not thrown.
JSON Patch (RFC 6902): Aframe's PATCH endpoints (update_transaction, update_custom_field) require an array of patch operations with Content-Type: application/json-patch+json. The MCP tools accept friendly flat-object inputs (e.g. { closingDate: "2026-08-15" }) and the client internally constructs the patch array. Callers never need to write RFC 6902 syntax.
Custom fields: Stored as a flat xactionFieldData map keyed by Merge Field Code (e.g. f_EarnestMoney). All values are strings regardless of underlying type — Aframe handles type coercion server-side.
Auth note
This service runs authless. Acceptable for development since the Railway subdomain is effectively unguessable, but not suitable for production. Claude.ai's custom connector UI does not currently support pasted bearer tokens (per the connector auth docs); production use will require OAuth 2.1 with PKCE.
Version history
- v0.2.0 — Added
get_transaction,update_transaction,update_custom_field. Fixed response envelope parsing to treaterroron 2xx as warnings rather than failures. JSON Patch helper for PATCH endpoints. - v0.1.0 — Initial POC with
create_transactionandadd_transaction_note.
Reference
- Aframe Open API: https://api.aframeonline.com/api-pub/swagger-ui/index.html
- MCP spec: https://modelcontextprotocol.io
- Claude custom connectors: https://support.claude.com/en/articles/11175166-get-started-with-custom-connectors-using-remote-mcp
- JSON Patch (RFC 6902): https://datatracker.ietf.org/doc/html/rfc6902
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.