Ramp

ramp_mcp

python

30

GitHub Stars

python

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "ramp-public-ramp_mcp": {
      "command": "uv",
      "args": [
        "run",
        "ramp-mcp",
        "-s",
        "transactions:read,reimbursements:read"
      ],
      "env": {
        "RAMP_ENV": "<demo|prd>",
        "RAMP_CLIENT_ID": "<CLIENT_ID>",
        "RAMP_CLIENT_SECRET": "<CLIENT_SECRET>"
      }
    }
  }
}

You run the Ramp MCP server to enable a local ETL pipeline with an in-memory database for rapid data analysis and task execution. This server retrieves data, loads it into a transient in-memory database, and exposes tools to process, query, fetch, and analyze information using Ramp client credentials and scoped access.

How to use

Start the MCP server from your command line using your Ramp client credentials and a set of scopes that match the tools you want to enable. The server runs as a local process that your MCP client can talk to and uses an ephemeral in-memory database for analysis.

How to install

Prerequisites you need before starting: install the runtime that runs this MCP server, and have Ramp client credentials ready.

Clone the MCP server repository.
Install the runtime you will use to run the MCP server.
Prepare Ramp client credentials and the scopes you intend to enable.
Start the MCP server using your credentials and the desired scopes.

Configuration and usage notes

To run the server with a concrete set of scopes, execute the following command structure, filling in your actual client credentials and desired scopes.

Security and environment considerations

Keep your Ramp client ID and client secret secure and do not expose them in shared environments. Set the Ramp environment to either demo or prd as appropriate for your testing or production needs.

Examples

RAMP_CLIENT_ID=... RAMP_CLIENT_SECRET=... RAMP_ENV=<demo|prd> uv run ramp-mcp -s <COMMA-SEPARATED-SCOPES>

Available tools

process_data

Process in-memory data as part of the ETL workflow to prepare data for analysis by the LLM.

execute_query

Run queries against the ephemeral in-memory database to extract insights or verify results.

clear_table

Clear data from an in-memory table to reset analysis contexts.

get_ramp_categories

Fetch available Ramp data categories for selective loading and analysis.

get_currencies

Retrieve currency information for financial datasets.

load_transactions

Load transactions data into the ephemeral database with read scope transactions:read.

load_reimbursements

Load reimbursements data into the ephemeral database with read scope reimbursements:read.

load_bills

Load bills data into the ephemeral database with read scope bills:read.

load_locations

Load locations data into the ephemeral database with read scope locations:read.

load_departments

Load departments data into the ephemeral database with read scope departments:read.

load_bank_accounts

Load bank accounts into the ephemeral database with read scope bank_accounts:read.

load_vendors

Load vendors into the ephemeral database with read scope vendors:read.

load_vendor_bank_accounts

Load vendor bank accounts into the ephemeral database with read scope vendors:read.

load_entities

Load entities into the ephemeral database with read scope entities:read.

load_spend_limits

Load spend limits into the ephemeral database with read scope limits:read.

load_spend_programs

Load spend programs into the ephemeral database with read scope spend_programs:read.

load_users

Load users into the ephemeral database with read scope users:read.