Skip to main content
Glama
Reamer Research MCP server

Reamer Research MCP server

by reamerlabs

Reamer Research MCP server

An MCP server and Claude Code plugin for Reamer Research, a backtesting and live trading engine for Python and C++. Your AI assistant writes a strategy file, backtests it on your bars, reads every closed trade, runs Monte Carlo and opens the run in visual trade diagnostics, all on your own machine.

Works with Claude Code, Claude Desktop, Cursor and any client that speaks MCP over stdio.

Tools

Tool

What it does

engine_status

Finds the installed engine: version, default settings, Evaluation Mode limits.

api_reference

The strategy API, order functions and every setting, for writing a strategy.

backtest

Runs a strategy file on CSV or .bin bars and returns the summary metrics and a run_id.

closed_trades

Every round trip of a run, sorted and paged.

monte_carlo

Resamples a run's per-trade returns into loss, ruin and drawdown statistics.

printed_output

Every value the strategy printed, stamped with its bar's time.

open_trade_diagnostics

Opens the run in visual trade diagnostics: each trade on its own bars, beside every value the strategy printed.

Each backtest runs in its own process under the Python that has the engine, so anything a strategy prints stays out of the MCP connection. Runs are stored in ~/.cache/reamer-research-mcp/runs/ (REAMER_MCP_RUNS moves it), each with the full JSON report and the trade diagnostics page.

Related MCP server: OpenFinClaw CLI

Install

1. The engine. Download the Reamer Research kit (free, no key, Linux x86-64, macOS on Apple silicon, Windows through WSL2) from reamerlabs.com/download, then:

tar xzf research-kit-*.tar.gz
pip install research-kit-*/python/reamer_research-*.whl

2. The server, in one of these ways.

Claude Code plugin, which also adds a Reamer Research skill (needs uv):

claude plugin marketplace add reamerlabs/reamer-research-mcp
claude plugin install reamer-research@reamerlabs

Claude Code, server only:

claude mcp add reamer-research -- uvx reamer-research-mcp

Claude Desktop, Cursor and other clients, in the client's MCP config:

{
  "mcpServers": {
    "reamer-research": {
      "command": "uvx",
      "args": ["reamer-research-mcp"]
    }
  }
}

Without uv, pip install reamer-research-mcp and use reamer-research-mcp as the command.

The server finds the engine in REAMER_PYTHON if set, else in its own Python, else in python3 on PATH. If you installed the wheel into a virtual environment, point REAMER_PYTHON at it:

"env": { "REAMER_PYTHON": "/home/you/venv/bin/python" }

The server needs Python 3.10 or later. The engine runs on Python 3.9 or later.

Try it

Ask your assistant:

Write an ATR band reversion strategy for Reamer Research, backtest it on research-kit-4.11.3-linux_x86_64/python/sample_data/SAMPLE_5min.csv with a 0.005 commission and a 0.01 spread, show me the five worst trades and run Monte Carlo.

Evaluation Mode

Without a licence, each backtest runs on one ticker over at most 30 days or 50,000 bars, whichever comes first, and live trading is off. A run past a limit returns an error that names the limit. A Reamer Research licence is $20 per seat per month and removes every limit; the same strategy files run unchanged.

Licence

This server, the plugin and the skill are under the MIT licence. The engine is commercial software under the Reamer Labs Software License Agreement, downloaded separately from reamerlabs.com/download.

Support: support@reamerlabs.com

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      A
      quality
      B
      maintenance
      MCP server for the FLOX trading framework. About 30 tools to run backtests, scaffold strategies, validate for lookahead bias, compute indicators, place orders, and query PnL from Claude/Cursor.
      38
      945 PyPI
      231
      MIT
    • A
      license
      B
      quality
      D
      maintenance
      Local-first MCP server for building and running reproducible Backtrader strategies offline, with tools to inspect datasets, manage drafts, and execute backtests via controlled subprocess runs. It supports multiple hosts including Claude, Codex, OpenCode, and OpenClaw through stdio.
      30
      1
      MIT
    • A
      license
      Not graded
      quality
      C
      maintenance
      Enables AI agents to run a full quant research workflow over MCP: pulling data, authoring and backtesting strategies, running statistical validation and risk checks, and recording findings for future sessions.
      1
      MIT