New · Linux server edition

Your research agent.
Always online.

Run the full RLXBT Rust engine headlessly on Ubuntu. Your agent can research, backtest, stress-test and evaluate live strategy signals 24/7 — without a desktop or Python runtime.

Docker · Ubuntu x86_64 & ARM64 · Same validated engine · Your data stays on your server

ubuntu · rlxbt-server

$ docker compose up -d

Pulling ghcr.io/sergio12s/rlxbt-server:0.2.7

license verified

headless API listening on 127.0.0.1:8142

agentwalk_forwardrobust

agentevaluate_signalenter_long

Same compiled strategy path in backtest and live evaluation.

One engine, two modes

Everything except the pixels.

The server edition removes the visual dashboard, not the research engine. Agents use the same backtesting, validation, reports and live-signal APIs as the desktop product.

🧪

Research & backtest

Load datasets, build strategy ideas, run parameter sweeps and keep curated reports through the headless API.

🛡️

Prove robustness

Run walk-forward, Monte Carlo, sensitivity and out-of-sample validation continuously from your agent.

📡

Evaluate live signals

Send each completed bar with features and position state; receive enter, exit or hold with absolute TP/SL levels.

Clean separation

RLXBT decides. Your agent executes.

RLXBT never stores exchange credentials or places orders. The execution agent owns market data, positions, risk limits and exchange connectivity; RLXBT supplies backtest-identical strategy decisions.

📈
Exchange
🤖
Your agent
RLXBT

Ubuntu deployment

Online in three steps.

Requires Docker Compose and a valid RLXBT license. No Python toolchain is installed.

terminal
mkdir rlxbt-server && cd rlxbt-server
curl -fsSLO https://rlxbt.com/downloads/rlxbt-server-compose.yml
curl -fsSLo .env https://rlxbt.com/downloads/rlxbt-server.env.example
chmod 600 .env
# Edit .env: add RLX_LICENSE_KEY and a random RLXBT_API_KEY
docker compose -f rlxbt-server-compose.yml up -d
curl http://127.0.0.1:8142/api/health
01

Download

Fetch the official Compose and environment templates from rlxbt.com.

02

Activate

Add your RLXBT license and generate a private API key in .env.

03

Connect

Start Docker and point your agent at the private RLXBT endpoint.

Production security: the supplied Compose file binds RLXBT to 127.0.0.1. Use a VPN, SSH tunnel, or authenticated TLS reverse proxy for remote access. Do not expose port 8142 directly to the public internet.

Agent onboarding

One URL teaches the agent.

Give any coding or trading agent the canonical guide. It explains installation, MCP and REST discovery, research workflow, live-signal state, security boundaries, upgrades and rollback.

Discovers the running engine instead of guessing tools or schemas.
Can research, backtest, validate, curate reports and evaluate live signals.
Knows RLXBT never holds exchange credentials or executes orders.
Keeps API keys out of chat and port 8142 off the public internet.
prompt for your agent
Read https://rlxbt.com/server-agent.md and follow it as the canonical RLXBT Server contract.
Use secrets only from environment configuration; never ask me to paste license, API, or exchange keys into chat.
First check /api/health, then discover /api/endpoints, /api/ai-instructions, and /api/strategy-schema.
RLXBT researches and decides; it never places exchange orders. Keep execution state, risk limits, idempotency, and the kill switch outside RLXBT.

Machine discovery is also published in /llms.txt and /ai.md. Secrets belong in environment configuration, never in the prompt.