Actually FOSS open source quantconnect lean engine

This project aims to create a true, actual, non-commercial, community edition or free and open source version of QuantConnect lean. It does not require any subscriptions. It simply makes the beautiful QuantConnect Lean engine available with all the brokers in a local manner for you. Trading is a private affair and I know how important a local platform is for you. It was for me. I've followed Jared Broad over the years, who is the genius CEO and founder of QuantConnect, is actually part Austin Powers, part Dr. Evil and part Mini-me, a confused individual. He makes sure all his code is open source, however, the source code is unusable for the majority of the world as it is. He wants you to use his cloud platform and subscriptions despite the engine being open source. Therefore, there are projects like this that need to crop up. I know if I searched for 'actually open source lean' or 'truly open source lean' or 'madly open source lean' or 'deeply open source lean', others must have also and come up with nothing...still nothing..someone please SEO this project. This project aims to help those people, because there isn't such a project currently and this has worked so far for me in my personal trading algorithms. You can fork this, privatize it, use it to make money, whatever.

Some caveats though:

if you make money from this, please share some with me as I'm still in a 9 to 5 x 50 year rut
It's a very rudimentary setup with shell scripts and plain docker. Should toally be pythonized
No warranties..absolutely not. This is fully open source and usable as is for any purposes without warranties
If u have pull requests for nice features, I will integrate them in for the world.

My receiving wallets:

Ethereum 0xd576227aCc7d74259F66a8AFd379C7f36eCee41A (includes Ethereum compatible networks like Polygon, Arbitrum, OP Mainnet, Base, Blast, Linea and Avalanche)
Solana ArTNKfq3MWF3JfZXAquE5dhAQQGWiUVNw1ufkuV8WwAU
Bitcoin bc1q3guagg7zrh3ssw3vgw88jsrj8lulqfzf508u0x
Dogecoin DSRYJhMLzyiQajrSpZQgc2hfPwpc2gEqae

Documentation

Supported Operating Systems

this uses bash scripts and docker

Linux - Ubuntu is what I use, but this is written in bash and generic so, in theory, it should work on all flavors that have a bash shell.
macOS bash shell? - untested waters for me
Windows subsystem for Linux (WSL) - i was using Ubuntu here also
Windows NOT supported unfortunately - unless you can get it to work with the Windows bash shell and docker somehow. Yucghk

Step 0.0 - trading motivation

money is a problem that can be solved - forest gump after pwning apple stock
right approach for trading:
- idea-> Backtest->live paper trade->live trade small->live trade big.
Probability of profit*avg profit - probability of loss*avg loss=your average trade profit or loss (pnl). If this is greater than all the commissions and fees, u become a money printing casino. Most important equation in trading.
Never trade with what you are not willing to lose completely.
Limit the risk of unlimited loss potential and stick with limited loss potential trades(option spreads, crypto futures, buying margin rather than naked shorting)
pigs get fat, hogs get ...
...

Step 0.0.1 - STOP AND READ THIS

instead of doing all this, u can pay like USD $1080/mo on quantconnect.com for an institution seat to run locally or $10/mo to run on their cloud servers which might cost more - 60/mo or something not sure. idc because i like my free on-premise live trading institution grade version below
definitely review their pricing page: https://www.quantconnect.com/pricing u will have to click choose tier to see the options.
u'll still need to write ur own algo. The config stuff, they take care of for you with their 'open source' lean-CLI
also u'll still need to like install docker, set up vs code
but U won't need my dockerfile stuff or brokerage stuff. They have backtesting in their lean-cli which is also open source but I have not cracked it yet - too much subscription code in it as opposed to one line in Lean's Brokerage code lol.

Step 0: Train and learn lean engine algo-writing

There's a bootcamp And tons of beautiful, documentation on the engine online with youtube videos, a discord group where Jared and his AI Mia Khalisi personally help you, a class reference (just google for class u need), a slack group?, etc.....Too much stuff to help you become experts at writing Lean algorithms but not at executing them :') - just remember my wallet addresses please

Detailed Documentation

Step 0.1: Install stuff

Install vs code i'm using vs code in windows subsystem on linux. Ubuntu linux works great for this. You could use an ubuntu VM or other linux. Install python, pylance vscode extensions
Set up code autocomplete for vs code (OPTIONAL, but very very helpful)
- Create python environment somewhere on your machine and install these packages:
  - install miniconda. google search for downloading the ubuntu installer shell script and run it using bash Miniconda.sh
- start new terminal after conda installed
- Create conda env: conda create --name lean python==3.11
- conda activate lean
- install stubs autocomplete package using pip:
  - pip install quantconnect-stubs
- in vs code press ctrl-shift-p or cmd-shift-p and select python interpreter: .conda/envs/lean/python
Install docker desktop or engine - either one works
- look at docker desktop installation instructions for ubuntu or ur linux
Install git
- apt install git
git clone this repo to another repo of yours using an origin replace.
You want to create a static copy of this actually_open_lean GitHub repository, including its submodule, into a new repository without any ongoing synchronization. This means you want a one-time, independent copy. Here's how to achieve that:
- git clone https://github.com/vrindger/actually_open_lean --recurse-submodules
  - yes you will need to add your computer's key to your Github account to be able to recurse the submodule Lean. You don't need Step 3.2 that last step to force private key..
  - if you already cloned it without the --recurse-submodules, you can run this command anywhere inside the cloned repo git submodule update --init --recursive and it will pull Lean for you
- cd actually_open_lean
- Create a new PRIVATE repository inside your Github account. Go to Your Repositories -> New -> Name -> Create
- Note the repo url for your new repository
- cd actually_open_lean
- git remote add new_origin <new_repo_url>
- git push -u new_origin --all
- git push -u new_origin --tags
- git remote remove old_origin
- git remote rename new_origin origin That should be it for setting up your repo!! Lean submodule should automatically be in your new repo.

`lean_algos` Your main script directory = lean_algos/MyLeanAlgorithmTemplate

lean_algos is the directory where you can place your algo directory(s). Each directory will contain a set of artifacts and be for one algorithm that will be built from scripts within the MyLeanAlgorithmTemplate directory

Check the template dir MyLeanAlgorithmTemplate for a 'non-functional' python Lean example with some goodies I left in. At a minimum you will always need:

A QCAlgorithm class written in either .cs or .py - algo.py - this is your actual trading algorithm file which can be written in either python or C#. C# for ultimate speed I suppose. This is what is worth 1080$ / month !!!!!!!!! you simply write your algo and Lean/this project does the rest. Love you jaredbroad - sickkkk object oriented stuff
config.json file (this will specify import config items, mostly leave unchanged). Change it to add:
1. your brokerage API keys (yes, KEEP YOUR FORKEDREPOS PRIVATE KEEP YOUR REPOS PRIVATE KEEP YOUR REPOS PRIVATE KEEP YOUR REPOS PRIVATE KEEP YOUR REPOS PRIVATE KEEP YOUR REPOS PRIVATE KEEP YOUR REPOS PRIVATE KEEP YOUR REPOS PRIVATE KEEP YOUR REPOS PRIVATE KEEP YOUR REPOS PRIVATE KEEP YOUR REPOS PRIVATE KEEP YOUR REPOS PRIVATE KEEP YOUR REPOS PRIVATE KEEP YOUR REPOS PRIVATE KEEP YOUR REPOS PRIVATE!!)
```
"alpaca-api-key": "redacted",
"alpaca-api-secret": "redacted",
```
- algorithm-location: "../../../Algorithm.Python/algo.py" Yes you need the 3 dots because this will be executed from a dll inside Lean/Launcher/Debug
- algorithm-language - Algorithm language selector - options CSharp, Python
- algorithm-type-name - this is the name of the QCAlgorithm class you wrote in either C# or python as explained above.
- environment - The config for the environment that will get run. Search for live-alpaca environment later in the example file to check what settings are under that env.
Dockerfile - This builds the docker file. Please add whatever stuff you need in here. For brokerages, you will need to copy and paste my line to comment out ValidateSubscription() from the Brokerage code and other stuff I have in there. Other than that I've added AWS, my own projects, several brokerages etc. This example shows you one brokerage, but it should at least get you up to executing the algorithm script you put in algo.py.
run_algo.sh - This is the final file that you will run using bash run_algo.sh in a Linux shell to run the docker container that will run the lean algo. It contains cleanup code that cleans old containers with Lean executions that you canceled with Ctrl-C keyboard interrupts. If history is important, you will want to remove those.
For cloning brokerage projects in the Dockerfile, add GITHUB_USERNAME and GITHUB_PAT (your personal access token) as environment variables to your local environment. In linux you would use export GITHUB_USERNAME=vrindger etc...

That's it! Run your algorithm by going into it's directory and executing bash run_algo.sh! You will see the magical Lean get instantiated locally if you did everything right and pay me at all my crypto wallets above.

Step 0.2: Decide what u want to trade and pick brokerages

Think of a trading strategy u think will work.
Which instruments does it trade? stocks, options, futures, future options, spot crypto, derivative crypto
Which brokerage do u want to use? https://www.quantconnect.com/docs/v2/cloud-platform/live-trading/brokerages
Identify lean brokerage github open source project: https://github.com/QuantConnect/ e.g. https://github.com/QuantConnect/Lean.Brokerages.Alpaca

Step 1 - Create your own Dockerfile for running your engine

- some things like which brokerage are set here. i added common ones i like

Look at the Dockerfile being used in my project such as: https://github.com/vrindger/lean/blob/main/Dockerfile_dev_debug

u might want to use a different brokerage than the ones I've used, e.g:

 RUN git clone https://github.com/QuantConnect/Lean.Brokerages.Binance $BINANCE_DIR && \
    cp -rf $BINANCE_DIR/QuantConnect.BinanceBrokerage $LEAN_DIR && \
    cp -rf $BINANCE_DIR/QuantConnect.BinanceBrokerage.ToolBox $LEAN_DIR

This is the secret sauce to this project hahaha - commenting out one line in the brokerage code. Hope you don't change the name to that function @jaredbroad lol:

# Disable QuantConnect license check
RUN sed -i 's/^\s*ValidateSubscription();/\/\/&/' $LEAN_DIR/QuantConnect.AlpacaBrokerage/AlpacaBrokerage.cs

Step 2 - Create your own trading algo! U can copy mine but use your own brokerage keys etc

Create a new dir in the lean_algos dir with your own config file and an algo.py or algo.cs file for C#.
To create algo, look at the template dir under lean_algos
MyLeanAlgorithm is my example algorithm using my config file(a config.json) which used to have my secret keys for my brokerage's paper account. The algorithm file is a .py file.
Yes, you will need every single file - the json file, an algo file, Dockerfile AND the run_algo.sh bash file. You can also split up build and run from the run_algo.sh file if you need to.

Step 3 - Run bash script

Before running the bash script, you will need to set environment variables for Git because you will be cloning any number of Brokerage projects for your project. So you can add the following to your ~/.bashrc file in linux:

export GITHUB_USERNAME=vrindger
export GITHUB_PAT=yourgithubpersonalaccesstokenhere

bash run_algo.sh or inside another algo scratch, you would run bash scratch.sh

Step 4 - Lean is running

If you see this output, this is Hello World for this project = you are good to go - you will need to add your API keys in your config and take it from here yourself:

20250320 11:29:05.708 TRACE:: AlpacaBrokerage.Initialize(): Option failed to connect to live feed 'opra', will retry with free feed
20250320 11:29:05.716 TRACE:: AlpacaStreamingClientWrapper.ConnectAndAuthenticateAsync(Option): try connecting free feed
20250320 11:29:05.716 TRACE:: StreamingClient_OnWarning(AlpacaStreamingClientWrapper): AlpacaOptionsStreamingClient set content type to: application/msgpack
20250320 11:29:05.850 TRACE:: StreamingClient_SocketOpened(AlpacaStreamingClientWrapper): SocketOpened
20250320 11:29:05.894 TRACE:: StreamingClient_Connected(AlpacaStreamingClientWrapper): Unauthorized
20250320 11:29:05.894 TRACE:: AlpacaBrokerage.Initialize(): Option failed to connect to live feed 'indicative'
20250320 11:29:05.909 ERROR:: Engine.Run():  System.InvalidOperationException: Connect(): Failed to connect to AlpacaStreamingClientWrapper
   at QuantConnect.Brokerages.Alpaca.AlpacaBrokerage.Connect() in QuantConnect.AlpacaBrokerage/AlpacaBrokerage.cs:line 628
   at QuantConnect.Brokerages.Alpaca.AlpacaBrokerage.SetJob(LiveNodePacket job) in QuantConnect.AlpacaBrokerage/AlpacaBrokerage.DataQueueHandler.cs:line 78
   at QuantConnect.Lean.Engine.DataFeeds.DataQueueHandlerManager.SetJob(LiveNodePacket job) in Engine/DataFeeds/DataQueueHandlerManager.cs:line 182
   at QuantConnect.Lean.Engine.DataFeeds.LiveTradingDataFeed.Initialize(IAlgorithm algorithm, AlgorithmNodePacket job, IResultHandler resultHandler, IMapFileProvider mapFileProvider, IFactorFileProvider factorFileProvider, IDataProvider dataProvider, IDataFeedSubscriptionManager subscriptionManager, IDataFeedTimeProvider dataFeedTimeProvider, IDataChannelProvider dataChannelProvider) in Engine/DataFeeds/LiveTradingDataFeed.cs:line 103
   at QuantConnect.Lean.Engine.Engine.Run(AlgorithmNodePacket job, AlgorithmManager manager, String assemblyPath, WorkerThread workerThread) in Engine/Engine.cs:line 177
20250320 11:29:05.910 TRACE:: JOB HANDLERS:
         DataFeed:             QuantConnect.Lean.Engine.DataFeeds.LiveTradingDataFeed
         Setup:                QuantConnect.Lean.Engine.Setup.BrokerageSetupHandler
         RealTime:             QuantConnect.Lean.Engine.RealTime.LiveTradingRealTimeHandler
         Results:              QuantConnect.Lean.Engine.Results.LiveTradingResultHandler
         Transactions:         QuantConnect.Lean.Engine.TransactionHandlers.BrokerageTransactionHandler
         Object Store:         QuantConnect.Lean.Engine.Storage.LocalObjectStore
         History Provider:     
         Brokerage:            QuantConnect.Brokerages.Alpaca.AlpacaBrokerage
         Data Provider:        QuantConnect.Lean.Engine.DataFeeds.DefaultDataProvider

20250320 11:29:05.911 TRACE:: StopSafely(): Waiting for 'RealTimeScheduleEventService' thread to stop...
20250320 11:29:05.912 TRACE:: StopSafely(): Waiting for 'Result Thread' thread to stop...
20250320 11:29:05.913 ERROR:: Algorithm.Initialize() Error: Connect(): Failed to connect to AlpacaStreamingClientWrapper Stack Trace: Connect(): Failed to connect to AlpacaStreamingClientWrapper
 Connect(): Failed to connect to AlpacaStreamingClientWrapper
20250320 11:29:05.913 TRACE:: LiveTradingResultHandler.Run(): Ending Thread...
20250320 11:29:05.914 TRACE:: LiveTradingResultHandler.SendFinalResult(): Starting...
20250320 11:29:05.964 TRACE:: LiveTradingResultHandler.SendFinalResult(): Finished storing results. Start sending...
20250320 11:29:05.964 TRACE:: LiveTradingResultHandler.SendFinalResult(): Ended
20250320 11:29:05.964 TRACE:: Engine.Run(): Disconnecting from brokerage...
20250320 11:29:05.966 ERROR:: Engine.Run(): Error running algorithm System.NullReferenceException: Object reference not set to an instance of an object.
   at QuantConnect.Extensions.SynchronouslyAwaitTask(Task task) in Common/Extensions.cs:line 3338
   at QuantConnect.Brokerages.Alpaca.AlpacaBrokerage.Disconnect() in QuantConnect.AlpacaBrokerage/AlpacaBrokerage.cs:line 669
   at QuantConnect.Lean.Engine.Engine.Run(AlgorithmNodePacket job, AlgorithmManager manager, String assemblyPath, WorkerThread workerThread) in Engine/Engine.cs:line 426
Engine.Main(): Analysis Complete.
20250320 11:29:05.967 TRACE:: Config.GetValue(): close-automatically - Using default value: False
Engine.Main(): Press any key to continue.

Random important notes about Lean:

there's a string ticker for instruments like 'BTCUSDT' or 'GOOGL'.
then there's a Lean Symbol object created from that ticker if u specify US stock like so: Market Symbol.create('SPY', SecurityType.EQUITY, Market.USA). symbols don't contain very much info.
then there's a Security that is added to the algorithm which contains the symbol and even more information like exchange and hours etc. It's the active Security that gets traded
In Lean, methods only (?not classes?, ManualUniverseSelectionModel) are automatically converted from camel case(C#) to snake case:
- Snake case methods, and CamelCase classes:
```
self.AddUniverseSelection(ManualUniverseSelectionModel(symbols)) 
self.add_universe_selection(ManualUniverseSelectionModel(symbols))
```

SEO

General Open Source & QuantConnect: foss foss foss foss fossest foss free and open source free open source free source open source QuantConnect LEAN QuantConnect LEAN open source fork LEAN trading engine open source open source algorithmic trading platform LEAN free open source QuantConnect alternative QuantConnect LEAN community edition open source backtesting engine LEAN LEAN open source trading framework open source quant trading LEAN LEAN engine github open source

Specific LEAN & Project Features:

fully open source QuantConnect LEAN
truly open source LEAN engine
unrestricted open source LEAN
no limitations open source LEAN
complete open source LEAN
LEAN engine source code access
LEAN open source contributions
modified QuantConnect LEAN open source
enhanced LEAN trading engine open source
extended LEAN framework open source

GitHub & Repository Focused:

LEAN engine github repository
open source QuantConnect LEAN github
github LEAN trading platform
LEAN engine open source code github
search github LEAN open source
LEAN open source project github
LEAN engine fork github
github open source trading algorithm LEAN
LEAN engine open source examples github
github LEAN open source documentation

Problem Solving & Alternatives:

QuantConnect LEAN limitations open source
bypass QuantConnect restrictions open source
alternative to QuantConnect LEAN open source
free LEAN backtesting open source
open source LEAN without cloud
local LEAN engine open source
self-hosted LEAN open source
run LEAN locally open source
QuantConnect LEAN cost open source
solve LEAN limitations open source

Technical & Development:

LEAN engine source code download
compile LEAN engine open source
modify LEAN engine source
build LEAN engine from source
LEAN engine development open source
contribute to LEAN engine
LEAN engine open source API
LEAN engine open source plugins
extend LEAN engine functionality
LEAN engine open source libraries

Keywords + Specific Terms:

"actually open source" LEAN
"fully unlocked" LEAN engine
"community maintained" LEAN
"completely free" LEAN trading
"no cloud required" LEAN
LEAN engine "local execution"
open source "algorithmic trading" LEAN
LEAN "backtesting" open source
"quant trading" LEAN github
LEAN engine "source code" access

Long-Tail & Specific Queries:

how to run QuantConnect LEAN locally open source
where to find open source QuantConnect LEAN on github
best open source alternative to QuantConnect LEAN
LEAN engine open source with no API restrictions
detailed documentation for open source LEAN engine
community support for open source LEAN trading
how to contribute to open source LEAN engine development
LEAN engine open source with full source code access
open source LEAN engine for advanced trading strategies
step-by-step guide to compiling open source LEAN engine

Targeting Specific User Needs:

open source LEAN for professional traders
LEAN engine open source for researchers
open source LEAN for students
LEAN engine open source for hobbyists
open source LEAN for quantitative analysts
LEAN engine open source for developers
LEAN open source for automated trading
open source LEAN for backtesting strategies
LEAN open source for custom indicators
LEAN open source for market data analysis

Variations & Misspellings:

quant connect open source
lean engine open source code
open source lean platform
quant connect free
lean trading engine free
quant connect community
opensource lean trading
lean source code
quantconnect github
lean github

Adding Descriptive Terms:

robust open source LEAN engine
scalable open source LEAN trading
customizable open source LEAN platform
powerful open source LEAN framework
flexible open source LEAN engine
reliable open source LEAN backtesting
efficient open source LEAN trading
modern open source LEAN engine
comprehensive open source LEAN documentation 100.active open source LEAN community

Name		Name	Last commit message	Last commit date
Latest commit History 28 Commits
base		base
lean_algos		lean_algos
tools		tools
.gitignore		.gitignore
.gitmodules		.gitmodules
README.md		README.md
todo.txt		todo.txt

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Repository files navigation

Actually FOSS open source quantconnect lean engine

My receiving wallets:

Documentation

Supported Operating Systems

Step 0.0 - trading motivation

Step 0.0.1 - STOP AND READ THIS

Step 0: Train and learn lean engine algo-writing

Detailed Documentation

Step 0.1: Install stuff

`lean_algos` Your main script directory = lean_algos/MyLeanAlgorithmTemplate

Step 0.2: Decide what u want to trade and pick brokerages

Step 1 - Create your own Dockerfile for running your engine

- some things like which brokerage are set here. i added common ones i like

Step 2 - Create your own trading algo! U can copy mine but use your own brokerage keys etc

Step 3 - Run bash script

Step 4 - Lean is running

Random important notes about Lean:

SEO

About

Uh oh!

Releases

Packages

Languages

vrindger/actually_open_lean

Folders and files

Latest commit

History

Repository files navigation

Actually FOSS open source quantconnect lean engine

My receiving wallets:

Documentation

Supported Operating Systems

Step 0.0 - trading motivation

Step 0.0.1 - STOP AND READ THIS

Step 0: Train and learn lean engine algo-writing

Detailed Documentation

Step 0.1: Install stuff

lean_algos Your main script directory = lean_algos/MyLeanAlgorithmTemplate

Step 0.2: Decide what u want to trade and pick brokerages

Step 1 - Create your own Dockerfile for running your engine

- some things like which brokerage are set here. i added common ones i like

Step 2 - Create your own trading algo! U can copy mine but use your own brokerage keys etc

Step 3 - Run bash script

Step 4 - Lean is running

Random important notes about Lean:

SEO

About

Topics

Resources

Uh oh!

Stars

Watchers

Forks

Releases

Packages 0

Languages

`lean_algos` Your main script directory = lean_algos/MyLeanAlgorithmTemplate

Packages