Skip to content

Create Publication

We are looking for publications that demonstrate building dApps or smart contracts!
See the full list of Gitcoin bounties that are eligible for rewards.


Beaker is a framework for building Smart Contracts using PyTeal. Beaker is designed to simplify writing, testing and deploying Algorand smart contracts. The Beaker source code available on github.

This page provides an overview of the features available in Beaker. For complete details see the Beaker's documentation.

High Level Overview

Beaker provides several packages that extend PyTeal and provide convenience functionality for testing and deploying smart contracts.

The Application class is Beaker's primary class. It is used to create ABI compliant Algorand smart contracts. Beaker also provides decorators to route specific application transactions to the proper functionality within a smart contract. Beaker facilitates management of local and global state, and box storage.

The ApplicationSpecification class is used to generate a JSON manifest that describes the contract methods, source, and state schema used. This manifest can be used by other modules and utilities to deploy the smart contract.

The ApplicationClient class can be used to connect to an Algorand node and interact with a specific Application.

Beaker's sandbox module can be used to quickly connect to the default docker sandbox installation to deploy and call a contract.


Beaker can be installed using pip package manager.

pip install beaker-pyteal

Alternatively, Beaker can be installed with AlgoKit using the beaker project template.

algokit init --template beaker

Either of these methods will also install PyTeal in addition to Beaker.


Beaker requires python version 3.10 or higher

Initialize Application

To create an application simply initialize a new Beaker Application object, supplying the name and description.

from beaker import Application

app = Application("MyRadApp", descr="This is a rad app")
Snippet Source

This is enough to generate the ApplicationSpecification that can be exported for use by other tools. This spec can be generated using the Application build method. Optionally you can use the export method to export the approval and clear TEAL programs, the ABI contract manifest, and the application specification.


At this point a complete smart contract has been written, albeit with no utility.

app_spec =
Snippet Source

Add Method Handlers

Method handlers can be added to provide functionality within the smart contract. This can be accomplished using the external decorator or by using a blueprint.


To provide a method that can be invoked by an application call transaction, Beaker provides the external decorator. This instructs the framework to expose the method publicly for incoming transactions. The method is then defined with its required method signature, where the parameter types in the method signature describe the input types and output type. These types of the arguments must be valid ABI data types, using PyTeal's ABI package. Arguments and output types are optional, omitting any arguments is perfectly valid.


Note that input types come first, and if a value is returned it should be denoted in the method signature at the end using the notation *, output: abi.ValidABIType, which provides a variable to write the output into.

import pyteal as pt

# use the decorator provided on the `app` object to register a handler
def add(a: pt.abi.Uint64, b: pt.abi.Uint64, *, output: pt.abi.Uint64) -> pt.Expr:
    return output.set(a.get() + b.get())
Snippet Source

In the example above the add method is defined to take two Uint64 arguments (a, b) and return a Uint64 (output).

The full method signature for the above is add(uint64,uint64)uint64 and will, by default, field only application call transactions with an OnComplete of NoOp.

On Complete handlers

There are other decorators that can be used to modify the behavior of the method handler including create, optin, closeout, clear, update, and delete. These decorators can be used to register handlers for specific OnComplete values. See the full docs for more details.


Beaker allows for a pattern called blueprints apply a set of method handlers to an Application. Adding handlers using a blueprint allows for code re-use and makes it easier to add behaviors, especially for applications that wish to adhere to some ARC standard.

Blueprints can be defined using a standard python method definition that accepts an Application as an argument and applies the handlers.

The code below defines a calculator blueprint that applies method handlers for a set of functions to implement a simple calculator. The blueprint must take an Application argument and optionally other arguments to modify the behavior of the blueprint

An instantiated Application can apply this blueprint using the .apply method passing blueprint method as an argument. If other arguments in the blueprint method are defined, they can be passed with standard python kwarg format (i.e. .apply(bp, arg1="hello"))

# passing the app to this method will register the handlers on the app
def calculator_blueprint(app: Application) -> Application:
    def add(a: pt.abi.Uint64, b: pt.abi.Uint64, *, output: pt.abi.Uint64) -> pt.Expr:
        return output.set(a.get() + b.get())

    def sub(a: pt.abi.Uint64, b: pt.abi.Uint64, *, output: pt.abi.Uint64) -> pt.Expr:
        return output.set(a.get() - b.get())

    def div(a: pt.abi.Uint64, b: pt.abi.Uint64, *, output: pt.abi.Uint64) -> pt.Expr:
        return output.set(a.get() / b.get())

    def mul(a: pt.abi.Uint64, b: pt.abi.Uint64, *, output: pt.abi.Uint64) -> pt.Expr:
        return output.set(a.get() * b.get())

    return app

calculator_app = Application("CalculatorApp", descr="This is a calculator app")

calculator_app_spec =
Snippet Source

An application that has a blueprint applied can also implement additional handlers or apply additional blueprints. Identical method signatures are not allowed. However, if necessary, an identical method signature still be registered by adding the override attribute to the decorator, external(override=True).

Add State

An Application can define the state it uses to store data. This is done by defining a class that contains some number of StateValue objects as attributes and passing an instance of that class to the Application constructor.

Beaker's GlobalStateValue class can be used to define and alter a contract's global state values. Global state values are defined by passing the TealType and a description the GlobalStateValue constructor.


TealType is specific to the AVM and only bytes, unit64 are acceptable values for state.

The code below illustrates creating global integer counter that is stored in state. First, a CounterState class is created with an instance of a GlobalStateValue as an attribute. This class is then instantiated and passed in the Application constructor. Two method handlers are also added to the app to increment and decrement the counter.

import pyteal as pt
from beaker import Application, GlobalStateValue

class CounterState:
    counter = GlobalStateValue(
        descr="A counter for showing how to use application state",

app = Application(
    "CounterApp", descr="An app that holds a counter", state=CounterState()

def increment() -> pt.Expr:
    return app.state.counter.set(app.state.counter + pt.Int(1))

def decrement() -> pt.Expr:
    return app.state.counter.set(app.state.counter - pt.Int(1))

app_spec =
Snippet Source

Similarly, a LocalStateValue can be used to alter and store local state values. The code below is identical to the previous example, except the counter is stored locally.

import pyteal as pt
from beaker import Application, LocalStateValue

class LocalCounterState:
    local_counter = LocalStateValue(
        descr="A counter for showing how to use application state",

local_app = Application(
    "CounterApp", descr="An app that holds a counter", state=LocalCounterState()

def user_increment() -> pt.Expr:
    return local_app.state.local_counter.set(local_app.state.local_counter + pt.Int(1))

def user_decrement() -> pt.Expr:
    return local_app.state.local_counter.set(local_app.state.local_counter - pt.Int(1))

local_app_spec =
Snippet Source

Beaker provides the BoxMapping and BoxList classes to work in conjunction with existing PyTeal box functionality.

In the example below a BoxMapping instance is defined in the MappingState class. Each entry in the map is keyed using the type of Address and stores a Uint64 value.

The method handler we define allows us to set an integer for the specific application caller's address.

import pyteal as pt
from import BoxMapping

class MappingState:
    users = BoxMapping(pt.abi.Address, pt.abi.Uint64)

mapping_app = Application(
    "MappingApp", descr="An app that holds a mapping", state=MappingState()

def store_user_value(value: pt.abi.Uint64) -> pt.Expr:
    # access an element in the mapping by key
    return mapping_app.state.users[pt.Txn.sender()].set(value)
Snippet Source

The BoxList class can be used to store a list of specific static ABI types. The example below creates a box named users that stores a list of five addresses. The store_user method is passed an address and an index. The passed-in address is then stored in the BoxList at the specified index.

import pyteal as pt
from import BoxList

class ListState:
    users = BoxList(pt.abi.Address, 5)

list_app = Application("ListApp", descr="An app that holds a list", state=ListState())

def store_user(user: pt.abi.Address, index: pt.abi.Uint64) -> pt.Expr:
    # access an element in the list by index
    return list_app.state.users[index.get()].set(user)
Snippet Source

Interacting with the Application

The contract can be deployed and tested using Beaker's sandbox module and the ApplicationClient class. The code below first retrieves the accounts from the currently running sandbox instance. A ApplicationClient (app_client) is then instantiated with an algod client, the Application class that is going to be used, and the first sandbox account (sandbox default starts with a couple of predefined accounts) which will be used to sign transactions.

from beaker import sandbox, client

# grab funded accounts from the sandbox KMD
accts = sandbox.get_accounts()

# get a client for the sandbox algod
algod_client = sandbox.get_algod_client()

# create an application client for the calculator app
app_client = client.ApplicationClient(
    algod_client, calculator_app, signer=accts[0].signer
Snippet Source

The instance of AppliationClient can deploy the calculator app now using the create method.

app_id, app_addr, txid = app_client.create()
print(f"Created app with id: {app_id} and address: {app_addr} in tx: {txid}")
Snippet Source

The contract can then be used to call the contract. In this example the contracts add method is called, and two integers are passed as method arguments. Finally, the return value is printed.

result ="add", a=1, b=2)
print(result.return_value)  # 3
Snippet Source

This is only a small sample of what Beaker can do. For more see Beaker's documentation