PyCaster Boilerplate

PyCaster is a python lib that makes it easy to build apps on Farcaster. It can be used to create Farcaster Frames. This boilerplate provides a demo of it's usage.

Using PyCaster to build your frames backend is easy. I wrote a blog post about it at https://pranavprakash.in/2024/02/12/building-a-farcaster-frame-in-python-with-flask/

If you have any questions, open an issue in this repo or cast to me on Farcaster

Getting Started

Install

1. Clone this repo

git clone git@github.com:merokudao/pycaster-boilerplate.git

2. Optional

Create a virtual environment using python -m venv venv and activate it using source venv/bin/activate

3. Install Dependencies

pip install -r requirements.txt

Environment Variables

Depending on your use case, you may need to set one or more of the following env variables that are needed by PyCaster.

Var	Required	Description
NEYNAR_API_KEY	Recommended	Used to perform various crucial calls related to user. If your app does user related functions (ex: check if the user follows a channel or has casted something), this will be needed.
REDIS_HOST	Yes	Highly recommended to increase speed of frame response. Without this, the pycaster lib has to be changed significantly to bypass.
MEROKU_API_KEY	Optional	Required if you're building on Meroku dApp Store Kit APIs
OPENAI_API_KEY	Optional	Required if you're using OpenAI
AWS_ACCESS_KEY_ID	Optional	If your frame needs to upload media to S3, this is needed.
AWS_SECRET_ACCESS_KEY	Optional	If your frame needs to upload media to S3, this is needed.

Create your own .env by copying from .env.example.

cp .env.example .env

Set env variables in terminal by

set -a; source .env; set +a

Run

Local

RUN using

flask run --host 0.0.0.0 --port 9091

Docker

Build

docker build -t pycasterdemo .

Run

docker run -p 9091:5000 --env-file .env pycasterdemo

Check

ruff .

Contributing

We'd love to accept contriutions. Please open an issue with what you'd like to build and we'll discuss and take it from there.