Superagent is a powerful tool that simplifies the configuration and deployment of LLM (Large Language Model) Agents to production. It provides a range of features and functionalities to make it easier for developers to build, manage and deploy AI agents to production including features such as built in memory and document retrieval via vector dbs, powerful tools, webhooks, cron jobs etc.
If you are looking for a plug-n-play way getting started be sure to checkout Superagent.sh.
Checkout the full documentation here.
You can follow the roadmap here
We post tutorials regularly on our Youtube channel. Make sure to check them out!
To get started with Superagent, follow these steps:
-
Clone the Superagent repository into a public GitHub repository or fork it from https://github.com/homanp/Superagent/fork. If you plan to distribute the code, keep the source code public.
git clone https://github.com/homanp/Superagent.git
-
To run the script, simply execute it using:
bash setup.sh
-
See the setup instructions for the UI in the
ui
folder.
In the .docker
folder there are multiple docker-compose files.
The main docker-compose.yml
file will start up the API and a Postgres DB in docker. You can optionally also run the UI in docker too.
The other docker compose files can be used individually, or in combination to start up just the bits you need.
follow the guide in .docker/README.md file to get started
To run this project, you will need to add the following environment variables to your .env file
Mandatory variables
DATABASE_URL
- A database connection string (with pooling)
DATABASE_MIGRATION_URL
- A database connection (without pooling), used when creating/applying migrations.
OPENAI_API_KEY
- An OpenAI API key
JWT_SECRET
- A secret key/string
VECTORSTORE
- Change this if you plan on supporting other vector databases.
PINECONE_ENVIRONMENT
- Pinecone environment (found in the pinecone dashboard)
PINECONE_API_KEY
- Pinecone API key (found in the pinecone dashboard)
PINECONE_INDEX
- The name of the Pinecone index you would like to use
SUPERAGENT_TRACING
- If you want to enable agent tracing
Optional variables
PSYCHIC_API_KEY
- Psychic.dev API key
BING_SUBSCRIPTION_KEY
- Bing Search subscription key (found in the Azure dashboard)
BING_SEARCH_URL
- Bing Search url (found in the Azure dashboard)
WOLFRAM_ALPHA_APPID
- Wolfram Alpha App ID (found in your Wolfram Alpha dashboard)
REPLICATE_API_TOKEN
- Replicate API token (found in uour replicate dashboard)
If you plan on using other Language Models
ANTHROPIC_API_KEY
COHERE_API_KEY
HUGGINGFACEHUB_API_TOKEN
AZURE_API_KEY
AZURE_API_TYPE
AZURE_API_BASE
AZURE_API_VERSION
- Running Superagent with NextJS
- Javascript
- Python
- Swift (community)
Superagent is an open-source project, and contributions are welcome. If you would like to contribute, you can create new features, fix bugs, or improve the infrastructure. Please refer to the CONTRIBUTING.md file in the repository for more information on how to contribute.
We appreciate your contributions and aim to make it easy for anyone to create and run LLM Agents in production using Superagent.
A big thanks to all partners that support the development of Superagent!
๐ฟ Fern: Fern helps create SDKs and client libraries from OpenAPI specs. Superagent uses Fern for all of the client libraries and SDKs we provide. A big shout out for the support!
We appreciate all the support you can give us, either with contributions, feedback, bug reports or feature requests. Drop a star and share Superagent to the world!