How to Install SuperAGI: Easy 10-Step Guide

SuperAGI

Introduction to SuperAGI

SuperAGI is essentially AutoGPT on steroids. It can use tools, run multiple agents in parallel, has a graphical user interface, and is super easy to install. In this guide, we will walk you through the process of installing SuperAGI and how to use it.

You can find more information about SuperAGI on GitHub. They say it is still under construction and still has many flaws. In my experience, it seems well-polished for a new AI Tool.

Step 1: Get Docker

The first thing you need to do is get Docker. Visit docker.com, download Docker, install it, and have it up and running.

Docker Download Page

It looks like this, and you can keep that on in the background.

Docker Desktop for Super AGI

Step 2: Clone the SuperAGI Repository

Next, open Visual Studio Code and click the little button in the top right that says “toggle panel”. This will open up your terminal within Visual Studio Code. If you don’t have Visual Studio Code, download it from https://code.visualstudio.com/download.

Visual Studio Code Terminal

On the SuperAGI GitHub repo page, click the green “Code” button and copy the URL.

SuperAGI GitHub Repo

Once the terminal opens up, type cd C:\ to change the directory to C Drive. From there, type mkdir \superagi to create a new folder and then cd \superagi to change the directory to the new folder you created. Now type git clone and paste the URL you just copied. Hit enter, and that will clone it to your new folder.

cd C:\

mkdir \superagi

cd \superagi

git clone https://github.com/TransformerOptimus/SuperAGI.git

Step 3: Open the Super AGI Folder

In Visual Studio Code, click the Explorer icon, then click “Open Folder”. Select SuperAGI and click “Select Folder”.

SuperAGI Folder

Again, in the top right, click “toggle panel” to open up your terminal.

Visual Studio Code Terminal

Step 4: Rename the Config Template

Right-click on config_template.yaml and rename it to config.yaml. This is where you’re going to store all your environment variables. This is where you will put your Pinecone key, OpenAI key, Google key, and anything else you want to use.

SuperAGI yaml config

Step 5: Create an API Key with OpenAI

If you don’t already have an account, sign up for OpenAI. Then browse to platform.openai.com/account/api-keys. Click “create new key”, name it, and click “create secret key”. Copy it and paste it into the config.yaml file in the OpenAI key section.

Step 6: Get a Pinecone API Key

Pinecone Vector DB Site

Create a free Pincone account; it is a vector database. If you’re using a free Pinecone account, you can only have one index at a time.

Once your account is created, go to Pinecone, click “API keys”, and click “create new API key”. Enter the key name, create the key, and copy the key value. Switch back to Visual Studio Code and paste the key value in the Pinecone API key section of the config.yaml file.

Step 7: Get the Google API Key

Go to Google Cloud Services, and create an account for free. Click “API and Services”, and create a new project. Search for “custom search API” and click “Create credentials”.

Google Cloud Services

Create a new API key, copy it, and paste it into the Google API key section of the config.yaml file.

SuperAGI GOOGLE_API_KEY

Step 8: Get the Search Engine ID

Go to the Programmable Search Engine website, click “get started”, and create a new search engine.

Programmable Search Engine Form

Copy the search engine ID and paste it into the Search Engine ID section of the config.yaml file.

SuperAGI config.yaml settings

Step 9: Install SuperAGI

With Docker up and running, type docker-compose up --build in the terminal and hit enter. This will build everything from the repository. Once it’s up and running, go to your browser and type localhost:3000 , and hit enter. You should now see SuperAGI.

Step 10: Create an Agent

Go to the “agents” tab, click “create agent”, and fill in the necessary details. You can add one or multiple goals here. Select the tools you want SuperAGI to use, click “create and run”, and your agent will start running. You can have multiple agents running in parallel.

Step 11: Monitor the Agent

If you want more information about what’s happening, switch back to Visual Studio Code and look through the logs of what the agent is doing. The agent will think, plan, and execute tasks based on your goals.

And that’s it! You’ve successfully installed Super AGI. Feel free to play around with it and explore its many functionalities. If you have any questions or need further assistance, please reach out in the comment section of this blog post. Happy exploring with SuperAGI!

SuperAGI GUI Demo

SuperAGI GUI Walkthrough

Features of SuperAGI

SuperAGI is now on version v0.0.6, with every new release, there have been major improvements and many features added for the SuperAGI Autonomous Agent!

Basic Google search capability that was released with SuperAGI’s version v0.0.1.

File Write

The ability to write files is a necessary feature for autonomous agents. This feature was also released in version v0.0.1.

Web Interactor

Harnessing the power of the python-playwright package, developers can traverse webpages and carry out tasks within the browser. The Web Interactor, operating within a Docker container, now boasts of real-time VNC support, offering users the exciting opportunity to witness agent browser activity as it unfolds.

Searx

Searx is a robust, privacy-centric internet metasearch engine, compiling results from more than 70 search services. Developers can relish its non-tracking and non-profiling attributes, ensuring a secure browsing experience. Searx also allows operating over the Tor network for an added layer of online anonymity.

GitHub

This tool offers a handy feature to pinpoint specific files or folders within a GitHub repository. Developers are further empowered with the ability to execute actions like adding or deleting files and folders. Integrating with SuperAGI’s Coding tool unlocks many productivity use cases for developers, enhancing their efficiency.

Slack

Slack integration empowers agents to dispatch messages in Slack channels, thereby facilitating the execution of Slack-centric workflows by the agent or existing Slack workflows.

Email

SuperAGI agents can send emails.

Web Scraping

Web scraping is a powerful tool that enables the extraction of vast amounts of data from the internet, unlocking new opportunities for data-driven insights and decision-making.

DALL.E2

DALL-E 2 is an advanced version of OpenAI’s DALL-E, an artificial intelligence model capable of generating diverse images from textual descriptions, offering improved capabilities and performance.

Stable Diffusion

Agents now have the power of stable diffusion models at their fingertips. Agents can conjure up strikingly photorealistic images! This opens up a thrilling world of possibilities, empowering agents to generate the vibrant multimedia assets necessary to achieve their objectives triumphantly.

Google Calendar

Empowering agents with the ability to create, update, and delete events automatically, they can seamlessly manage and schedule appointments directly onto the user’s calendar. Something that I am personally excited about is if I can get this working!

Coding

Agents now support coding competing with other Agent based code generators like GPT-Engineer, which we covered in our article about it. Have a read, and let us know which agent you think is better at coding.

Write Specs

Agents are granted the capability to craft meticulous specifications for software projects. They can also generate comprehensive specifications derived from user inputs, encapsulating the project’s requirements, objectives, and design considerations in a detailed manner.

Write Unit Tests

This enables agents to create unit tests for their code, using either the user’s specifications or those produced by the Write Specs Tool above. Self-testing code AI continues to amaze me!

Agents are empowered to construct unit tests for their code, drawing from specifications either supplied by the user or produced by the innovative Write Specs Tool.

Conclusion

In conclusion, SuperAGI is a revolutionary tool that is redefining the landscape of autonomous agents. It’s easy-to-use interface and robust functionality make it an invaluable asset for developers and AI enthusiasts. With the ability to manage multiple agents simultaneously, SuperAGI offers unparalleled performance and unparalleled efficiency.

Its continuous evolution and improvement, as evidenced by the regular addition of new features, demonstrate its commitment to staying at the forefront of AI technology.

Whether you’re a seasoned developer or just starting your journey in AI, SuperAGI is a tool that can help you achieve your goals and push the boundaries of what’s possible. So, why wait? Start exploring the world of SuperAGI today!

Want to know how SuperAGI stacks up against Auto-GPT? Read our Latest Post, SuperAGI vs AutoGPT.

My name is Lachie, and I am the author of this how-to guide. Check out how I use OpenAi’s Chat GPT-4 to improve my daily life.

Leave a Reply

Up ↑

%d