SuperAGI: V0.0.6 Now Live [Updated Jun 2023]

1 min


Build Autonomous AI Agents


SuperAGI is an open-source repository that provides the infrastructure to build and deploy autonomous AI agents for various applications. It allows developers to run multiple agents concurrently, customize them with different models and tools, and interact with them through a graphical user interface and an action console. SuperAGI is designed to enable rapid and reliable development of useful autonomous agents.

SuperAGI Setup


  1. Download the repository using git clone in your terminal or directly from github page in zip format.
  2. Use cd SuperAGI to enter the directory and make a duplicate of config_template.yaml and name it config.yaml.
  3. To use the config.yaml file. you need to obtain four keys from different sources: OpenAI API Key, Google key, Custom search engine ID, and Pinecone API key. Do not include any quotes or spaces when you paste the keys in the file. You can find the links to get the keys below:
Keys Accessing the keys
OpenAI API Key Sign up and create an API key at OpenAI Developer
Google API key Create a project in Google Cloud Console and enable the API you need (for example: Google Custom Search JSON API). Then, create an API key in the “Credentials” section.
Custom search engine ID Navigate to Google Programmable Search Engine to create a custom search engine for your application and generate the search engine ID.
Pinecone API key Register at Pinecone and generate an API key in your account dashboard.

SuperAGI Features

✅Create and launch self-governing AI agents
✅Enhance agent skills with tools
✅Manage multiple agents simultaneously
✅Visual interface for users
✅Action console for commands
✅Various vector databases
✅Multi-modal agent communication
✅Agent path fine-tuning
✅Performance monitoring
✅Token usage optimization
✅Agent memory storage
✅Loop detection heuristics
✅Parallel agents
✅Resource manager

To use Pinecone free plan, you can only create 1 pod and 1 index. To avoid errors, update the index name in where you define memory

 memory = VectorFactory.get_vector_storage("PineCone", "my-current-indexname", OpenAiEmbedding())

Before proceeding, check if you have Docker on your system. If not, you can download it from this link.

To launch SuperAGI, make sure Docker Desktop is active and execute : docker-compose up --build from the SuperAGI folder. Then, visit localhost:3000 in your browser to view SuperAGI.


Developers are working on this project and it might have some bugs. Before reporting a problem, please look at the existing issues. If you don’t see your problem there, please make a new issue and describe what went wrong.

[Update May 5, 2023] SuperAGI v0.0.3 now Live

What’s New in v0.0.3?
⚙️ Improvements to the Core
  • Added a dedicated resource manager(v1) to manage input and output files used by agents
  • Resource manager supports uploading input files and downloading output files
  • Resource manager supports .txt & .pdf as input
  • Added support for DB migration using Alembic top support customization in SuperAGI DB models
  • Improved agent activity feed
  • Improved agent details section in GUI – Get clear visibility of multiple goals
📑 Documentation Updates
  • Updated the docker command to run SuperAGI: docker-compose up –build
🚀 What’s Next in v0.0.4?
  • New tools – DALL-E2, GitHub, Web Interaction tool, Human interaction tool
  • Agent Performance Metrics
  • Open-source LLM models – LLaMa, Vicuna, Alpaca and more
  • Vector DBs – Chroma DB
  • Improvements in resource manager (v2)

[Update Jun 8, 2023] SuperAGI v0.0.4 now Live

What’s New in v0.0.4?

⚙️ Core Enhancements
  • Dropped Pinecone dependency
  • Customize Max Iteration limit per agent run to control API calls
  • Monitor token usage – visible in GUI
  • Improved task prioritization with Agents’ task queue
  • Track API calls per agent
  • Set OpenAI API key from GUI
🛠 Tools
  • Added DALL.E2 for image generation
  • Web Scraping – To extract data from webpages
  • Coding Tool (all programming languages) – To code and save output file in resource manager
📑 Documentation Update
  • Updated Docker install instructions

[Update Jun 15, 2023] SuperAGI v0.0.5 now Live

What’s New in v0.0.5?

⚙️ Core Enhancements
  • Open-source LLMs: Text-generation-webui supports them, allowing efficient local LLM hosting. Developers can use them to enhance agents. Text Generation Web UI’s latest Docker image has multi-GPU offloading for GGMLs, boosting SuperAGI. Developers can also configure SuperAGI to use TGWUI and download models from
  • Agent Templates: Predefined agents for various purposes. Soon on Marketplace.
  • Test Coverage: Added tests for Image Generator, GitHub tool, etc. More tests to improve code quality and stability.
  • Task Queue Optimization: Optimized task queue agents: Improved workflow and task handling.
  • Track API calls per agent
  • Set OpenAI API key from GUI
🛠 Tools
  • Web Interaction Tool: Uses Python-playwright to automate browser tasks. Runs in a Docker container with VNC support.
  • GitHub Tool: Searches, adds, or deletes files and folders in GitHub repos. Works with Coding tool for developer productivity.
  • Slack Tool: Sends messages in Slack channels for Slack-based workflows.
  • Searx Tool: Privacy-focused metasearch engine aggregating 70+ services. No tracking or profiling. Option to use Tor network for anonymity.
📑 Documentation Update

[Update Jun 23, 2023] SuperAGI v0.0.6 now Live

What’s New in v0.0.6?

⚙️ Improvements to the Core

🤖 Introduced SuperCoder agent template type.
SuperCoder template is a powerful tool that enables users to create complete coding projects with ease. Users only need to specify their objectives and follow some simple steps to generate high-quality code.
One of its capabilities is to generate mini retro games and apps using the appropriate code and testing methods. Some examples of its creations are:
💬 Now add ‘Agent Instructions’ in the agent setup.
The agent’s workflow can be optimized for accuracy and performance by giving it detailed guidance on how to execute each task.
✅ Introduced ‘Restricted mode’ with permission controls.
The agents running on restricted will constantly seek human approvals to execute critical tasks like writing files, sending emails, moving to the next iteration, etc. Users can either “Approve”, “Give feedback” or “Deny” these permission requests in the Action console to let the agent proceed to the next step.
🛠 Added new primitive of toolkits.
Toolkit is a collection of tools. For example, the file manager is a toolkit containing tools like the read file & write file tools. It helps the user to easily attach tools to an agent at the time of provisioning. Also allowing contributors to manage & publish new tools for SuperAGI.
🏋️ Integrated NGINX
To streamline the installation process NGINX acts as a reverse proxy, helping in running the Front-End and Back-End on the same URL.
🧠 Introducing Tool Memory
To allow agents to access and read responses from other tools. Tools can build upon each other’s outputs and cross-communicate to support more complex and integrated workflows. Allows for state management for tool output.
✅ Implemented GitHub Actions
To automate the unit testing and linting for the codebase. Ensuring that code meets quality standards and is error-free before merging changes.

🛠 New Tools Added

🏞 Stable Diffusion:
Agents can use stable diffusion models to generate photorealistic images. Allowing agents to generate multimedia assets required to complete the goals.
🗓 Google Calendar:
Enabling agents to create/update/delete events to manage and schedule appointments automatically to the user’s calendar.
✒️ Write Specs Tool:
Allows agents to write specifications for software projects. Also generate detailed specifications based on user inputs, outlining the requirements, goals, and design considerations for a project.
🧪 Write Unit Tests Tool:
Allows agents to write unit tests for their code based on specifications provided by the user or generated by the Write Specs Tool.

📑 Documentation Update

✍️  Added v1 API documentation to SuperAGI Docs
✍️ Added SuperAGI architecture diagrams to README

Join Guidady AI Mail List

Subscribe to our mailing list and get interesting stuff and updates to your email inbox.

Thank you for subscribing.

Something went wrong.

Like it? Share with your friends!

79 shares, 1 point


Your email address will not be published. Required fields are marked *


I am an IT engineer, content creator, and proud father with a passion for innovation and excellence. In both my personal and professional life, I strive for excellence and am committed to finding innovative solutions to complex problems.
Choose A Format
Personality quiz
Series of questions that intends to reveal something about the personality
Trivia quiz
Series of questions with right and wrong answers that intends to check knowledge
Voting to make decisions or determine opinions
Formatted Text with Embeds and Visuals
The Classic Internet Listicles
The Classic Internet Countdowns
Open List
Submit your own item and vote up for the best submission
Ranked List
Upvote or downvote to decide the best list item
Upload your own images to make custom memes
Youtube and Vimeo Embeds
Soundcloud or Mixcloud Embeds
Photo or GIF
GIF format