![midjourney prompt: The ninth planet around the sun](https://raw.githubusercontent.com/kreneskyp/ix/master/ix_350.png)
Amidst the swirling sands of the cosmos, Ix stands as an enigmatic jewel,
where the brilliance of human ingenuity dances on the edge of forbidden
knowledge, casting a shadow of intrigue over the galaxy.
- Atreides Scribe, The Chronicles of Ixian Innovation
- QA chat bots
- Code generation
- Data extraction
- Data analysis
- Data augmentation
- Research assistants
- OpenAI
- Google PaLM (Experimental)
- Anthropic (Experimental)
- Llama (Experimental)
No-code editor for creating and testing agents. The editor provides an interface to drop and connect nodes into a graph representing the cognitive logic of an agent. Chat is embedded in the editor to allow for rapid testing and debugging.
MetaphorCreateShort_V2.mp4
Create your own teams of agents and interact with them through a single interface. Chat room support multiple agents.
By default it includes the IX moderator agent, which delegates tasks to other agents. You can @mention
specific
agents to complete the tasks.
fizzbuzz.mkv_V2.mp4
The smart input bar auto-completes agent @mentions
and file & data {artifacts}
created by tasks.
smart_input.mkv_V2.mp4
The agent runner backend is dockerized and is triggered with a celery message queue. This allows the backend to scale horizontally to support a fleet of agents running in parallel.
IX implements a component config layer that maps LangChain components to the configuration graph. The config layer powers a number of other systems and features. For example, component field and connector definitions are used to render nodes and forms dynamically in the no-code editor.
Windows Linux Subsystem (windows only)
- Open powershell
- run `wsl --install` to install and/or activate WSL
Docker
Install Docker Desktop for your OS:https://www.docker.com/products/docker-desktop/
Detailed install instructions:
Python
Python 3.8 or higher is required for the CLI. The app python version is managed by the image.The quickest way to start IX is with the agent-ix CLI. The CLI starts a preconfigured docker cluster with docker-compose. It downloads the required images automatically and starts the app cluster.
pip install agent-ix
ix up
Scale agent workers with the scale
command. Each worker will run agent processes in parallel. The limit to the number
of workers is based on available memory and CPU capacity.
ix scale 5
The client may start a specific version, including the unstable dev
image built on master
branch.
ix up --version dev
You chat with an agent that uses that direction to investigate, plan, and complete tasks. The agents are capable of searching the web, writing code, creating images, interacting with other APIs and services. If it can be coded, it's within the realm of possibility that an agent can be built to assist you.
-
Setup the server and visit http://0.0.0.0:8000, a new chat will be created automatically with the default agents.
-
Enter a request and the IX moderator will delegate the task to the agent best suited for the response. Or
@mention
an agent to request a specific agent to complete the task. -
Customized agents may be added or removed from the chat as needed to process your tasks
IX provides the moderator agent IX, a coder agent, and other example agents. Custom agents may be built using the chain editor or the python API.
- Navigate to the chain editor
- Click on the root connector to open the component search
- Drag agents, chains, tools, and other components into the editor
- Connect the components to create a chain
- Open the test chat to try it out!
Chains python API docs
Before getting started, ensure you have the following software installed on your system:
Windows Linux Subsystem (windows only)
- Open powershell
- run `wsl --install` to install and/or activate WSL
Docker
Install Docker Desktop for your OS:https://www.docker.com/products/docker-desktop/
Detailed install instructions:
Git & Make
- Mac:
brew install git make
- Linux:
apt install git make
- Windows (WSL):
apt install git make
git clone https://github.com/kreneskyp/ix.git
cd ix
Setup config in .env
cp .env.template .env
OPENAI_API_KEY=YOUR_KEY_HERE
The image will build automatically when needed in most cases. Set NO_IMAGE_BUILD=1
to skip rebuilding the image.
Use the image
target to build and start the IX images. The dev_setup
target will build the frontend and
initialize the database. See the developer tool section for more commands to manage the dev environment.
make dev_setup
The IX cluster runs using docker-compose. It will start containers for the web server, app server, agent workers, database, redis, and other supporting services.
make cluster
Web and app container logs
make server
Agent worker container logs
make worker
Visit http://0.0.0.0:8000 to access the user interface. From there you may create and edit agents and chains. The platform will automatically spawn agent processes to complete tasks as needed.
Adjust the number of active agent workers with the scale
target. The default is 1 agent worker to handle tasks. There
is no hard limit on agents, but the number of workers is limited by available memory and CPU capacity.
make scale N=5
Here are some helpful commands for developers to set up and manage the development environment:
make up
/make cluster
: Start the application in development mode at http://0.0.0.0:8000.make server
: watch logs for web and app containers.make worker
: watch logs for agent worker containers.
make image
: Build the Docker image.make frontend
: Rebuild the front end (GraphQL, relay, webpack).make webpack
: Rebuild JavaScript only.make webpack-watch
: Rebuild JavaScript on file changes.make dev_setup
: Builds frontend and generates database.make node_types_fixture
: Builds database fixture for component type definitions.
make migrate
: Run Django database migrations.make migrations
: Generate new Django database migration files.
make bash
: Open a bash shell in the Docker container.make shell
: Open a Django shell_plus session.
Dump fixtures with the dump_agent
django command. This command will gather and dump the agent and chain, including
the component graph.
-
make bash
-
./manage.py dump_agent -a alias