https://github.com/substratusai/sandboxai
Run AI generated code in isolated sandboxes
https://github.com/substratusai/sandboxai
ai-agent-tools
Last synced: about 1 month ago
JSON representation
Run AI generated code in isolated sandboxes
- Host: GitHub
- URL: https://github.com/substratusai/sandboxai
- Owner: substratusai
- Created: 2025-01-30T20:28:50.000Z (4 months ago)
- Default Branch: main
- Last Pushed: 2025-02-05T21:31:58.000Z (3 months ago)
- Last Synced: 2025-03-28T04:35:01.553Z (about 2 months ago)
- Topics: ai-agent-tools
- Language: Go
- Homepage:
- Size: 249 KB
- Stars: 44
- Watchers: 3
- Forks: 5
- Open Issues: 3
-
Metadata Files:
- Readme: docs/README.md
- Contributing: docs/contributing.md
Awesome Lists containing this project
README
# SandboxAI
*Unleash your AI agents ... in a sandbox!*
SandboxAI is an open source runtime for securely executing AI-generated Python code and shell commands in isolated sandboxes.
* **Self-host**
* Docker based single host backend
* Kubernetes based multi-node backend (coming soon)
* **Simple SDK**
* `pip install sandboxai-client` is all you need
* Go SDK included
* **Container based**
* You define the environment
* Integrate with existing tooling
* **Flexible control**
* Full-agency - Set the agent loose within the container
* Limited-agency - Give the agent access to specific commands
* **Open API**
* Simple RESTful interface
* Generate your own client from the API spec
## Architecture
SandboxAI was designed to fit nicely into your AI software stack.
## Quickstart (Single-host)
Make sure you have docker installed.
TIP: Free alternatives to docker desktop...
* On a mac? Checkout [colima](https://github.com/abiosoft/colima).
* On linux? Use [docker engine](https://docs.docker.com/engine/install/).
```bash
docker ps
```
Install the python client.
```bash
pip install sandboxai-client
```
Run a sandbox on your local machine (using an embedded sandbox server to manage the docker containers).
```python
from sandboxai import Sandbox
with Sandbox(embedded=True) as box:
print(box.run_ipython_cell("print('hi')").output)
print(box.run_shell_command("ls /").output)
```
## Contributing
We welcome contributions to SandboxAI. See the [contributing doc](contributing.md) for more information.
## Contact
Let us know about features you are interested in seeing or reach out with questions. [Visit our Discord channel](https://discord.gg/JeXhcmjZVm) to join the discussion!
Or just reach out on LinkedIn if you want to connect:
* [Nick Stogner](https://www.linkedin.com/in/nstogner/)
* [Sam Stoelinga](https://www.linkedin.com/in/samstoelinga/)