Metadata-Version: 2.4
Name: shoud-bots
Version: 1.0.0
Summary: Official Python SDK for Shoud Bots — Hardware-isolated microVM sandboxes for AI agents
Author-email: Vantyrix Tek <dev@shoud.online>
License: Apache-2.0
Project-URL: Homepage, https://shoud.online
Project-URL: Documentation, https://shoud.online/docs
Keywords: shoud,sandbox,microvm,ai-agents,code-interpreter,e2b,gvisor,langchain
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.8
Classifier: Programming Language :: Python :: 3.9
Classifier: Programming Language :: Python :: 3.10
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Classifier: Operating System :: OS Independent
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Classifier: Topic :: Software Development :: Build Tools
Requires-Python: >=3.8
Description-Content-Type: text/markdown
Requires-Dist: requests>=2.25.0

# shoud-bots

Official Python SDK for **Shoud Bots** — High-performance, hardware-isolated microVM sandboxes for AI agents.

## Features

- ⚡ **Sub-118ms Cold Boot**: Spin up full Linux microVMs in milliseconds.
- 🛡️ **Zero Host Kernel Access**: Secured by Google gVisor Sentry userspace kernel.
- 🐍 **Full Data Science Stack**: Python 3.11, pandas, numpy, scipy, matplotlib.
- 💰 **Prepaid Per-Second Billing**: Zero monthly lock-in. Unused session reservations are auto-refunded.

## Installation

```bash
pip install shoud-bots
```

## Quickstart

```python
import os
from shoud import Sandbox

# Set your API key or export SHOUD_API_KEY=shoud_live_...
os.environ["SHOUD_API_KEY"] = "shoud_live_YOUR_KEY"

# Context manager automatically closes the sandbox and refunds unused credits
with Sandbox.create(timeout_seconds=300) as sandbox:
    # Run commands
    result = sandbox.commands.run("python3 -c 'import math; print(math.factorial(10))'")
    print(result.stdout)  # 3628800

    # Write files
    sandbox.files.write("calc.py", "print('Executed safely in gVisor sandbox')")
    res = sandbox.commands.run("python3 calc.py")
    print(res.stdout)
```

## Using with LangChain / CrewAI / AutoGen

```python
from shoud import Sandbox

sandbox = Sandbox.create()

def code_interpreter_tool(python_code: str) -> str:
    """Safely execute Python code inside an isolated Shoud microVM."""
    result = sandbox.run_code(python_code)
    return result.stdout if result.exit_code == 0 else result.stderr
```

## Documentation

Full documentation is available at [https://shoud.online/docs](https://shoud.online/docs).
