likhonsheikh/code-interpreter-sandbox

🚀 Advanced Code Interpreter Sandbox

A powerful, feature-rich code execution environment on HuggingFace Spaces that rivals e2b.dev! This sandbox provides a comprehensive Python development environment with advanced capabilities.

✨ Features

🔐 Secure Code Execution

• —Safe, sandboxed Python code execution
• —Timeout protection (10s default)
• —Memory limit controls
• —Syntax error handling
• —Exception capture and display

📁 File System Support

• —Upload files directly to the workspace
• —Download results and outputs
• —Multi-file management
• —Real-time file operations (read, delete, list)
• —Temporary file storage with session isolation

📦 Package Management

• —On-the-fly package installation via pip
• —Support for popular data science libraries
• —Package version management
• —Quick install of common packages (numpy, pandas, matplotlib, plotly, etc.)
• —Pre-installed essential packages

📊 Data Visualization

• —Built-in support for:
• —Matplotlib - Static plots and charts
• —Plotly - Interactive visualizations
• —Seaborn - Statistical data visualization
• —Bokeh - Interactive plots
• —Altair - Declarative visualization
• —Pillow - Image processing

💾 Session Persistence

• —Maintains state throughout session
• —File persistence
• —Package installation history
• —Session information tracking
• —Uptime monitoring

🔄 Real-time Output

• —Streamed stdout/stderr capture
• —Live code execution feedback
• —Error highlighting
• —Output mode selection (stdout, stderr, or both)

📝 Multi-file Support

• —Create and manage multiple files
• —File editor interface
• —Switch between files easily
• —Automatic file detection
• —File metadata tracking

🎨 Superior UX

• —Clean, modern Gradio interface
• —Dark theme support
• —Intuitive tabbed interface
• —Responsive design
• —Syntax highlighting in code editor
• —Progress indicators

🚀 Getting Started

1. Run Your Code

# Write Python code in the editor
print("Hello, World!")

# Import any installed package
import numpy as np
arr = np.array([1, 2, 3, 4, 5])
print(f"Array: {arr}")
print(f"Mean: {np.mean(arr)}")

2. Install Packages

• —Go to "Package Manager" tab
• —Enter package names (comma-separated)
• —Click "Install"
• —Example: numpy, pandas, matplotlib, plotly

3. Upload Files

• —Go to "File Manager" tab
• —Upload files using the file picker
• —Files are stored in your workspace
• —Access files directly in your code

4. Data Visualization Example

import matplotlib.pyplot as plt
import numpy as np

# Generate data
x = np.linspace(0, 10, 100)
y = np.sin(x)

# Create plot
plt.figure(figsize=(10, 6))
plt.plot(x, y, 'b-', label='sin(x)')
plt.xlabel('x')
plt.ylabel('y')
plt.title('Sine Wave')
plt.legend()
plt.grid(True)
plt.show()

📚 Pre-installed Libraries

The following packages are pre-installed and ready to use:

• —NumPy
• —Pandas
• —Matplotlib
• —Plotly
• —Seaborn
• —SciPy
• —Scikit-learn
• —Pillow
• —Requests
• —BeautifulSoup4
• —NetworkX
• —SymPy

🔧 Available Packages

You can install additional packages using the Package Manager. Popular options:

• —tensorflow - Machine learning
• —torch - Deep learning
• —transformers - Hugging Face transformers
• —streamlit - Web app framework
• —dash - Interactive dashboards
• —openai - OpenAI API client
• —langchain - LLM application framework
• —wordcloud - Text visualization
• —geopandas - Geospatial analysis
• — sqlalchemy - Database ORM

🎯 Use Cases

Data Analysis

import pandas as pd
import matplotlib.pyplot as plt

# Load and analyze data
df = pd.read_csv('your_data.csv')
print(df.head())
df.describe().plot()
plt.show()

Machine Learning

from sklearn.ensemble import RandomForestClassifier
from sklearn.model_selection import train_test_split

# Train a model
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2)
model = RandomForestClassifier()
model.fit(X_train, y_train)

Web Scraping

import requests
from bs4 import BeautifulSoup

# Scrape a webpage
url = "https://example.com"
response = requests.get(url)
soup = BeautifulSoup(response.content, 'html.parser')
print(soup.title)

API Integration

import requests
import json

# Call an API
response = requests.get('https://api.github.com/users/octocat')
data = response.json()
print(json.dumps(data, indent=2))

⚠️ Security Notes

• —Code runs in a secure sandbox environment
• —No network access to external resources (except for package installation and explicit API calls)
• —Files are isolated per session
• —Execution timeouts prevent infinite loops
• —Maximum memory usage is limited

🆚 Compared to e2b.dev

🎨 Interface Overview

Code Executor Tab

• —Write and execute Python code
• —Choose output mode (stdout, stderr, or both)
• —Real-time execution feedback
• —Syntax highlighting

File Manager Tab

• —Upload files to workspace
• —View and manage all files
• —Read file contents
• —Delete unwanted files

Package Manager Tab

• —Install packages on-the-fly
• —View installed packages
• —Batch installation support

Session Info Tab

• —View session details
• —Monitor uptime
• —Track installed packages
• —Workspace information

🚀 Deployment

This application is designed for HuggingFace Spaces and includes:

• —Optimized requirements.txt
• —Pre-configured dependencies
• —Docker support
• —GPU acceleration ready

📝 Tips & Tricks

1. 1.Use variables across executions: Variables persist within a session
2. 2.Install packages first: Install required packages before using them
3. 3.Save important outputs: Use file operations to save results
4. 4.Check session info: Monitor your session status
5. 5.Explore the interface: Each tab provides different functionality

🤝 Contributing

Contributions are welcome! Areas for improvement:

• —Additional language support
• —Enhanced visualization options
• —More pre-installed packages
• —Improved error handling
• —Better performance optimization

📄 License

This project is open source and available under the MIT License.

🙏 Acknowledgments

• —Built with Gradio
• —Powered by HuggingFace Spaces
• —Inspired by e2b.dev

Ready to code? Just start typing in the Code Executor tab! 🚀