python-executor

Execute Python code in a sandboxed environment with 100+ pre-installed libraries.

Updated Sep 14, 2026
One-click install
npx skills add https://github.com/desarrolloainia/nuevo_circuito_mir --skill python-executor-desarrolloainia
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: python-executor
Source: https://github.com/desarrolloainia/nuevo_circuito_mir/tree/main/backend/.agents/skills/python-executor
Command: npx skills add https://github.com/desarrolloainia/nuevo_circuito_mir --skill python-executor-desarrolloainia

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Running Python code for data processing, web scraping, or media manipulation requires local environment setup, dependency management, and security isolation. This Skill executes Python code in a sandboxed environment via inference.sh with 100+ pre-installed libraries, removing setup overhead. ## Core Features & Use Cases - Sandboxed Execution: Run Python 3.10 code in an isolated subprocess with configurable timeout (1-300 seconds) and memory (8GB or 16GB). - Rich Library Stack: Pre-installed NumPy, Pandas, Matplotlib, Selenium, Playwright, MoviePy, Pillow, OpenCV, trimesh, and more for data, web, image, video, and 3D tasks. - Automatic File Output: Files saved to the outputs/ directory are automatically detected and returned, including charts, CSVs, videos, and 3D models. - Use Case: Scrape a website with requests and BeautifulSoup, analyze the data with Pandas, generate a Matplotlib chart, and receive the PNG output in a single run. ## Quick Start Ask the AI to run a Python script that fetches data from an API and saves a chart to the outputs folder using the python-executor skill.

Frequently Asked Questions about python-executor

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run Python code in a sandboxed environment?▼

Use the belt CLI command 'belt app run infsh/python-executor' with a JSON input containing your code string. The code runs in an isolated Python 3.10 subprocess and returns printed output plus any files saved to the outputs/ directory.

What Python libraries are pre-installed for web scraping and data analysis?▼

The environment includes requests, httpx, BeautifulSoup, Selenium, Playwright, and Scrapy for scraping, plus NumPy, Pandas, SciPy, Matplotlib, Seaborn, and Plotly for data processing and visualization. Over 100 libraries are available without installation.

Does the Python executor support GPU or machine learning libraries?▼

No, the environment is CPU-only and does not include GPU or ML libraries. For AI-based image or video generation, use the dedicated ai-image-generation or ai-video-generation apps instead.

How do I get files back from executed Python code?▼

Save files to the outputs/ directory in your code, such as plt.savefig('outputs/chart.png') or df.to_csv('outputs/data.csv'). Files in that directory are automatically detected and included in the response.

Why is my Python script timing out during execution?▼

The default timeout is 30 seconds. Increase it by setting the timeout field in the input JSON, up to a maximum of 300 seconds, for long-running tasks like video rendering or large dataset processing.

Can I process large datasets that need more memory?▼

Yes, use the high_memory variant by running 'belt app run infsh/python-executor@high_memory' to get 16GB RAM instead of the default 8GB, which suits larger Pandas DataFrames or 3D mesh workloads.