Repolode
Submit
e2b-dev

e2b-dev/code-interpreter

Python and JS/TS SDKs for running AI-generated code in isolated cloud sandboxes and reading back its output.

View on GitHubWebsite
Stars
2,421
Forks
233
Open issues
27
Last commit
yesterday

What is it?

E2B Code Interpreter is an SDK that lets AI applications execute code they generate inside secure, isolated cloud sandboxes. Developers create a sandbox from Python or JavaScript/TypeScript, run code in it, and read back the results and printed output.

The sandbox keeps state between calls, so a variable set in one run (for example, x = 1) is still available in the next. This makes it suitable for multi-step workflows such as data analysis driven by an LLM. The project is the Python and JS SDK layer; sandbox templates for customizing packages and runtimes are included in the repository.

The SDK sources have moved to the E2B monorepo, and this repository keeps the sandbox template and a chart data extractor.

Key features

  • 01Runs AI-generated code in isolated cloud sandboxes.
  • 02Provides Python and JavaScript/TypeScript SDKs.
  • 03Keeps variable state across multiple code executions in one sandbox.
  • 04Returns execution text and output from each run.
  • 05Supports building custom sandbox templates with extra packages or runtimes.

Use cases

  • Letting an LLM agent run Python to analyze data and return results.
  • Executing untrusted model-generated code without touching the host system.
  • Building chat apps that compute answers by running code, such as with OpenAI or Anthropic models.

Quick start

  1. 1

    Install the Python SDK

    pip install e2b-code-interpreter
  2. 2

    Get an E2B API key from the dashboard and set it as the E2B_API_KEY environment variable

  3. 3

    Run the Python example that creates a Sandbox and executes code in it

Things to know

  • Requires an E2B account and API key to create sandboxes.
  • Sandboxes run in E2B's cloud, so it depends on a network connection and the hosted service.

Topics

#code-execution#sandbox#llm#ai-agents#python#typescript#jupyter#ai#ai-data-analysis#anthropic

FAQ

What is code-interpreter?
Python and JS/TS SDKs for running AI-generated code in isolated cloud sandboxes and reading back its output. E2B Code Interpreter is an SDK that lets AI applications execute code they generate inside secure, isolated cloud sandboxes. Developers create a sandbox from Python or JavaScript/TypeScript, run code in it, and read back the results and printed output.
Is code-interpreter free to use? What license does it have?
code-interpreter is open source under the Apache-2.0 license. Check the license terms before using it in commercial products.
Is code-interpreter still maintained?
Actively maintained. The last commit to code-interpreter was on Oct 20, 2026.
What can I use code-interpreter for?
Letting an LLM agent run Python to analyze data and return results. Executing untrusted model-generated code without touching the host system. Building chat apps that compute answers by running code, such as with OpenAI or Anthropic models.
How do I install code-interpreter?
Install the Python SDK: pip install e2b-code-interpreter Get an E2B API key from the dashboard and set it as the E2B_API_KEY environment variable Run the Python example that creates a Sandbox and executes code in it
Where can I get code-interpreter?
The source code and installation instructions are on GitHub: https://github.com/e2b-dev/code-interpreter

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects