Python sandboxes for llms

Minimal AI coding agent (~1,000 lines of Python) inspired by Claude Code. Works with any LLM. Think NanoGPT for coding agents. Formerly NanoCoder.

Unified execution environment for Python code, shell commands, and programmatic MCP tool calls.

High-performance AI pipeline engine with a C++ core and 50+ Python-extensible nodes. Build, debug, and scale LLM workflows with 13+ model providers, 8+ vector databases, and agent orchestration, all from your IDE. Includes VS Code extension, TypeScript/Python SDKs, and Docker deployment.

UEFN Toolbelt: 362 Python automation tools across 55 categories, with a PySide6 dashboard and an experimental, authenticated same-user loopback bridge for local AI control. Complements Epic's official UEFN MCP; Toolbelt is not exposed through Epic's MCP server.

A native Python agent CLI built on DeepAgents CLI, featuring an independent memory Agent that captures learnings after each task and delivers efficient AI coding assistance through hierarchical memory management.

Open-source, self-hosted Claude Code - a terminal AI assistant and the Python framework behind it. Tool-calling, sandboxed execution, multi-agent teams, skills, checkpoints, unlimited context - on Pydantic AI, any model.

Authorized security testing workspace. v2 TypeScript terminal product on release; v1 Python on pypi-release; Go branch is demo only.

Build an agent harness and control it end-to-end. Open-source SDK for production AI agents in Python & TypeScript - any model, any cloud.

AI-native agent harness for coding workflows by python: multi-model LLM orchestration, stateful sessions, tool governance, traceable delivery, and provider routing for GPT, Claude, DeepSeek, Qwen, Kimi, GLM, and MiniMax.

Cowork/Claude Code skill for Unreal Engine 5 development via MCP, pixel streaming, and Python editor scripting

OpenBrowser is a framework for intelligent browser automation. It combines direct CDP communication with a CodeAgent architecture, where the LLM writes Python code executed in a persistent namespace, to navigate, interact with, and extract information from web pages autonomously.

Build resilient agents.

AI coding agent with one Python core and three front-ends — headless CLI, Textual TUI, and an Electron desktop. Works with any OpenAI-compatible API, with risk-tiered permissions, event-sourced replayable sessions, and a fail-closed OS-level sandbox.

Agent Skill evaluation harness for paired variants, trace artifacts, and runner adapters

Protocol-layer harness for DeepSeek: Python witness stack — posterior verification that keeps the protocol honest. dsh doctor --node probes included.

Agent Skills marketplace: framework-aware skills for code review, documentation, test-plan generation, AI-writing detection, architectural analysis, and git workflows — for Python, Go, Rust, Elixir, React, Remix, and iOS/Swift. Works with Claude Code, Codex, and any agent that supports Agent Skills.

AG2 (formerly AutoGen): The Open-Source AgentOS.Join us at: https://discord.gg/sNGSwQME3x

MCP Server Framework and Tool Development library for building custom capabilities into agents.

🌐Web Agent Protocol (WAP) - Record and replay user interactions in the browser with MCP support

The open-source memory and observability layer for AI agents — persistent memory, loop detection, hash-chained audit trails, and a live dashboard, automatic on pip install.

Omnigent is an open-source AI agent framework and meta-harness: orchestrate Claude Code, Codex, Cursor, Pi, and custom agents — swap harnesses without rewriting, enforce policies and sandboxing, and collaborate in real time from any device.

Cross-language-safe code-intelligence MCP for AI agents -- 13 languages, family-gated call graph (CodeGraph 745 vs TSA 6 cross-language mis-wires, v1.21.0 same-session -- ~124x by count, ~390x by rate; re-verifying vs current develop). Run miswire-audit on your repo. 8 facade tools, TOON output, 100% local. Python.

Experiment task scheduling made easy.