AppWorld: A Controllable World of Apps and People for Benchmarking Function Calling and Interactive Coding Agents
How to install appworld?
stonybrooknlp/appworld
$git clone https://github.com/stonybrooknlp/appworldInstalls into the current project.
›Prefer a prompt? Paste this to your agent
Install appworld by running `git clone https://github.com/stonybrooknlp/appworld`, then use it for the current task and follow its documentation at https://github.com/stonybrooknlp/appworld.
README.md
| 1 | <h2 align="center"> |
| 2 | <a href="https://appworld.dev/"><img src="/images/banner-light.svg" alt="AppWorld Banner" width="100%"></a> |
| 3 | </h2> |
| 4 | |
| 5 | <div> |
| 6 | <h3 align="center">A Controllable World of Apps and People for <br> Benchmarking Function Calling & Interactive Coding Agents <br><br>🏆 ACL'24 Best Resource Paper 🏆</h3> |
| 7 | <p align="center"> |
| 8 | :link: <a href="https://appworld.dev/">Website</a> |
| 9 | :earth_americas: <a href="https://appworld.dev/task-explorer">Task Explorer</a> |
| 10 | :hammer_and_wrench: <a href="https://appworld.dev/api-explorer">API Explorer</a> |
| 11 | :bar_chart: <a href="https://appworld.dev/leaderboard">Leaderboard</a> |
| 12 | <br> |
| 13 | :movie_camera: <a href="https://appworld.dev/video">Videos</a> |
| 14 | :bird: <a href="https://x.com/harsh3vedi/status/1818311843976233198">Tweet</a> |
| 15 | :speech_balloon: <a href="https://towardsdatascience.com/appworld-a-controllable-world-of-apps-and-people-for-benchmarking-interactive-coding-agents-37517dd9d498">Blog</a> |
| 16 | :page_facing_up: <a href="https://arxiv.org/abs/2407.18901">Paper</a> |
| 17 | </p> |
| 18 | </div> |
| 19 | |
| 20 | <h2 align="center"> |
| 21 | <a href="https://github.com/StonyBrookNLP/appworld/actions/workflows/ci.yml"> |
| 22 | <img src="https://github.com/StonyBrookNLP/appworld/actions/workflows/ci.yml/badge.svg" alt="AppWorld CI"> |
| 23 | </a> |
| 24 | <a href="https://www.python.org/"> |
| 25 | <img alt="Build" src="https://img.shields.io/badge/Python-3.11+-1f425f.svg?color=purple", alt="Python version" |
| 26 | </a> |
| 27 | <a href="https://github.com/charliermarsh/ruff"> |
| 28 | <img alt="Build" src="https://img.shields.io/badge/linter-ruff-green" alt="Ruff"> |
| 29 | </a> |
| 30 | <a href="https://mypy-lang.org"> |
| 31 | <img alt="Build" src="https://img.shields.io/badge/type%20checked-mypy-039dfc" alt="MyPy"> |
| 32 | </a> |
| 33 | <a href="https://opensource.org/licenses/Apache-2.0"> |
| 34 | <img alt="License" src="https://img.shields.io/badge/License-Apache_2.0-blue.svg" alt="Apache 2.0 License"> |
| 35 | </a> |
| 36 | <a href="https://pypi.org/project/appworld/"> |
| 37 | <img src="https://badge.fury.io/py/appworld.svg" alt="PyPI Version"> |
| 38 | </a> |
| 39 | <a href="https://pepy.tech/projects/appworld"> |
| 40 | <img src="https://img.shields.io/pepy/dt/appworld" alt="PyPI Downloads"> |
| 41 | </a> |
| 42 | <a href="http://github.com/stonybrooknlp/appworld"> |
| 43 | <img src="https://img.shields.io/badge/PRs-welcome-black.svg" alt="PRs welcome"> |
| 44 | </a> |
| 45 | <a href="https://github.com/stonybrooknlp/appworld#electric_plug-introducing-appworld-mcp-server-and-client"> |
| 46 | <img src="https://badge.mcpx.dev?type=server" title="MCP Server"/> |
| 47 | </a> |
| 48 | <a href="https://github.com/stonybrooknlp/appworld#electric_plug-introducing-appworld-mcp-server-and-client"> |
| 49 | <img src="https://badge.mcpx.dev?type=client" title="MCP Client"/> |
| 50 | </a> |
| 51 | </h2> |
| 52 | |
| 53 | - :information_source: [About](#information_source-about) |
| 54 | - :racing_car: [TLDR](#racing_car-tldr) |
| 55 | - :floppy_disk: [Installation](#floppy_disk-installation) |
| 56 | - :rotating_light: [Release Disclaimer](#rotating_light-release-disclaimer) |
| 57 | - :earth_africa: [AppWorld Walkthrough](#earth_africa-appworld-walkthrough) |
| 58 | - :wave: [Primer](#wave-primer) |
| 59 | - :globe_with_meridians: [Task Worlds](#globe_with_meridians-task-worlds) |
| 60 | - :repeat: [Interactive Coding and API Calls](#repeat-interactive-coding-and-api-calls) |
| 61 | - :runner: [A Minimal Agent in Action](#runner-a-minimal-agent-in-action) |
| 62 | - :bar_chart: [Evaluating the Agent](#bar_chart-evaluating-the-agent) |
| 63 | - :no_entry_sign: [Agent Development Restrictions](#no_entry_sign-agent-development-restrictions) |
| 64 | - :lock: [Code Execution Safety](#lock-code-execution-safety) |
| 65 | - :satellite: [Serving AppWorld Environment/APIs with/out Docker](#satellite-serving-appworld-environmentapis-without-docker) |
| 66 | - :electric_plug: [Introducing AppWorld MCP Server and Client](#electric_plug-introducing-appworld-mcp-server-and-client) (⭐ **NEW** ⭐) |
| 67 | - :link: [Starting MCP Server](#link-starting-mcp-server) |
| 68 | - :robot: [Co |
Related
6 picksType
- langchain-mcp-adaptersMake Anthropic Model Context Protocol (MCP) tools compatible with LangChain and LangGraph agents.Jul 2026
- mcporterTypeScript runtime and CLI for connecting to configured Model Context Protocol servers.Jul 2026
- mcp-proxyA MCP server which proxies requests to a remote MCP server over streamable HTTP or SSE.Jul 2026
- mcp-atlassianThe Model Context Protocol (MCP) Atlassian integration is an open-source implementation that bridges Atlassian products (Jira and Confluence) with AI language models following Anthropic's MCP specification.Jul 2026
- open-webuiOpen WebUIJul 2026
- rulesyncUnified AI rules management CLI tool that generates configuration files for various AI development toolsJul 2026
