local-code-community

Source code of Local Code, a private AI agent for Mac, iPad and iPhone. Runs DeepSeek 4.1 on device, with Python, Node.js and a shell built into the app.

What is local-code-community?

Source code of Local Code, a private AI agent for Mac, iPad and iPhone. Runs DeepSeek 4.1 on device, with Python, Node.js and a shell built into the app.

Overview of local-code-community

What it is

Local Code is a private AI agent for Mac, iPad and iPhone, written mostly in Swift. It combines an LLM inference engine for Apple Silicon, a coding harness tuned to each model, and an agent runtime with Python, Node.js, a shell and other tools built in. The agent can work through tasks on files you choose, such as analyzing data, charting, writing documents and slide decks, or building software. It can run on-device models like DeepSeek 4.1 Flash 1017, or use cloud models from Anthropic, OpenAI, OpenRouter or an OpenAI-compatible server. When using cloud models, prompts, file contents and tool output go to that provider. The agent's tools run inside the app's sandbox and can reach only the folders you allow.

Key features

  • Durable tasks and threads: tasks and chat threads are saved, so you can close the app and pick them up later.
  • Resumable generation: an interrupted reply, reasoning included, continues where it stopped instead of starting over.
  • Steering: you can send a message at any time, even while the agent is working, to change its course.
  • Long-term context cache: the model's context is saved to disk, including images and recurrent state, so a long conversation isn't read again from the start.
  • In-app sandbox with built-in tools: shell (OSH), Python 3.14 with common libraries, Node.js 24, Lean 4 and Z3, LibreOffice, FFmpeg, web search, WebKit browser, Draw Things CLI, and personal data commands (calendar, photos, notes, mail, imessage) after granting access.
  • Built-in skills for common tasks like creating office documents, drawing charts, making videos, deploying static sites, connecting to MCP servers, and more.
  • Supports on-device models: DeepSeek 4.1 Flash 1017, DeepSeek 4 Flash 0731, Qwen 3.8 27B, with various quantizations. Model suggestion based on device memory.

Use cases

  • Analyzing data and creating charts.
  • Writing documents and slide decks.
  • Building software.
  • Proving code properties with Lean.
  • Generating images and videos.
  • Browsing the web and testing pages.
  • Reading and editing personal data (calendar, photos, notes, mail, messages) after granting access.

Getting started

See the repository README for installation and usage instructions.

License and pricing

Licensed under GPL-3.0. Open source.

Based on the project's public README and repository metadata.

Add this badge to your README

Featured on YunoDB Badge styles & help

Maintain this project? Show that it is featured on YunoDB.

Markdown[![Featured on YunoDB](https://yunodb.com/badge/drawthingsailocal-code-community.svg)](https://yunodb.com/tools/drawthingsailocal-code-community?utm_source=github&utm_medium=badge)
HTML<a href="https://yunodb.com/tools/drawthingsailocal-code-community?utm_source=github&utm_medium=badge"><img src="https://yunodb.com/badge/drawthingsailocal-code-community.svg" alt="Featured on YunoDB" height="20"></a>