Metadata-Version: 2.4
Name: punt-lux
Version: 0.18.0
Summary: The paintbrush for Claude. A visual output surface for AI agents — diagrams, tables, images, and interactive controls.
Author: Punt Labs
License-Expression: MIT
Classifier: Development Status :: 3 - Alpha
Classifier: Intended Audience :: Developers
Classifier: License :: OSI Approved :: MIT License
Classifier: Programming Language :: Python :: 3.13
Classifier: Programming Language :: Python :: 3.14
Classifier: Topic :: Multimedia :: Graphics
Classifier: Topic :: Software Development :: Libraries
Requires-Dist: typer>=0.15.0,<1
Requires-Dist: rich>=13.0.0,<14
Requires-Dist: fastmcp>=3.0.0,<4
Requires-Dist: pydantic>=2.0.0,<3
Requires-Dist: starlette>=0.46.0
Requires-Dist: uvicorn>=0.34.0
Requires-Dist: ruff>=0.13.0 ; extra == 'dev'
Requires-Dist: mypy>=1.18.1 ; extra == 'dev'
Requires-Dist: pyright>=1.1.400 ; extra == 'dev'
Requires-Dist: pytest>=8.4.2 ; extra == 'dev'
Requires-Dist: pytest-cov>=6.0.0 ; extra == 'dev'
Requires-Dist: imgui-bundle==1.92.600 ; extra == 'display'
Requires-Dist: pillow>=11.0.0 ; extra == 'display'
Requires-Dist: numpy>=2.0.0 ; extra == 'display'
Requires-Dist: pyopengl>=3.1.0 ; extra == 'display'
Requires-Dist: setproctitle>=1.3.0 ; extra == 'display'
Requires-Dist: pyobjc-framework-cocoa>=10.0 ; sys_platform == 'darwin' and extra == 'display'
Requires-Dist: pyobjc-framework-quartz>=10.0 ; sys_platform == 'darwin' and extra == 'display'
Requires-Python: >=3.13
Project-URL: Homepage, https://github.com/punt-labs/lux
Project-URL: Repository, https://github.com/punt-labs/lux
Project-URL: Bug Tracker, https://github.com/punt-labs/lux/issues
Provides-Extra: dev
Provides-Extra: display
Description-Content-Type: text/markdown

# lux

> A visual output surface for AI agents.

[![License](https://img.shields.io/github/license/punt-labs/lux)](LICENSE)
[![CI](https://img.shields.io/github/actions/workflow/status/punt-labs/lux/test.yml?label=CI)](https://github.com/punt-labs/lux/actions/workflows/test.yml)
[![PyPI](https://img.shields.io/pypi/v/punt-lux)](https://pypi.org/project/punt-lux/)
[![Python](https://img.shields.io/pypi/pyversions/punt-lux)](https://pypi.org/project/punt-lux/)

Lux gives AI agents a window they can draw into. It runs an ImGui display server on the local machine, connected by Unix socket IPC. Agents send JSON element trees via MCP tools; the display renders them at 60fps. The protocol is the API surface --- if an agent can describe it as JSON, Lux renders it.

The design follows Smalltalk's Morphic model: every visible element is a composable, nestable object. Windows contain tabs, tabs contain groups, groups contain buttons and plots. The long-term goal is a live environment where the MCP server is the message bus and Lux is the rendering layer, with the agent as the programmer at the keyboard.

**Platforms:** macOS, Linux

**Stage:** alpha --- protocol is stable, published on PyPI as `punt-lux`

*A Claude Code plugin displaying a project issue board --- the agent fetches live data from DoltDB via `bd list --json`, builds a filterable table with detail panel, and renders it in a single tool call. Filters and row selection run at 60fps with zero MCP round-trips.*

![Beads issue board with filterable table and detail panel](docs/beads-board.png)

*The same list/detail pattern generalizes to any tabular data. Search, combo filters, pagination, and a detail panel --- all driven by a single `show_table()` call.*

![Filterable data explorer with detail panel](docs/data-explorer-filtered.png)

*Dashboards compose metric cards, charts, and tables. `show_dashboard()` builds the layout from structured data --- no manual element positioning needed.*

![Dashboard with metrics, charts, and data table](docs/dashboard.png)

## Quick Start

```bash
curl -fsSL https://raw.githubusercontent.com/punt-labs/lux/8553ef7/install.sh | sh
```

Restart Claude Code twice. The Lux display window opens automatically when agents send visual output.

<details>
<summary>Manual install (if you already have uv)</summary>

```bash
uv tool install 'punt-lux[display]'
```

Then install the plugin via the marketplace:

```bash
claude plugin marketplace add punt-labs/claude-plugins
claude plugin install lux@punt-labs
```

</details>

<details>
<summary>Lightweight install (library use only)</summary>

If you only need `DisplayClient` to send scenes from Python (no display server):

```bash
uv add punt-lux
```

This pulls ~2 MB of lightweight deps. The 66 MB display stack (imgui-bundle, numpy, Pillow, PyOpenGL) is only needed for `lux display` and is available via `punt-lux[display]`.

</details>

<details>
<summary>Verify before running</summary>

```bash
curl -fsSL https://raw.githubusercontent.com/punt-labs/lux/8553ef7/install.sh -o install.sh
shasum -a 256 install.sh
cat install.sh
sh install.sh
```

</details>

<details>
<summary>Run a demo</summary>

```bash
lux display &
uv run python demos/dashboard.py
```

Demos are in `demos/` --- each connects as a client and drives the display:

| Demo | What it shows |
|------|--------------|
| `interactive.py` | Sliders, checkboxes, combos, text inputs, color pickers |
| `containers.py` | Windows, tab bars, collapsing headers, groups |
| `dashboard.py` | Multi-window layout with draw canvases and live controls |
| `data_viz.py` | Tables, plots, progress bars, spinners, markdown |
| `menu_bar.py` | Custom menus, event handling, periodic refresh |

</details>

## Features

- **24 element kinds** --- text, buttons (arrow, small), images, sliders, checkboxes, combos, inputs (text, number), radios, color pickers (alpha, full picker), selectables, trees, tables, plots, progress bars, spinners, markdown, draw canvases, modals, groups, tab bars, collapsing headers, windows, separators
- **Frames** --- scenes target named frames (inner windows) via `frame_id`. Frames persist after disconnect, can be adopted by new clients, and support initial sizing (`frame_size`) and ImGui window flags (`frame_flags`)
- **Layout nesting** --- windows contain tab bars contain groups contain any element, arbitrarily deep
- **Incremental updates** --- `update` patches individual elements by ID without replacing the scene
- **World menu** --- per-client namespaced menus. Each connected MCP server gets its own submenu. Items registered via `register_tool` are routed only to the owning client
- **Interaction events** --- button clicks, slider changes, table row selections, menu clicks queue as events the agent reads via `recv`
- **Frame auto-focus** --- frames automatically focus (brought to front) when they receive a scene update
- **Persistent tabs** --- each `show()` call opens a dismissable tab; same `scene_id` replaces content in-place. Users can close individual tabs
- **Themes** --- 11 themes via `set_theme`: `imgui_colors_dark`, `imgui_colors_light`, `imgui_colors_classic`, `darcula`, `darcula_darker`, `material_flat`, `photoshop_style`, `grey_flat`, `cherry`, `light_rounded`, `microsoft_style`
- **Auto-spawn** --- `DisplayClient` starts the display server on first connection if it isn't running
- **Unix socket IPC** --- length-prefixed JSON frames, no HTTP overhead, no threads

## MCP Tools

Agents interact with Lux through 11 MCP tools exposed by `lux serve`:

| Tool | What it does |
|------|-------------|
| `show(scene_id, elements)` | Replace the display with a new element tree. Supports `frame_id`, `frame_size`, `frame_flags` for windowed frames |
| `show_table(scene_id, columns, rows)` | Display a filterable data table with optional detail panel |
| `show_dashboard(scene_id, ...)` | Display a dashboard with metric cards, charts, and a table |
| `update(scene_id, patches)` | Patch elements by ID (set fields or remove) |
| `set_menu(menus)` | Add custom menus to the menu bar |
| `register_tool(id, label)` | Register a World menu item routed only to the calling server via `recv()` |
| `set_theme(theme)` | Switch display theme (11 themes available) |
| `display_mode(mode)` | Get or set display mode (`y`/`n`) --- advisory signal for consumer plugins |
| `clear()` | Remove all content from the display |
| `ping()` | Round-trip latency check |
| `recv(timeout)` | Read the next interaction event (clicks, row selections, menu clicks) |

## What It Looks Like

### Show text and a button

```json
{"tool": "show", "input": {
  "scene_id": "hello",
  "elements": [
    {"kind": "text", "id": "t1", "content": "Hello from the agent"},
    {"kind": "button", "id": "b1", "label": "Click me"}
  ]
}}
```

Returns `"ack:hello"`. When the user clicks the button:

```json
{"tool": "recv", "input": {"timeout": 5.0}}
```

Returns `"interaction:element=b1,action=click,value=True"`.

### Multi-window dashboard

```json
{"tool": "show", "input": {
  "scene_id": "dash",
  "elements": [
    {"kind": "window", "id": "w1", "title": "Controls", "x": 10, "y": 10,
     "children": [
       {"kind": "slider", "id": "vol", "label": "Volume", "value": 50}
     ]},
    {"kind": "window", "id": "w2", "title": "Chart", "x": 320, "y": 10,
     "children": [
       {"kind": "plot", "id": "p1", "title": "Trend",
        "series": [{"label": "y", "type": "line",
          "x": [1,2,3,4], "y": [10,20,15,25]}]}
     ]}
  ]
}}
```

### Update a single element

```json
{"tool": "update", "input": {
  "scene_id": "dash",
  "patches": [
    {"id": "vol", "set": {"value": 75}}
  ]
}}
```

## Element Kinds

| Category | Kinds |
|----------|-------|
| Display | `text`, `button` (arrow, small variants), `image`, `separator` |
| Interactive | `slider`, `checkbox`, `combo`, `input_text`, `input_number`, `radio`, `color_picker` (alpha, picker modes) |
| Lists | `selectable`, `tree` |
| Data | `table`, `plot`, `progress`, `spinner`, `markdown` |
| Canvas | `draw` (line, rect, circle, triangle, polyline, text, bezier) |
| Layout | `group`, `tab_bar`, `collapsing_header`, `window`, `modal` |

All elements with an `id` support an optional `tooltip` field (string shown on hover).

## CLI Commands

| Command | What it does |
|---------|-------------|
| `lux display` | Start the display server (ImGui window) |
| `lux serve` | Start the MCP server (stdio transport) |
| `lux enable` | Enable visual output for this project |
| `lux disable` | Disable visual output for this project |
| `lux status` | Check if the display server is running |
| `lux doctor` | Check installation health (Python, fonts, plugin) |
| `lux install` | Install the Claude Code plugin via the marketplace |
| `lux uninstall` | Uninstall the Claude Code plugin |
| `lux show beads` | Display the beads issue board (no LLM needed) |
| `lux version` | Print version |

## Architecture

```text
Agent (Claude Code)
  │ MCP (stdio)
  ▼
lux serve (FastMCP)
  │ Unix socket (JSON frames)
  ▼
lux display (ImGui + OpenGL)
  │ renders at 60fps
  ▼
Window on screen
```

The display server and MCP server are separate processes. The MCP server is a thin adapter that translates MCP tool calls into protocol messages sent over the Unix socket. The display server runs the ImGui render loop, polls the socket each frame via `select()` with zero timeout, and renders whatever scene the agent last sent.

Client code can also use `DisplayClient` directly as a Python library, bypassing MCP. The demos do this.

## Documentation

[Architecture (PDF)](docs/architecture.pdf) |
[Design Log](DESIGN.md) |
[Changelog](CHANGELOG.md) |
[Contributing](CONTRIBUTING.md)

## Development

```bash
uv sync --extra dev            # Install dependencies
uv run ruff check .            # Lint
uv run ruff format --check .   # Check formatting
uv run mypy src/ tests/        # Type check (mypy)
uv run pyright                 # Type check (pyright)
uv run pytest                  # Test
```

## Acknowledgements

Lux is a thin orchestration layer. The rendering is done by [Dear ImGui](https://github.com/ocornut/imgui), Omar Cornut's immediate-mode GUI library. ImGui handles all the hard problems --- text layout, widget state, input handling, GPU rendering --- and does so in a single-pass retained-mode-free architecture that maps naturally to Lux's "send JSON, render this frame" model. The 60fps render loop, the composable widget tree, and the ability to drive a full UI from a socket with no threading are all consequences of ImGui's design.

Python bindings come from [imgui-bundle](https://github.com/pthom/imgui_bundle) by Pascal Thomet, which packages ImGui, ImPlot, and several other ImGui extensions into a single pip-installable wheel with complete type stubs. imgui-bundle is what makes "install one Python package, get a GPU-accelerated UI" possible.

[FastMCP](https://github.com/jlowin/fastmcp) provides the MCP server layer.

## License

MIT
