{
  "markdown": "# glass\n\n[![CI](https://github.com/fixed-width/glass/actions/workflows/ci.yml/badge.svg)](https://github.com/fixed-width/glass/actions/workflows/ci.yml)\n[![glass MCP server](https://glama.ai/mcp/servers/fixed-width/glass/badges/score.svg)](https://glama.ai/mcp/servers/fixed-width/glass)\n\n**Give your coding agent hands for the app it's building** — launch it, drive it, and verify the\nresult, without burning a screenshot on every step.\n\nA Rust [MCP](https://modelcontextprotocol.io) server that gives an AI coding agent a closed **build →\nsee → interact → debug** loop over external native GUI applications.\n\nglass lets an agent launch a GUI app, capture what is on screen, inject mouse and keyboard input, read\nthe app's logs, and detect visual changes — so a coding agent can build and debug UI applications\nindependently instead of asking the user \"does this look right?\".\n\nglass drives apps as an external black box, so it works with any native GUI app regardless of toolkit\nor language. It has two Linux backends (**X11** and **Wayland**), a **Windows** backend, an\n**Android** backend (an AVD emulator, driven over `adb` from any host), an **iOS** backend (native\napps in the Simulator over `xcrun simctl`, with input and the accessibility tree via `idb_companion`,\nincluding a two-finger pinch), and a **macOS** backend, behind a platform-agnostic core.\n\n## See it\n\n![An agent debugging a GTK app under glass](docs/assets/hero-debug-loop.gif)\n\nAn agent building a GUI app runs it under glass, reproduces a bug **from the accessibility tree**\n(no screenshots), fixes the code, and re-verifies — the loop it otherwise can't close on its own.\n[Try it yourself](#try-it-in-60-seconds).\n\n## Try it in 60 seconds\n\n1. Download glass for your platform from the [Releases page](https://github.com/fixed-width/glass/releases/latest)\n   and [connect it to your agent](docs/how-to/connect-an-agent.md) — glass works with any MCP host\n   ([see which are verified](docs/reference/host-compatibility.md)).\n2. Get the example app — clone this repo, or download\n   [`examples/tasks-demo/tasks_demo.py`](examples/tasks-demo/tasks_demo.py) (on Linux it needs\n   `sudo apt install python3-gi gir1.2-gtk-4.0`).\n3. Paste this to your agent:\n\n   > Use glass to run `examples/tasks-demo/tasks_demo.py` with accessibility on. There's a bug: clicking\n   > **Add** doesn't add the typed task. Reproduce it by driving the UI and checking the\n   > accessibility tree (don't just screenshot), then find and fix the bug in the code and verify a\n   > task actually appears.\n\nYour agent launches the app, reproduces the bug from the accessibility tree, fixes the one-line\nwiring bug, and confirms the task appears — the whole build → see → interact → debug loop, start to\nfinish. (`glass-mcp doctor` checks your environment if anything's off.)\n\n## The loop in practice\n\nAn agent can fill in a form, click Save, and check that the app reports success. When the app exposes\nan accessibility tree, the agent works with named controls and verifies changes from text:\n\n```jsonc\nglass_start { \"run\": [\"python3\", \"app.py\"] }\nglass_do { \"actions\": [\n  { \"action\": \"set_value\",\n    \"target\": { \"query\": \"account\", \"role\": \"TextField\", \"states\": [\"enabled\"] },\n    \"text\": \"hello\" },\n  { \"action\": \"click_element\",\n    \"target\": { \"query\": \"Save\", \"role\": \"Button\", \"states\": [\"enabled\"] },\n    \"mode\": \"auto\" },\n  { \"action\": \"wait_for_element\", \"name\": \"Saved\" }\n] }\nglass_logs\n```\n\nUse `glass_find_elements` to inspect candidates when the intended target is not unique or known well\nenough to act on directly. Use `glass_a11y_snapshot` to explore the app's controls. See the\n[tool reference](docs/reference/tools.md#glass_click_element) for click modes and platform behavior.\n\nFor a canvas or custom-rendered app with no accessibility tree, drive it by pixels instead —\n`glass_screenshot`, `glass_click {x,y}`, and `glass_diff`, which returns `changed_pct` + a `bbox` as\ntext, so routine checks between screenshots cost no vision tokens. Why the loop is shaped this way:\n[the build → see → interact → debug loop](docs/explanation/the-loop.md).\n\nTo review a run after the app stops, opt into [session evidence recording](docs/how-to/record-session-evidence.md)\nwith `--trace-dir`. Glass retains tool inputs and requested results in bounded storage, then\n`glass-mcp trace inspect` and `trace export` validate the trace and package its evidence in a ZIP.\nRecording is off by default; retained app content can contain sensitive data.\n\n## Install at a glance\n\nDownload the latest build for your platform from the\n[Releases page](https://github.com/fixed-width/glass/releases/latest), then set up your host:\n\n- **Linux** — [docs/how-to/setup-linux.md](docs/how-to/setup-linux.md) (X11 or Wayland; `Xvfb` /\n  `sway` + bubblewrap)\n- **Windows** — [docs/how-to/setup-windows.md](docs/how-to/setup-windows.md) (a prebuilt `.exe` +\n  Sandboxie)\n- **macOS** — [docs/how-to/setup-macos.md](docs/how-to/setup-macos.md) (install the notarized `.dmg`;\n  no build needed)\n- **Android** — [docs/how-to/setup-android.md](docs/how-to/setup-android.md) (an AVD emulator, from any\n  host)\n- **iOS** — [docs/how-to/setup-ios.md](docs/how-to/setup-ios.md) (the Simulator, macOS host only)\n\nEvery asset is listed in [docs/reference/platforms.md](docs/reference/platforms.md#release-artifacts).\nPrefer to compile, or on an architecture with no published asset? See\n[docs/how-to/build-from-source.md](docs/how-to/build-from-source.md) — it is a single `cargo build`.\n\nThen [connect glass to your agent](docs/how-to/connect-an-agent.md) and run `glass-mcp doctor` to check\nthe environment. New here? Follow [the tutorial](docs/tutorial/first-drive.md) for a guaranteed first\nsuccess.\n\nThe full toolbox remains the default. For fewer agent tool definitions, start with\n`glass-mcp --tool-profile lean` and use `glass_do` for actions. Inspect either inventory without\nstarting a session using `glass-mcp tools --json`; see [tool profiles](docs/reference/tools.md#tool-profiles).\n\n## Drive it well — the `glass-drive` skill\n\nglass needs no app integration and no skill to run, but an agent drives it far more reliably with the\nopen [glass-drive](docs/how-to/drive-glass-well.md) Agent Skill — it stops the agent spending its first\nturns rediscovering the verify-cheaply-then-look loop. **Installing it is the single highest-leverage\nthing you can add** when pointing an agent at glass.\n\n## Platform support\n\n**✓** supported · **◑** partial · **–** not supported · **🚧** planned.\n\n<!-- KEEP IN SYNC with docs/reference/platforms.md (the canonical matrix) and the code. -->\n\n| Capability | Linux (X11 + Wayland) | Windows | Android (AVD) | iOS (Simulator) | macOS |\n|---|:--:|:--:|:--:|:--:|:--:|\n| Capture · input · windows · clipboard · logs | ✓ | ✓ | ✓ | ✓ | ✓ |\n| Accessibility (semantic addressing) | ✓ AT-SPI | ✓ UI Automation | ✓ UIAutomator | ✓ idb | ✓ AX |\n| Containment / sandboxing | ✓ bubblewrap | ✓ Sandboxie | ✓ the emulator VM | ✓ the Simulator | ✓ Seatbelt |\n| Display isolation (app off your desktop) | ✓ headless Xvfb / sway | ◑ virtual display · VM tier | ✓ headless emulator | ✓ headless simctl boot | 🚧 |\n\nFull matrix, per-capability detail, and system requirements:\n[docs/reference/platforms.md](docs/reference/platforms.md). Transport is MCP over stdio (default) or\nnetwork HTTP.\n\n## Documentation\n\nThe full docs — tutorial, how-to guides, reference, and explanations — are under\n**[`docs/`](docs/README.md)**. See [`CHANGELOG.md`](CHANGELOG.md) for release notes, and\n[Stability and versioning](docs/reference/stability.md) for what a 1.0 release guarantees.\n\nContributing? [`CONTRIBUTING.md`](CONTRIBUTING.md) has the gates a PR has to pass, and\n[Verify a change](docs/how-to/verify-a-change.md) covers platform code your own host cannot run.\n\n## License\n\nglass is **open core**, licensed **Apache-2.0** — see [`LICENSE-APACHE`](LICENSE-APACHE).\n",
  "bytes": 7908,
  "sha": "0bac31559e92079331b0261d78eb4a1ca6f751c3d43c85dff1e91118e484e4c3",
  "repo_slug": "fixed-width/glass",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_io_github_fixed_width_glass_9c7c8356/readme"
}