Metadata-Version: 2.4
Name: holo-desktop
Version: 0.1.0
Summary: Vision-language agent that drives real macOS, Linux, and Windows apps. Powered by Holo3.
Project-URL: Homepage, https://hcompany.ai
Project-URL: Repository, https://github.com/hcompai/holo-desktop
Project-URL: Issues, https://github.com/hcompai/holo-desktop/issues
Project-URL: Documentation, https://hub.hcompany.ai/quickstart
Project-URL: Changelog, https://github.com/hcompai/holo-desktop/blob/main/CHANGELOG.md
Project-URL: Holo3-35B-A3B, https://huggingface.co/Hcompany/Holo3-35B-A3B
Project-URL: Holo3-122B-A10B, https://huggingface.co/Hcompany/Holo3-122B-A10B
Author-email: H Company <contact@hcompany.ai>
License: MIT License
        
        Copyright (c) 2026 H Company
        
        Permission is hereby granted, free of charge, to any person obtaining a copy
        of this software and associated documentation files (the "Software"), to deal
        in the Software without restriction, including without limitation the rights
        to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
        copies of the Software, and to permit persons to whom the Software is
        furnished to do so, subject to the following conditions:
        
        The above copyright notice and this permission notice shall be included in all
        copies or substantial portions of the Software.
        
        THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
        IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
        FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
        AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
        LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
        OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
        SOFTWARE.
License-File: LICENSE
Keywords: acp,agent,computer-use,desktop-automation,h-company,holo3,mcp,vlm
Classifier: Development Status :: 3 - Alpha
Classifier: Intended Audience :: Developers
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: MacOS
Classifier: Operating System :: Microsoft :: Windows
Classifier: Operating System :: POSIX :: Linux
Classifier: Programming Language :: Python :: 3.10
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Classifier: Topic :: Software Development :: Libraries :: Python Modules
Classifier: Typing :: Typed
Requires-Python: >=3.10
Requires-Dist: a2a-sdk[http-server]>=1.0.3
Requires-Dist: agent-client-protocol>=0.10
Requires-Dist: httpx>=0.27
Requires-Dist: jinja2>=3.1
Requires-Dist: mcp>=1.27
Requires-Dist: pillow>=10.0
Requires-Dist: protobuf>=4.21
Requires-Dist: pybase64>=1.4
Requires-Dist: pydantic>=2.6
Requires-Dist: python-dotenv>=1.0
Requires-Dist: pyyaml>=6.0
Requires-Dist: rich>=13.7
Requires-Dist: tyro>=0.9
Requires-Dist: uvicorn>=0.46.0
Description-Content-Type: text/markdown

# Holo Desktop

[![PyPI](https://img.shields.io/pypi/v/holo-desktop.svg)](https://pypi.org/project/holo-desktop/)
[![CI](https://github.com/hcompai/holo-desktop/actions/workflows/ci.yml/badge.svg?branch=main)](https://github.com/hcompai/holo-desktop/actions/workflows/ci.yml)
[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](LICENSE)

Tell your computer what to do. Holo gets it done. Open-source agent powered by [Holo3](https://huggingface.co/Hcompany/Holo3-35B-A3B), H Company's open-weight vision-language model. Use the hosted API, or run everything directly on your computer for full privacy.

## Quickstart

```bash
pip install holo-desktop
holo run "Catch me up on Slack"
```

On first run:

1. [`cua-driver`](https://github.com/trycua/cua) auto-installs (macOS, Linux, Windows).
2. Your browser opens to sign in at [portal.hcompany.ai](https://portal.hcompany.ai). Skip with `--base-url` for a local model.
3. macOS only: grant your terminal **Accessibility** and **Screen Recording** in *System Settings → Privacy & Security*.

## Run from the terminal

```bash
holo run "Open Safari and go to hcompany.ai"
```

Pin Holo to a specific window with `--app`:

```bash
holo run "Reply to the highlighted Slack thread" --app Slack
```

See `holo run --help` for all flags.

## Use from Python

```python
from holo_desktop import Holo

answer = Holo().run("Tell me how many unread emails I have")
print(answer)
```

The `Holo` class supports `pause`, `resume`, `stop`, and mid-run `send` for interactive embedding.

## Models

Holo defaults to the [H Company Models API](https://hcompany.ai/holo-models-api). Your first `holo run` opens your browser, signs you in at [portal.hcompany.ai](https://portal.hcompany.ai), and saves a key to `~/.holo/.env`. Run `holo login` to do this ahead of time. Holo3-35B is on the free tier; the 122B requires a paid plan.

To run everything on your own hardware instead, pass `--base-url` to any OpenAI-compatible server. No `holo login` needed, and no screenshots, keystrokes, or app content ever leave your machine.

```bash
holo run --base-url http://localhost:8000/v1 "Open Safari and go to hcompany.ai"
```

Holo3-35B-A3B fits comfortably on a recent MacBook Pro or Mac Mini at Q4. NVIDIA's [DGX Spark](https://www.nvidia.com/en-us/products/workstations/dgx-spark/) runs both the 35B and 122B at higher precision and gives you the best agent quality on a single box. Multi-GPU rigs and rack servers serve the FP8 stack at full throughput.

vLLM and llama.cpp both work:

<details>
<summary>vLLM (Holo3-35B-A3B-FP8)</summary>

Per-request `reasoning_effort` is honored via `chat_template_kwargs`; think tokens are decoded with `--reasoning-parser qwen3`.

```bash
export VLLM_ATTENTION_BACKEND=FLASHINFER
export TORCH_CUDA_ARCH_LIST=12.1a

vllm serve Hcompany/Holo3-35B-A3B-FP8 \
  --host 0.0.0.0 --port 8000 \
  --tensor-parallel-size 1 --gpu-memory-utilization 0.85 \
  --max-model-len 65537 --max-num-batched-tokens 8192 --max-num-seqs 1 \
  --kv-cache-dtype fp8 --attention-backend flashinfer --enable-prefix-caching \
  --load-format fastsafetensors \
  --enable-auto-tool-choice --tool-call-parser qwen3_coder --reasoning-parser qwen3 \
  --chat-template-content-format openai \
  --limit-mm-per-prompt '{"image": 1}' \
  --mm-processor-cache-gb 4 --mm-processor-cache-type shm \
  --trust-remote-code
```
</details>

<details>
<summary>llama.cpp (Holo3-35B-A3B GGUF)</summary>

Quants by [mradermacher/Holo3-35B-A3B-GGUF](https://huggingface.co/mradermacher/Holo3-35B-A3B-GGUF) (community).

Reasoning behavior is fixed at server launch (`--reasoning auto` separates `<think>` from content). `chat_template_kwargs` is silently ignored, so per-request `reasoning_effort` falls back to logit-bias steering on the `</think>` token.

```bash
llama-server -hf mradermacher/Holo3-35B-A3B-GGUF:Q4_K_M \
  --host 0.0.0.0 --port 8000 \
  --jinja --reasoning auto \
  -c 65536 -ngl 99 \
  --chat-template-kwargs '{"enable_thinking": true}'
```
</details>

## Use inside another agent

Holo runs as a sub-agent of Claude Code, Cursor, Codex, and other [MCP](https://modelcontextprotocol.io) / [ACP](https://agentclientprotocol.com) hosts. When your main agent needs to read a screen or click through an app, it delegates to Holo and gets the answer back.

One command wires Holo into every supported host on your machine:

```bash
holo install               # everything detected
holo install cursor        # one host
holo install list          # see what's available
```

Each host gets the MCP server in its config, plus a [Skill](https://docs.claude.com/en/docs/agents-and-tools/agent-skills/overview) (where supported) that teaches the parent when to delegate to Holo.

| id              | host                                                            | skill auto-load              |
| --------------- | --------------------------------------------------------------- | ---------------------------- |
| `claude-code`   | [Claude Code](https://docs.anthropic.com/en/docs/claude-code)   | `~/.claude/skills/`          |
| `claude-desktop`| [Claude Desktop](https://claude.ai/download)                    | —                            |
| `codex`         | [Codex](https://github.com/openai/codex)                        | `~/.agents/skills/`          |
| `copilot`       | [GitHub Copilot CLI](https://github.com/github/copilot-cli)     | —                            |
| `cursor`        | [Cursor](https://cursor.com)                                    | —                            |
| `gemini`        | [Gemini CLI](https://github.com/google-gemini/gemini-cli)       | —                            |
| `hermes`        | [Hermes](https://nousresearch.com)                              | —                            |
| `openclaw`      | [OpenClaw](https://github.com/openclaw/openclaw)                | `~/.openclaw/skills/`        |
| `opencode`      | [OpenCode](https://opencode.ai)                                 | `~/.config/opencode/skills/` |

### ACP

`holo acp` runs Holo as an [ACP](https://agentclientprotocol.com) sub-agent over stdio.

**Hermes** ([NousResearch](https://github.com/NousResearch/hermes-agent)):

```python
delegate_task(acp_command="holo acp", task="Open Authy and grab my AWS 2FA code")
```

**OpenClaw** — `~/.openclaw/openclaw.json`:

```json
{ "runtimes": { "holo": { "runtime": "acp-standard", "command": "holo", "args": ["acp"] } } }
```

**Zed** — `~/.config/zed/settings.json`:

```json
{ "agent_servers": { "Holo Desktop": { "command": "holo", "args": ["acp"] } } }
```

**JetBrains, Neovim** (avante.nvim, CodeCompanion.nvim): point any ACP client at `holo acp`.

## Develop

```bash
git clone https://github.com/hcompai/holo-desktop && cd holo-desktop
make setup
```

See [`CONTRIBUTING.md`](CONTRIBUTING.md) and [`RELEASING.md`](RELEASING.md).

## Resources

- Models: [Holo3-35B-A3B](https://huggingface.co/Hcompany/Holo3-35B-A3B) · [Holo3-122B-A10B](https://huggingface.co/Hcompany/Holo3-122B-A10B)
- Docs: [Quickstart](https://hub.hcompany.ai/quickstart) · [Models API](https://hcompany.ai/holo-models-api)
- [H Company](https://hcompany.ai)
