# Mininglamp-AI/Mano-P

Mano-P: Open-source GUI-VLA agent for edge devices. #1 on OSWorld (specialized, 58.2%). Runs locally on Apple M4 Mac mini/MacBook — no data leaves your device.Mano-P 是一个开源 GUI-VLA 项目，支持在 Mac mini/MacBook 上或通过算力棒本地运行推理，实现纯视觉驱动的跨平台 GUI 自动化操作。数据完全本地处理，支持复杂多步骤任务规划与执行。

Repository: https://github.com/Mininglamp-AI/Mano-P
Canonical: https://ross.abutalabs.com/products/mano-p
License: Apache-2.0
License Family: permissive
Topics: computer-use-agents, desktop-automation, edge-computing, gui-automation, vision-language-action, multimodal-ai, gui-grounding, local-inference, mano, on-device-ai, osworld, visual-language-model, mano-p
Last push: 2026-06-25T03:00:01+00:00

## Health v2 (maintenance only)
Score: 55/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 89, release rhythm 35, longevity 11
- inputs: {"age_days": 161, "days_push": 69, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, young
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2589, forks 242 (observed 2026-08-28T04:07:02.820709+00:00)

## What it is
Mano-P is an open-source GUI-VLA (vision-language-action) agent model and SDK for edge devices, enabling purely vision-driven cross-platform GUI automation. It runs inference locally on Apple Silicon Macs or compute sticks so that all computer-use data stays on-device.

## Use cases
- automate desktop GUI tasks with a vision agent
- run a computer-use agent fully locally on a Mac mini
- build custom CUA skills and workflows
- perform multi-step task planning and execution via screenshots
- keep screen automation data private on-device
- benchmark GUI agents on OSWorld

## When to choose
- you need privacy-first GUI automation with no data leaving the device
- you want to run a computer-use agent on Apple Silicon or edge hardware
- you're building custom agent skills on top of a GUI-VLA model
- you need cross-platform visual GUI grounding without accessibility APIs

## When to avoid
- you need a cloud-scale agent with large frontier-model reasoning
- your automation targets only mobile or web APIs rather than desktop GUIs
- you lack Apple Silicon or compatible edge hardware for local inference
- you need extensive ecosystem maturity beyond the phased open-source release

## Facets
- artifact type: library
- maturity: active
- function: agent-framework, computer-vision, machine-learning, llm-inference, workflow-automation
- domain: computer-vision, privacy, developer-tools
- platform: python, cross-platform
- tags: gui-automation, vision-language-action, computer-use-agent, edge-devices, local-inference, on-device-ai, osworld, desktop-automation, gui-grounding, ai-agents, automation, macos, gpu, desktop

## Member repositories
- Mininglamp-AI/Mano-P (main) score 55

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:02.820709+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T02:22:21.477345+00:00, confidence not recorded.
  - readme: https://github.com/Mininglamp-AI/Mano-P (fetched 2026-08-28T04:07:02.820709+00:00, sha badc1a576416)
- Data as of 2026-08-30T08:39:29.467469+00:00.