{"$schema":"https://wellknown.network/schemas/agent-record-v1.json","schemaVersion":"1","id":"ag_fkb4tpbpc56d","handle":"mcp-server-omniparser","url":"https://wellknown.network/agents/mcp-server-omniparser","links":{"self":"https://wellknown.network/agents/mcp-server-omniparser/record.json","html":"https://wellknown.network/agents/mcp-server-omniparser","markdown":"https://wellknown.network/agents/mcp-server-omniparser/record.md","api":"https://wellknown.network/api/v1/agents/mcp-server-omniparser","status":"https://wellknown.network/api/v1/agents/mcp-server-omniparser/status","claim":"https://wellknown.network/agents/mcp-server-omniparser/claim","claimApi":"https://wellknown.network/api/v1/claims","claimDescriptor":"https://wellknown.network/agents/mcp-server-omniparser/claim.json","badge":"https://wellknown.network/agents/mcp-server-omniparser/badge.svg","openapi":"https://wellknown.network/openapi.json","history":"https://wellknown.network/api/v1/agents/mcp-server-omniparser/history","tools":"https://wellknown.network/api/v1/agents/mcp-server-omniparser/tools"},"ard":{"identifier":"urn:air::server:mcp-server-omniparser","type":"application/mcp-server-card+json"},"kind":"mcp_server","declared":{"name":"mcp-server-omniparser","summary":"MCP Server for UI automation using Microsoft OmniParser model for screen parsing and element detection","description":"# OmniParser MCP Server \n \n[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT) \n[![Python 3.10+](https://img.shields.io/badge/python-3.10+-blue.svg)](https://www.python.org/downloads/) \n[![MCP](https://img.shields.io/badge/MCP-Compatible-green.svg)](https://modelcontextprotocol.io/) \n \n一個基於Microsoft OmniParser模型的MCP (Model Context Protocol) Server，用於UI自動化和屏幕解析。 \n \n## 🌟 功能特性 \n \n- 🖥️ **屏幕截圖解析**: 使用OmniParser模型解析UI截圖，識別可交互元素 \n- 🖱️ **鍵鼠模擬**: 支持鍵盤和滑鼠操作模擬 \n- 🪟 **視窗管理**: 連接和管理特定應用程式視窗 \n- 🌐 **多平台支持**: 支持桌面應用、瀏覽器和遊戲 \n- 🔧 **MCP協議**: 完全兼容MCP協議，可與支持MCP的AI助手集成 \n \n## 📋 系統要求 \n \n- Python 3.10+ \n- Windows 10/11 (推薦，完整功能支持) \n- CUDA支持的GPU (推薦) \n- 8GB+ RAM \n- 5GB+ 可用磁盤空間 \n \n## 🚀 快速開始 \n \n### 1. 克隆項目 \n \n```bash \ngit clone https://github.com/yourusername/omniparser-mcp-server.git \ncd omniparser-mcp-server \n``` \n \n### 2. 安裝依賴 \n \n```bash \n# 創建虛擬環境 (推薦) \npython -m venv venv \nsource venv/bin/activate  # Linux/Mac \n# 或 venv\\Scripts\\activate  # Windows \n \n# 安裝依賴 \npip install -r requirements.txt \npip install -e . \n \n# Windows 用戶需要額外安裝 \npip install pywin32 psutil \n``` \n \n### 3. 下載OmniParser模型 \n \n```bash \n# 下載模型權重 (需要先安裝 huggingface-cli) \npip install huggingface_hub \n \n# 下載到 weights 目錄 \nmkdir -p weights \nfor f in icon_detect/{train_args.yaml,model.pt,model.yaml} icon_caption/{config.json,generation_config.json,model.safetensors}; do \n    huggingface-cli download microsoft/OmniParser-v2.0 \"$f\" --local-dir weights \ndone \nmv weights/icon_caption weights/icon_caption_florence \n``` \n \n### 4. 配置設定 \n \n編輯 `config.json` 確保模型路徑正確，然後啟動服務器： \n \n```bash \nomniparser-mcp-server \n``` \n \n## 🔧 MCP 配置 \n \n### 安裝MCP服務器 \n \n```bash \n# 使用 uvx 安裝（推薦） \nuvx install mcp-server-omniparser \n \n# 或從源碼安裝 \nuvx install git+https://github.com/win10ogod/omniparser-mcp-server.git \n``` \n \n### Claude Desktop 配置 \n \n在 Claude Desktop 配置文件中添加： \n \n**配置文件位置：** \n- macOS: `~/Library/Application Support/Claude/claude_desktop_config.json` \n- Windows: `%APPDATA%\\Claude\\claude_desktop…","publisher":{"name":"Your Name","url":null},"homepage":"https://github.com/yourusername/omniparser-mcp-server","repository":"https://github.com/yourusername/omniparser-mcp-server","version":"0.1.0.3","license":"MIT","protocols":["mcp"],"tags":["mcp","ui-automation","omniparser","computer-vision","screen-parsing"],"pricing":null,"endpoints":[{"url":"pypi:mcp-server-omniparser","type":"package_pypi","auth":null,"probeable":false}],"skills":null,"tools":null,"extra":null,"attribution":{"kind":"pypi","name":"pypi","license":"pypi","repoUrl":"pypi","summary":"pypi","version":"pypi","description":"pypi","homepageUrl":"pypi","publisherName":"pypi"}},"derived":{"capabilities":[{"slug":"ai.model-access","name":"Model Access","confidence":0.905,"provenance":"derived"},{"slug":"dev.version-control","name":"Version Control","confidence":0.745,"provenance":"derived"},{"slug":"media.image-understanding","name":"Image Understanding","confidence":0.638,"provenance":"derived"}],"categories":["ai","dev","media"],"language":"fr"},"observed":{"status":"unknown","statusReason":"Distributed as a package to run locally; no network endpoint to check.","lastOkAt":null,"lastProbedAt":null,"statusComputedAt":null,"reliability30d":null,"latestObservations":[],"tools":null,"package":{"name":"mcp-server-omniparser","registry":"pypi","observedAt":"2026-10-22T11:30:47.314Z","publishedAt":"2025-06-25T04:31:54.214845Z","latestVersion":"0.1.0.3"},"toolSurface":null,"endpointFacts":[]},"verification":{"claimed":false,"claimedAt":null,"proofs":[]},"provenance":{"sources":[{"source":"pypi","key":"mcp-server-omniparser","url":"https://pypi.org/project/mcp-server-omniparser/","firstSeenAt":"2026-09-10T04:21:59.144Z","fetchedAt":"2026-10-22T11:28:35.695Z","normalizedAt":"2026-10-22T11:28:35.695Z"}]},"firstSeenAt":"2026-09-10T04:21:59.144Z","updatedAt":"2026-10-22T11:30:47.314Z"}