{"$schema":"https://wellknown.network/schemas/agent-record-v1.json","schemaVersion":"1","id":"ag_wyq8msgedffa","handle":"iflow-mcp-x-plug-osworld-mcp","url":"https://wellknown.network/agents/iflow-mcp-x-plug-osworld-mcp","links":{"self":"https://wellknown.network/agents/iflow-mcp-x-plug-osworld-mcp/record.json","html":"https://wellknown.network/agents/iflow-mcp-x-plug-osworld-mcp","markdown":"https://wellknown.network/agents/iflow-mcp-x-plug-osworld-mcp/record.md","api":"https://wellknown.network/api/v1/agents/iflow-mcp-x-plug-osworld-mcp","status":"https://wellknown.network/api/v1/agents/iflow-mcp-x-plug-osworld-mcp/status","claim":"https://wellknown.network/agents/iflow-mcp-x-plug-osworld-mcp/claim","claimApi":"https://wellknown.network/api/v1/claims","claimDescriptor":"https://wellknown.network/agents/iflow-mcp-x-plug-osworld-mcp/claim.json","badge":"https://wellknown.network/agents/iflow-mcp-x-plug-osworld-mcp/badge.svg","openapi":"https://wellknown.network/openapi.json","history":"https://wellknown.network/api/v1/agents/iflow-mcp-x-plug-osworld-mcp/history","tools":"https://wellknown.network/api/v1/agents/iflow-mcp-x-plug-osworld-mcp/tools"},"ard":{"identifier":"urn:air::server:iflow-mcp-x-plug-osworld-mcp","type":"application/mcp-server-card+json"},"kind":"mcp_server","declared":{"name":"iflow-mcp_x-plug-osworld-mcp","summary":"OSWorld-MCP: A comprehensive MCP server for computer-use agents with 158 validated tools","description":"# OSWorld-MCP: Benchmarking MCP Tool Invocation in Computer-Use Agents\n\n## 🔔 Updates\n**2025-10-28:** We released our paper and project page! 🎉 \n\n📄 [Read the Paper](https://arxiv.org/abs/2510.24563)  |  🌐 [Visit the Project Page](https://osworld-mcp.github.io)\n\n---\n\n## 📑 Overview & Key Highlights\n\nOSWorld-MCP is a comprehensive and fair benchmark for evaluating computer-use agents in real-world scenarios.  \nIt jointly measures **Model Context Protocol (MCP)** tool invocation capabilities, **graphical user interface (GUI)** operation skills, and **decision-making** performance.  \nDesigned as an extension of **OSWorld**, it significantly improves realism, balance, and comparability in evaluation.\n\n**Key Features & Findings**\n- **158 validated MCP tools**, spanning **7 common applications** (LibreOffice Writer, Calc, Impress, VS Code, Google Chrome, VLC, OS utilities). Among them, **25 distractor tools** for robustness testing\n- **250 tool-beneficial tasks** → 69% of benchmark tasks benefit from MCP tools\n- Multi-round tool invocation possible, posing real decision-making challenges\n- **MCP tools boost model accuracy & efficiency** — e.g., OpenAI o3: 8.3% → 20.4% (15 steps)\n- Highest observed Tool Invocation Rate (**TIR**) = 36.3% (Claude-4-Sonnet, 50 steps) → indicating ample room for improvement\n- MCP tools improve agent metrics\n- Higher tool invocation correlates with higher accuracy\n- Combining tools introduces significant challenges\n\n**Architecture Overview**  \n\n![OSWorld-MCP Architecture](images/architecture.png)  \n*Figure: OSWorld-MCP evaluation framework integrating GUI actions and MCP tool invocations.*\n\n---\n\n## ⚙️ Installation & Usage\n\n### 1️⃣ Preparation: Code Setup\n```bash\n# Clone OSWorld base repo\ngit clone https://github.com/xlang-ai/OSWorld.git\n\n# Clone OSWorld-MCP\ngit clone https://github.com/X-PLUG/OSWorld-MCP.git\n```\nIntegrate **OSWorld-MCP** files into OSWorld to enable MCP support.\n\n---\n\n### 2️⃣ Preparation: Docker Environment\n1. Copy…","publisher":null,"homepage":null,"repository":null,"version":"0.1.2","license":null,"protocols":["mcp"],"tags":["mcp"],"pricing":null,"endpoints":[{"url":"pypi:iflow-mcp_x-plug-osworld-mcp","type":"package_pypi","auth":null,"probeable":false}],"skills":null,"tools":null,"extra":null,"attribution":{"kind":"pypi","name":"pypi","summary":"pypi","version":"pypi","description":"pypi"}},"derived":{"capabilities":[{"slug":"ai.evaluation","name":"Evaluation & Benchmarks","confidence":0.859,"provenance":"derived"},{"slug":"dev.version-control","name":"Version Control","confidence":0.745,"provenance":"derived"}],"categories":["ai","dev"],"language":"en"},"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":"iflow-mcp_x-plug-osworld-mcp","registry":"pypi","observedAt":"2026-09-15T20:21:39.694Z","publishedAt":"2026-02-12T10:10:13.306339Z","latestVersion":"0.1.2"},"toolSurface":null,"endpointFacts":[]},"verification":{"claimed":false,"claimedAt":null,"proofs":[]},"provenance":{"sources":[{"source":"pypi","key":"iflow-mcp_x-plug-osworld-mcp","url":"https://pypi.org/project/iflow-mcp_x-plug-osworld-mcp/","firstSeenAt":"2026-09-09T21:25:09.472Z","fetchedAt":"2026-09-15T20:20:52.972Z","normalizedAt":"2026-09-15T20:20:52.972Z"}]},"firstSeenAt":"2026-09-09T21:25:09.472Z","updatedAt":"2026-09-15T20:21:39.694Z"}