{"$schema":"https://wellknown.network/schemas/agent-record-v1.json","schemaVersion":"1","id":"ag_j9fhr4wsy8qj","handle":"iflow-mcp-hillyess-datahill","url":"https://wellknown.network/agents/iflow-mcp-hillyess-datahill","links":{"self":"https://wellknown.network/agents/iflow-mcp-hillyess-datahill/record.json","html":"https://wellknown.network/agents/iflow-mcp-hillyess-datahill","markdown":"https://wellknown.network/agents/iflow-mcp-hillyess-datahill/record.md","api":"https://wellknown.network/api/v1/agents/iflow-mcp-hillyess-datahill","status":"https://wellknown.network/api/v1/agents/iflow-mcp-hillyess-datahill/status","claim":"https://wellknown.network/agents/iflow-mcp-hillyess-datahill/claim","claimApi":"https://wellknown.network/api/v1/claims","claimDescriptor":"https://wellknown.network/agents/iflow-mcp-hillyess-datahill/claim.json","badge":"https://wellknown.network/agents/iflow-mcp-hillyess-datahill/badge.svg","openapi":"https://wellknown.network/openapi.json","history":"https://wellknown.network/api/v1/agents/iflow-mcp-hillyess-datahill/history","tools":"https://wellknown.network/api/v1/agents/iflow-mcp-hillyess-datahill/tools"},"ard":{"identifier":"urn:air::server:iflow-mcp-hillyess-datahill","type":"application/mcp-server-card+json"},"kind":"mcp_server","declared":{"name":"iflow-mcp_hillyess-datahill","summary":"IPython Data Analysis MCP Server - 基于 IPython 内核的轻量级数据分析 MCP 工具","description":"# IPython 数据分析 MCP 服务器 / IPython Data Analysis MCP Server\n\n[🇨🇳 中文](#中文版本) | [🇺🇸 English](#english-version)\n\n---\n\n## 中文版本\n\n基于真正 IPython 内核的轻量级数据分析 MCP (Model Context Protocol) 工具，提供完整的交互式 Python 数据分析环境，支持会话管理、数据加载、实时数据查看等核心功能。\n\n### 🚀 核心特性\n\n- **真正的 IPython 环境**: 基于 IPython InteractiveShell，支持所有 IPython 功能\n- **多会话管理**: 独立的会话空间，变量隔离，持久化状态\n- **智能数据加载**: 支持 CSV/Excel/JSON，自动编码检测，智能变量命名\n- **实时监控**: 内存使用监控、变量管理、执行历史追踪\n- **完整功能支持**: Python代码、IPython魔法命令、系统命令执行\n- **智能采样**: 大数据集友好的列数据查看，避免上下文溢出\n\n### 📋 功能清单\n\n#### 17个核心工具函数\n\n1. **会话管理**\n   - `create_ipython_session` - 创建新的 IPython 会话\n   - `list_ipython_sessions` - 列出所有活跃会话\n   - `get_session_status` - 获取会话详细状态\n   - `delete_ipython_session` - 删除指定会话\n\n2. **代码执行**\n   - `execute_code` - 执行 Python 代码、魔法命令、系统命令\n   - `get_execution_history` - 获取执行历史记录\n\n3. **数据加载**\n   - `load_csv_file` - 加载 CSV 文件（自动编码检测）\n   - `load_excel_file` - 加载 Excel 文件（支持 .xlsx/.xls）\n   - `load_json_file` - 加载 JSON 文件\n\n4. **数据操作与查看**\n   - `list_dataframes` - 列出会话中所有 DataFrame\n   - `get_dataframe_info` - 获取 DataFrame 详细信息\n   - `preview_dataframe` - 预览 DataFrame 数据\n   - `get_dataframe_summary` - 获取统计摘要\n   - `sample_column_data` - 智能采样查看列数据\n\n5. **内存与变量管理**\n   - `check_memory_usage` - 检查内存使用情况\n   - `get_variable_info` - 获取变量详细信息\n   - `clear_variables` - 清理变量释放内存\n\n### 🛠️ 安装配置\n\n#### 方法一：使用 uvx 直接运行（推荐）\n\n无需克隆项目，直接使用 uvx 从 GitHub 运行：\n\n```bash\n# 安装 uvx（如果还没有安装）\npip install uvx\n\n# 直接运行 MCP 服务器\nuvx --from git+https://github.com/Hillyess/dataHill.git DATA_MCP.py\n```\n\n#### 方法二：本地安装开发\n\n```bash\n# 1. 克隆项目\ngit clone git@github.com:Hillyess/dataHill.git\ncd dataHill\n\n# 2. 创建虚拟环境\nconda create -n data-analyzer python=3.10\nconda activate data-analyzer\n\n# 3. 安装依赖\npip install -r requirements.txt\n\n# 4. 测试安装\npython DATA_MCP.py\n```\n\n#### 配置 MCP 客户端\n\n##### Claude Desktop 配置\n\n编辑 Claude Desktop 配置文件：\n\n**macOS**: `~/Library/Application Support/Claude/claude_desktop_config.json`\n**Windows**: `%APPDATA%\\Claude\\claude_desktop_config.json`\n\n**推荐配置（使用 uvx）**：\n```json\n{\n  \"mcpServers\": {\n    \"…","publisher":{"name":"Hillyess","url":null},"homepage":"https://github.com/iflow-mcp/hillyess-datahill","repository":"https://github.com/iflow-mcp/hillyess-datahill","version":"0.1.0","license":"MIT","protocols":["mcp"],"tags":["data-analysis","ipython","mcp","model-context-protocol","pandas"],"pricing":null,"endpoints":[{"url":"pypi:iflow-mcp_hillyess-datahill","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":"documents.spreadsheets","name":"Spreadsheets","confidence":1,"provenance":"derived"},{"slug":"dev.version-control","name":"Version Control","confidence":0.745,"provenance":"derived"}],"categories":["dev","documents"],"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":"iflow-mcp_hillyess-datahill","registry":"pypi","observedAt":"2026-10-28T04:28:31.393Z","publishedAt":"2026-02-07T22:28:11.069601Z","latestVersion":"0.1.0"},"toolSurface":null,"endpointFacts":[]},"verification":{"claimed":false,"claimedAt":null,"proofs":[]},"provenance":{"sources":[{"source":"pypi","key":"iflow-mcp_hillyess-datahill","url":"https://pypi.org/project/iflow-mcp_hillyess-datahill/","firstSeenAt":"2026-09-09T19:23:37.275Z","fetchedAt":"2026-10-28T04:27:39.465Z","normalizedAt":"2026-10-28T04:27:39.465Z"}]},"firstSeenAt":"2026-09-09T19:23:37.275Z","updatedAt":"2026-10-28T04:28:31.393Z"}