{"$schema":"https://wellknown.network/schemas/agent-record-v1.json","schemaVersion":"1","id":"ag_rzetmg44n72w","handle":"spss-mcp","url":"https://wellknown.network/agents/spss-mcp","links":{"self":"https://wellknown.network/agents/spss-mcp/record.json","html":"https://wellknown.network/agents/spss-mcp","markdown":"https://wellknown.network/agents/spss-mcp/record.md","api":"https://wellknown.network/api/v1/agents/spss-mcp","status":"https://wellknown.network/api/v1/agents/spss-mcp/status","claim":"https://wellknown.network/agents/spss-mcp/claim","claimApi":"https://wellknown.network/api/v1/claims","claimDescriptor":"https://wellknown.network/agents/spss-mcp/claim.json","badge":"https://wellknown.network/agents/spss-mcp/badge.svg","openapi":"https://wellknown.network/openapi.json","history":"https://wellknown.network/api/v1/agents/spss-mcp/history","tools":"https://wellknown.network/api/v1/agents/spss-mcp/tools"},"ard":{"identifier":"urn:air::server:spss-mcp","type":"application/mcp-server-card+json"},"kind":"mcp_server","declared":{"name":"spss-mcp","summary":"MCP server for IBM SPSS Statistics (macOS + Windows): paper-ready charts, deep result parsing, and safe syntax execution","description":"# spss-mcp (macOS + Windows)\n\n> 让 SPSS 成为 Agent 的「统计引擎 + 制图工厂」：论文级图片、深度结果解析、方法真机验证、安全执行。\n\n**简体中文** ｜ [English](README.en.md)\n\n[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](LICENSE)\n[![PyPI](https://img.shields.io/pypi/v/spss-mcp.svg)](https://pypi.org/project/spss-mcp/)\n[![Python](https://img.shields.io/badge/Python-3.10%2B-blue.svg)]()\n[![Platform](https://img.shields.io/badge/platform-macOS%20%7C%20Windows-blue.svg)]()\n\n面向 **IBM SPSS Statistics** 的 MCP（Model Context Protocol）服务器，供 Cursor / Claude Code / Codex 等 Agent 调用本机 SPSS。\n\n本仓库基于上游 [`flupke91/spss-studio-mcp`](https://github.com/flupke91/spss-studio-mcp)（MIT），在其能力之上增加了 **macOS / SPSS 27+** 适配与若干兼容性修复，并在 **macOS + IBM SPSS Statistics 27** 真机验证通过。\n\n## 能力概览\n\n- **论文级出图**：11 类 `spss_chart_*`，PNG / TIFF / EMF（1950×1500 @300 dpi）\n- **深度结果解析**：OMS 文本 → Markdown + JSON + 统计摘要\n- **分析方法**：描述 / t / ANOVA / 回归 / 因子 / 信度 / 生存 / 混合模型 / 中介调节等\n- **安全执行**：危险命令拦截、路径白名单、`dry_run`、审计日志\n- **跨平台引擎**：Windows（`stats.exe`）与 macOS（`SPSS Statistics.app` + 捆绑 Python3 XD）\n\n## 本 fork 相对上游的主要改动\n\n| 项 | 说明 |\n|----|------|\n| macOS 探测 | 自动发现 `/Applications/IBM SPSS Statistics *` |\n| XD 启动 | 设置 `SPSS_HOME`、`DYLD_*`，修复捆绑 `libpython` 加载 |\n| 出图 | macOS HTML 内嵌 JPEG → 自动转 PNG，再缩放至 300 dpi |\n| 路径 | 相对路径解析为绝对路径，避免 SPSS Error #61 |\n| 解析 | 避免把 ANOVA 表中的 Error SS 误判为 `Error #` |\n| 样例数据 | 宽表 / 长表纵向数据对齐，便于 RM-ANOVA 与 MIXED |\n\n## 系统要求\n\n- Python 3.10+\n- 本机已授权的 **IBM SPSS Statistics**（已在 **27（macOS）** 验证；上游在 **32（Windows）** 验证）\n- macOS：安装包内需含 `Resources/Python3`（XD API）\n\n## 快速开始\n\n优先从 **PyPI** 安装（无需 clone）：\n\n```bash\npip install spss-mcp\n# 或：uv pip install spss-mcp\nspss-mcp status   # 应显示 SPSS batch / python: OK\n```\n\n免安装即跑（类似 `npx`）：\n\n```bash\nuvx spss-mcp status\nuvx spss-mcp serve\n```\n\nmacOS 若未自动发现安装，可设置：\n\n```bash\nexport SPSS_INSTALL_PATH=\"/Applications/IBM SPSS Statistics 27\"\n```\n\n配置 Cursor（`~/.cursor/mcp.json`）：\n\n```json\n{\n  \"mcpServers\": {\n    \"spss-mcp\": {\n      \"command\": \"uvx\",\n      \"args\": [\"spss-mcp\", \"serve\", \"--transport\", \"…","publisher":{"name":"spss-mcp Contributors","url":null},"homepage":"https://github.com/kinginsun/spss-mcp/tree/main/docs","repository":"https://github.com/kinginsun/spss-mcp/tree/main/docs","version":"1.1.2","license":"MIT","protocols":["mcp"],"tags":["chart","data-analysis","macos","mcp","psychology","social-science","spss","statistics"],"pricing":null,"endpoints":[{"url":"pypi:spss-mcp","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":"analytics.reporting","name":"Reporting & Dashboards","confidence":1,"provenance":"declared"}],"categories":["analytics"],"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":"spss-mcp","registry":"pypi","observedAt":"2026-10-30T20:23:29.451Z","publishedAt":"2026-10-30T12:04:13.407691Z","latestVersion":"1.1.2"},"toolSurface":null,"endpointFacts":[]},"verification":{"claimed":false,"claimedAt":null,"proofs":[]},"provenance":{"sources":[{"source":"pypi","key":"spss-mcp","url":"https://pypi.org/project/spss-mcp/","firstSeenAt":"2026-10-30T20:22:43.138Z","fetchedAt":"2026-10-30T20:22:43.138Z","normalizedAt":"2026-10-30T20:22:43.138Z"}]},"firstSeenAt":"2026-10-30T20:22:43.138Z","updatedAt":"2026-10-30T20:23:29.451Z"}