GDMU MCP Server
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@GDMU MCP Server查询2024-2025学年第一学期的成绩"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
🎓 ADE — Academic Decision Engine
确定性 · 可解释 · 可重放的学业决策引擎 从「我觉得能毕业」到「证据链证明你能毕业」
ADE 是一套面向高校教务的确定性学业决策引擎:把培养方案编译成知识投影,把学生成绩单解析成课程匹配结果,用纯规则引擎做出毕业判定,并把每次判定冻结成不可变的审计快照——五层证据链、可重放、可解释。
仓库同时保留 正方 V-9.0 教务系统 MCP 桥,让 AI 客户端直接查询课表、成绩、考试与选课。
✨ 亮点
能力 | 说明 | |
🎯 | 确定性 | 同输入必同输出;Replay Hash 与审计快照字节级可复现(ADR-0026) |
🔍 | 可解释 | 五层证据链: |
🔁 | 可重放 |
|
🧩 | Academic Matcher | 课程本体(别名/等价/替代/依赖)驱动的课程解析,Rule Accuracy 95% |
🗺️ | 多专业 | 护理学 / 临床医学 / 医学检验 / 医学影像 / 口腔医学;50 CANONICAL、500 合成成绩单 |
🤖 | Copilot | 自然语言 → 结构化意图 → 带证据的回答(计算路径无 LLM) |
🧭 | Planner + What-if | 补修计划生成与「删一门课会怎样」模拟 |
🚀 | Release 工程 |
|
Related MCP server: Canvas MCP Server
🏗 架构总览
flowchart TD
A[CLI / REST / MCP / SDK / Playground] --> B[AcademicCopilotFacade]
B --> C[Copilot: 意图识别]
B --> D[Planner + What-if]
C --> E[DecisionContext + RuleEngine]
D --> E
E --> F[AuditSnapshot + 五层证据链]
F --> G[Repository 层]
G --> H[Knowledge / Rule / Ontology]
H --> I[Compiler: PDF → AST → IR → plan JSON]完整架构见 Architecture Handbook 与 6 张架构图。
🚀 5 分钟快速开始
cd academic-copilot-framework/packages/ade
# 1. 构建知识库(5 个专业)
python -m compiler.cli knowledge build --verify
# 2. 健康检查(版本 / Hash / 知识库 / 规则 / Ontology / Benchmark / Replay)
python -m compiler.cli doctor然后跑真实示例:
cd academic-copilot-framework/examples
python 01_basic_status.py # Student → Audit → Response
python 02_risk.py # 毕业风险评估
python 03_plan.py # 补修计划
python 04_what_if.py # 补修课程 → 重新判定浏览器打开:
Playground — 交互式成绩单 / 判定 / 证据
Benchmark Dashboard — 指标可视化
Demo 学生 Alice / Bob / Charlie 已在
demo/students/,无需导数据。
📊 Benchmark
指标 | 数值 |
Rule Accuracy(护理学,10 CANONICAL) | 95% |
Overall Accuracy | 80% |
Matcher Accuracy | 100% |
Alias / Equivalent Match Rate | 100% |
Unmatched Courses | 0 |
专业 / CANONICAL / 合成成绩单 | 5 / 50 / 500 |
性能(10,000 学生) | 判定 4.6s · 审计 5.9s · 峰值内存 2.6MB |
详细数据:multi_major_benchmark.md · performance.md · Release Notes
📚 文档
文档 | 位置 |
架构手册(10 篇) | |
API 参考(OpenAPI + 6 模块) | |
ADR 索引(31 条架构决策) | |
快速开始 / FAQ / 贡献指南 | |
学术论文(Markdown + PDF) | |
安全策略 / 行为准则 / 引用格式 | SECURITY · CODE_OF_CONDUCT · CITATION.cff |
🗂 仓库结构
.
├── academic-copilot-framework/ # ADE 学业决策引擎
│ ├── packages/ade/ # 核心:compiler/knowledge/rule/decision/
│ │ # audit/matcher/copilot/planner/simulation
│ ├── validation/ # 验证工作区:golden cases + benchmark
│ ├── docs/ # 架构 / API / ADR / 图
│ ├── examples/ # 8 个可运行示例
│ ├── demo/ # 开箱即用的演示学生
│ ├── playground/ # 交互式 Playground
│ ├── benchmark/ # Benchmark 可视化
│ ├── plugins/ # 插件 SDK(Notification/Rule/Planner/Matcher)
│ ├── installer/ # Windows 安装器脚手架
│ └── paper/ # 学术论文
├── gdmu_mcp/ server.py # 正方教务 MCP 桥
└── docs/ # 逆向与教务系统分析文档🧪 测试
cd academic-copilot-framework/packages/ade && python -m pytest tests -q # 426 tests
cd academic-copilot-framework/validation && python -m pytest tests -q # 241 testsCI 流水线(ade-release.yml):push → Compile → Build → Validation → Ruff → Benchmark → Doctor → Docker → Release
🔌 MCP 桥(正方 V-9.0 教务系统)
面向 Codex / Claude Desktop / Cursor 等 AI 客户端的 MCP 桥,覆盖课表、成绩、考试、选课、毕业规划全场景(19 个工具,专为 CAS 单点登录 + MFA 设计)。配置与工具清单见仓库根 config.example.env 与 docs。
🤝 贡献
欢迎贡献!请先阅读 CONTRIBUTING 与 ADR 索引,保持核心管线确定性与 Repository 边界。
免责声明
本工具面向个人学业数据分析与研究场景设计。
仅限本人账号使用:禁止用于分析他人数据、批量采集或任何未经授权的访问。
仅限学习与研究用途:使用者须遵守所在高校的学生管理规定与相关法律法规。
不构成毕业建议:分析结果仅供辅助参考,正式毕业判定以学校教务处出具的文件为准。
作者不承担他人滥用责任:使用者对自身行为负全部责任。
使用本工具即表示你已阅读、理解并同意上述条款。
📄 许可
引用本项目请使用 CITATION.cff。
This server cannot be installed
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Related MCP Servers
- -licenseCquality-maintenanceEnables AI assistants to interact with Hangzhou Dianzi University's academic system through automatic login and course schedule retrieval. Supports secure authentication and structured academic data access for HDU students.Last updated2
- Alicense-qualityDmaintenanceEnables AI assistants to interact with Canvas LMS through 25 comprehensive tools. Supports course management, assignments, grades, messaging, calendar events, and file access through natural language.Last updated15512MIT
- Flicense-qualityDmaintenanceEnables AI assistants to access educational domain tools including grading, cognitive diagnosis, knowledge tracing, learning path recommendations, and sentiment analysis through FastAPI-powered small models.Last updated2

MCPaeroeduofficial
AlicenseAqualityBmaintenanceEnables AI assistants to interact with the AeroEduVision platform, allowing exam generation, course management, knowledge graph queries, and student progress tracking through natural language commands.Last updated24MIT
Related MCP Connectors
TRON energy exchange for AI agents. 54 tools, 30 prompts, 21 resources. A2A + ACP.
Chinese metaphysics (bazi, qimen, 5-element) as decision-support tools for AI agents.
Connect AI assistants to Stellary projects, boards, documents, and governed agent workflows.
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/S4saK1/GDMU-Academic-Runtime'
If you have feedback or need assistance with the MCP directory API, please join our Discord server