
Learning Commons Knowledge Graph
MCPLearning Commons 官方的 K-12 教育知识图谱远程 MCP,可按课标编号查到美国各州课标原文,拆解成可教的细分技能,并追溯数学课标前后衔接的学习进阶,帮老师备课、出题时对齐课标。需在 Learning Commons 平台创建 API Key;数据以美国课标为主,服务仍处于早期试用阶段。
Knowledge Graph
Learning Commons • Platform • Docs
What is Knowledge Graph?
Knowledge Graph is a structured collection of enriched educational datasets that connects academic standards, curricula, and learning science data. Datasets are modeled as graphs made up of entities (the elements in a dataset) and relationships (how those elements connect).
Knowledge Graph datasets use a standardized schema, so that edtech developers can build AI-powered tools without having to build a data pipeline themselves.
Check out common use cases and tutorials for Knowledge Graph.
Access Knowledge Graph
Knowledge Graph is database-neutral and interoperable. Start with the Quickstart, or pick an access method:
| Access method | When to use |
|---|---|
| Knowledge Graph Explorer | For visually exploring Knowledge Graph datasets in the browser |
| Agent plugins | For asking ChatGPT or Claude about Knowledge Graph data |
| Local files | For offline access and custom processing of Knowledge Graph data — download all data directly, or download by dataset using the Dataset catalog ↗ Can be used across graph databases, relational systems, in-memory tools, and AI pipelines without requiring any specialized infrastructure |
| REST API | For real-time programmatic access to Knowledge Graph data in an application |
| MCP server | For using Knowledge Graph data natively with an LLM |
Datasets
Knowledge Graph datasets are grouped into categories. Each category shares a schema. Each dataset covers a specific educational domain and includes one or more file downloads.
Download a dataset from its docs page below, or from the Dataset catalog. To download the full graph as JSONL (nodes.jsonl and relationships.jsonl), see Local files.
Access types: Open (freely downloadable), Open + Gated (some downloads open, others require approval), and Gated (approval required — contact support@learningcommons.org).
Standards
Official expectations for student learning across academic subjects and jurisdictions, plus durable skills and learner competencies. Schema: Standards.
| Dataset | Access | Downloads |
|---|---|---|
| U.S. academic standards | Open | Math, ELA, Social Studies, Science |
| U.S. academic standards crosswalks | Open | Math crosswalks, ELA crosswalks |
| English Language Development Standards Framework | Open | WIDA ELD Standards Framework |
| XQ Competencies | Open | XQ Competencies |
| Carnegie Skills Progressions | Open | Carnegie Skills Progressions |
Learning components
Granular skills or concepts that break academic standards into teachable parts. Schema: Learning components.
| Dataset | Access | Downloads |
|---|---|---|
| Math learning components | Open | Math learning components |
| ELA learning components | Open | ELA learning components |
| Eedi Misconceptions Graph | Open | Eedi Misconceptions Graph |
Learning progressions
How standards build towards and relate to each other. Schema: Learning progressions.
| Dataset | Access | Downloads |
|---|---|---|
| Math Coherence Map | Open | Math Coherence Map |
Curriculum
Standards-aligned lessons and assessments from a publisher. Schema: Curriculum.
| Dataset | Access | Downloads |
|---|---|---|
| IM® v.360 | Open + Gated | Scope and sequence (Open); Instructional materials (Gated); Assessment materials (Gated) |
| OpenSciEd | Gated | OpenSciEd curriculum (Gated) |
Instructional guidance
Grade- and subject-specific learner profiles, research-backed factors, and instructional strategies. Schema: Instructional guidance.
| Dataset | Access | Downloads |
|---|---|---|
| Learner Variability Navigator | Gated | Math models; Literacy models |
| English Learners Success Forum Guidelines | Open | Math guidelines; ELA guidelines; Science guidelines |
Additional resources
License
Use of Knowledge Graph is subject to our Terms of Use.
Knowledge Graph datasets are not all licensed the same way. Each dataset (and sometimes each download) is labeled Open, Open + Gated, or Gated:
- Open — Typically CC BY 4.0 (attribution required) or CC0
- Open + Gated — Includes both Open and Gated downloads (e.g., the IM® v.360 dataset's Scope and sequence is Open, while Instructional materials and Assessment materials are Gated)
- Gated — Not covered by an open license; requires approval and separate licensing agreement from the publisher
For the terms that apply to a specific dataset or download, reference that dataset’s page and the license docs.
Support
Questions or feedback: open an issue or email support@learningcommons.org. See also Support and release notes.
To report a security issue, contact security@learningcommons.org.
相關資源

Gemini Notebook CLI & MCP Server (notebooklm-mcp-cli)
MCP非官方 CLI 与 MCP,让 agent 操作 NotebookLM(现名 Gemini Notebook):批量导入资料、带出处问答,生成音频概览、测验、闪卡、思维导图和幻灯片,适合备课与整理文献。依赖未公开接口、可能失效,需登录 Google 账号并装好 uv。
jacob-bd/gemini-notebook-mcp-cli6.5k
Wikipedia MCP Server
MCP让 agent 检索维基百科,读取条目全文、摘要、指定章节和相关条目,适合备课查资料、辅导作业和事实核对。默认查英文维基,启动参数加 --language zh-hans 可切到简体中文等语言。无需账号,需本机安装 uv(Python)并能正常访问维基百科。
Rudra-ravi/wikipedia-mcp307
OpenZIM MCP Server
MCP离线检索维基百科、维基词典、Stack Exchange 等 ZIM 格式知识库,可全文搜索、按章节读取条目、查看链接关系,断网的教室也能用。需先从 Kiwix 下载 .zim 文件(几百 MB 到数十 GB)放进命令里的 ~/zim-files 文件夹,并在本机安装 uv。
cameronrye/openzim-mcp151
DeepTutor
CLI港大 HKUDS 开源的个性化辅导系统:导入教材建知识库后答疑、分步解题、出题测验,适合教师备课出题与学生自学;deeptutor 命令可输出 JSON 供 Claude Code 等调用。需要 Python 3.11+、Node.js 20+ 和大模型 API Key。
HKUDS/DeepTutor58.6k
notebooklm-py
CLI非官方 Python 库与 CLI,让 agent 操控 Google NotebookLM(现更名 Gemini Notebook):批量导入资料、基于来源带引用问答,生成播客、视频、幻灯片、测验与闪卡并批量下载,自带 agent skill。依赖未公开接口,需登录 Google 账号。
teng-lin/notebooklm-py20.2k
Gemini Notebook (NotebookLM) CLI & MCP Server
CLI非官方的 NotebookLM(现名 Gemini Notebook)命令行 nlm 与 MCP 服务器:老师和学生可让 AI 批量建笔记本、导入网页和文档,生成音频概述、测验、闪卡、思维导图和幻灯片,用于备课与复习。依赖未公开接口,需登录 Google 账号,可能随官方改版失效。
jacob-bd/gemini-notebook-mcp-cli6.5k