BC Curriculum MCP Server

BC Curriculum MCP Server

Enables natural-language querying of the British Columbia K-12 curriculum, including Big Ideas, Curricular Competencies, Content standards, grade progressions, and cross-subject connections, grounded in official BC Ministry of Education data.

Category
访问服务器

README

BC Curriculum MCP Server

An MCP (Model Context Protocol) server that gives BC teachers structured access to the entire BC Ministry of Education K–12 curriculum. Connect with one line of config — no installs, no accounts, no technical setup.

Author: Paul de Groot — Vancouver, BC

What It Does

This server crawls curriculum.gov.bc.ca and makes the full BC K–12 curriculum queryable through 12 MCP tools. Teachers and curriculum designers can ask natural-language questions about Big Ideas, Curricular Competencies, Content standards, grade progressions, cross-subject connections, and more — grounded in the actual published curriculum data.

Data Coverage

  • Grades K–12 (Kindergarten through Grade 12)
  • 9 English-stream subjects: ADST, Arts Education, Career Education, English Language Arts, Languages, Mathematics, Physical & Health Education, Science, Social Studies
  • Big Ideas, Curricular Competencies, and Content (the three-column KDU structure) with elaborations
  • Core Competencies (Communication, Thinking, Personal & Social)
  • First Peoples Principles of Learning (FPPL)
  • Assessment resources and Instructional samples
  • Full-text search across all curriculum content

Quick Start

Option 1: Remote server (no install needed)

Add this to your MCP config:

{
  "mcpServers": {
    "bc-curriculum": {
      "type": "http",
      "url": "https://bc-curriculum-mcp.fly.dev/mcp"
    }
  }
}

Or via Claude Code CLI:

claude mcp add bc-curriculum --transport http --url https://bc-curriculum-mcp.fly.dev/mcp

Option 2: Run locally via npx

npx bc-curriculum-mcp-server

This runs the server in stdio mode — compatible with any MCP client. You'll need a local bc-curriculum.sqlite database (clone the repo, install Playwright, and run the crawler — see For Developers below).

Option 3: Install globally via npm

npm install -g bc-curriculum-mcp-server
bc-curriculum-mcp

Once connected, ask things like:

  • "What are the Big Ideas for ADST grade 10?"
  • "Show me how science competencies build from grade 3 to 7"
  • "Find curriculum connections between math and ADST for grade 9"
  • "What are the First Peoples Principles of Learning?"
  • "What are the curricular competencies for Kindergarten math?"

Available Tools

Tool What it does
bc_search_curriculum Full-text search across all curriculum data
bc_get_course_curriculum Get Big Ideas, Competencies, and Content for a course
bc_list_courses List all available courses (filter by subject/grade)
bc_get_grade_progression Trace how curriculum builds across grade levels
bc_get_competency_connections Find competencies shared across subjects
bc_get_core_competencies Get Communication, Thinking, Personal/Social competencies
bc_get_fppl Get First Peoples Principles of Learning
bc_get_assessment_resources Get assessment practices and guidance
bc_get_crawl_status Check data freshness and completeness
bc_search_cross_curricular Find curriculum connections shared between two or more subjects at a grade
bc_get_curriculum_changes Show what changed in curriculum since a given date
bc_get_course_history View crawl history and change timeline for a course

Architecture

┌──────────────────────┐       ┌─────────────────────────┐
│  Your Machine        │       │  Fly.io (free tier)     │
│                      │       │                         │
│  Playwright crawler  │──────▶│  Express + SQLite       │
│  builds SQLite DB    │ SFTP  │  read-only MCP server   │
│                      │       │  256MB RAM              │
└──────────────────────┘       └─────────────────────────┘
                                        │
                                        ▼
                               Claude Desktop / Cowork / VS Codee / Antigravity
                               connects via MCP over HTTPS

The crawler runs locally using Playwright + headless Chromium to render the JavaScript-heavy curriculum.gov.bc.ca pages. It produces a SQLite database that gets uploaded to Fly.io. The deployed server is lightweight and read-only — no browser, no crawler, just serves pre-built data over MCP.

For Developers

Prerequisites

  • Node.js 18+
  • npm

Setup

git clone https://github.com/pdg6/bc-curriculum-mcp-server.git
cd bc-curriculum-mcp-server
npm install
npx playwright install chromium
npm run build

Crawl

# Quick test — just ADST
npm run crawl -- --subject adst

# Full K–12 crawl (~60-90 minutes)
npm run crawl:all

# Re-crawl everything (overwrites existing data)
npm run crawl:all -- --force

Run Locally

# stdio transport (for local MCP clients)
npm start

# HTTP transport (for remote connections)
TRANSPORT=http npm start

How It Works

The BC curriculum is organized in a three-column structure for every course:

  1. Big Ideas — High-level conceptual understandings
  2. Curricular Competencies — What students can do (grouped by domain)
  3. Content — What students should know (the KDU items)

Each element can have elaborations — expandable details that clarify scope and provide examples. This server captures all three columns plus elaborations for every course across all 9 subjects and all grades K–12.

The Core Competencies (Communication, Thinking, Personal & Social) and First Peoples Principles of Learning are cross-cutting frameworks that apply across all subjects and are stored separately.

Data Source

All curriculum data is sourced from curriculum.gov.bc.ca, the official BC Ministry of Education curriculum website. The crawler respects the site with 1.5-second delays between requests.

License

MIT — see LICENSE.

The curriculum content itself is published by the BC Ministry of Education. This project provides structured access to that publicly available data.

推荐服务器

Baidu Map

Baidu Map

百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。

官方
精选
JavaScript
Playwright MCP Server

Playwright MCP Server

一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。

官方
精选
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。

官方
精选
本地
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。

官方
精选
本地
TypeScript
VeyraX

VeyraX

一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。

官方
精选
本地
Kagi MCP Server

Kagi MCP Server

一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。

官方
精选
Python
graphlit-mcp-server

graphlit-mcp-server

模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。

官方
精选
TypeScript
Exa MCP Server

Exa MCP Server

模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。

官方
精选
mcp-server-qdrant

mcp-server-qdrant

这个仓库展示了如何为向量搜索引擎 Qdrant 创建一个 MCP (Managed Control Plane) 服务器的示例。

官方
精选
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选