Co-Pilot / 辅助式
更新于 24 days ago

content

Ppauljbernard
0.0k
pauljbernard/content
84
Agent 评分

💡 摘要

一个分层知识系统,用于创建可重用的、符合标准的教育内容,涵盖各种学科和州。

🎯 适合人群

课程开发者教育内容作者学区管理员K-8教育教师教育技术集成者

🤖 AI 吐槽:这个系统就像教育工作者的瑞士军刀,但我希望他们不要在课堂混乱中丢失它。

安全分析中风险

自述文件未指明任何直接的安全风险,但系统的复杂性可能导致配置错误。确保适当的访问控制和定期审计以降低风险。

Content Repository

Multi-Curriculum Educational Content Development System

This repository provides:

  1. HMH Multi-Curriculum Knowledge Base - A hierarchical, reusable knowledge system for creating standards-aligned instructional materials across any state, any subject, any grade level
  2. Professor Framework Integration - AI-powered content development with 92 specialized skills and 22 autonomous agents

Table of Contents


HMH Multi-Curriculum Knowledge Base

Status: Week 3 Complete (50 files, 94% of Phase 1)

What It Is

A hierarchical knowledge resolution system that enables creating standards-aligned, state-compliant instructional materials with 85-97% knowledge reuse across curricula.

Key Innovation: Write once, reuse everywhere. Universal instructional frameworks, assessment guides, and accessibility standards are shared across all states and subjects. State-specific standards and compliance requirements are layered on top through a 5-level hierarchy.

Current Coverage

States: 3 of 51 (6%)

  • Texas (TEKS standards)
  • California (CCSS/NGSS standards)
  • Florida (MAFS/B.E.S.T./NGSSS standards)

Subjects: 3 of 10+ core subjects (30%)

  • Mathematics (K-8) - 8 Math Language Routines, TEKS/CCSS-M/MAFS alignment
  • ELA (K-8) - 4 Literacy Routines, TEKS ELA/CCSS ELA alignment
  • Science (K-8) - NGSS 3-dimensional learning, 8 Science & Engineering Practices

Grade Levels: K-8 (elementary and middle school)

Curricula Supported:

  • HMH Into Math Texas Edition
  • HMH Into Math California Edition
  • HMH Into Math Florida Edition
  • HMH Into Reading Texas Edition

Knowledge Reuse Demonstrated

  • Florida Math: 97% reuse (only 1 new file, 30 files reused)
  • California Math: 93% reuse (2 new files, 28 files reused)
  • Texas ELA: 90% reuse (3 new files, 27 files reused)

Comparison to Traditional Approach:

  • Traditional: ~500-750 unique files per curriculum (no reuse)
  • This System: 50 shared files support 4 curricula (98% file reduction)

Architecture

5-Level Hierarchical Resolution (specific → general, first match wins):

1. Program-Specific       (e.g., Into Math TX specific features)
   ↓ (if not found)
2. Subject-District       (e.g., TEKS Math alignment)
   ↓ (if not found)
3. Subject-Common         (e.g., Math MLRs, universal for all states)
   ↓ (if not found)
4. District-Wide          (e.g., Texas compliance, ELPS)
   ↓ (if not found)
5. Universal              (e.g., UDL, DOK, WCAG, assessment best practices)

Example Resolution for "How to scaffold vocabulary for emergent bilinguals in Texas Math":

  1. Checks /subjects/mathematics/districts/texas/into-math/ (program-specific)
  2. Checks /subjects/mathematics/districts/texas/ (TEKS math)
  3. Finds /subjects/mathematics/common/mlr/ (Math Language Routines - universal)
  4. Finds /districts/texas/language/elps-alignment.md (Texas language standards)
  5. Finds /universal/frameworks/eb-scaffolding-guide.md (universal EB support)

Result: Combines universal MLR strategies with Texas-specific ELPS requirements and universal EB scaffolding techniques.

Directory Structure

reference/hmh-knowledge-v2/
├── universal/                    # 15 files - applies to ALL curricula
│   ├── frameworks/               # UDL, DOK, EB scaffolding, sentence frames
│   ├── assessment/               # Item types, rubrics, answer keys (8 files)
│   ├── accessibility/            # WCAG 2.1 AA compliance
│   ├── content-equity/           # CEID bias-free content standards
│   └── vendor/                   # HMH-specific quality checklists
├── subjects/
│   ├── mathematics/
│   │   ├── common/               # 12 files - ALL states use these
│   │   │   ├── mlr/              # 8 Math Language Routines (MLR1-MLR8)
│   │   │   ├── vocab-guidelines.md
│   │   │   └── problem-solving-framework.md
│   │   └── districts/
│   │       ├── texas/            # TEKS Math alignment
│   │       ├── california/       # CCSS-M alignment
│   │       └── florida/          # MAFS alignment
│   ├── ela/
│   │   ├── common/               # 5 files - ALL states use these
│   │   │   └── literacy-routines/  # Close Reading, Think-Pair-Share, etc.
│   │   └── districts/
│   │       └── texas/            # TEKS ELA alignment
│   └── science/
│       └── common/               # 2 files - ALL NGSS states use these
│           ├── ngss-alignment-guide.md
│           └── science-practices-framework.md
├── districts/
│   ├── texas/                    # 4 files - applies to ALL subjects in TX
│   │   ├── compliance/           # SBOE, IPACC requirements
│   │   └── language/             # ELPS (Texas language standards)
│   ├── california/               # 2 files - applies to ALL subjects in CA
│   │   ├── compliance/           # CA adoption criteria
│   │   └── language/             # ELD Standards
│   └── florida/                  # 3 files - applies to ALL subjects in FL
│       ├── compliance/           # FL statutory compliance
│       └── language/             # ESOL/WIDA Standards
└── publishers/
    └── hmh/                      # HMH content schema

config/curriculum/
├── hmh-math-tx.json              # Defines resolution order for TX Math
├── hmh-math-ca.json              # Defines resolution order for CA Math
├── hmh-math-fl.json              # Defines resolution order for FL Math
└── hmh-ela-tx.json               # Defines resolution order for TX ELA

Key Files by Category

Universal (15 files) - Reused by ALL Curricula:

  • UDL Principles Guide
  • DOK (Depth of Knowledge) Framework
  • EB Scaffolding Guide (Emergent Bilinguals)
  • Sentence Frames Library
  • WCAG 2.1 AA Compliance Guide
  • 8 Assessment Files (item types, rubrics, answer keys, validation, etc.)
  • CEID Guidelines (Content Equity, Inclusion & Diversity)
  • Vendor Quality Checklist

Mathematics Common (12 files) - Reused by ALL Math Programs:

  • 8 Math Language Routines (MLR1-MLR8):
    • MLR1: Stronger and Clearer Each Time
    • MLR2: Collect and Display
    • MLR3: Critique, Correct, and Clarify
    • MLR4: Information Gap
    • MLR5: Co-Craft Questions
    • MLR6: Three Reads
    • MLR7: Compare and Connect
    • MLR8: Discussion Supports
  • MLR Overview & Placement Rules
  • Math Vocabulary Guidelines
  • Problem-Solving Framework

ELA Common (5 files) - Reused by ALL ELA Programs:

  • Close Reading Protocol
  • Think-Pair-Share
  • Annotation Protocol
  • Turn-and-Talk
  • (More literacy routines to be added)

Science Common (2 files) - Reused by ALL NGSS States:

  • NGSS Alignment Guide (3-dimensional learning)
  • Science Practices Framework (8 SEPs with CER framework)

State-District Files (9 files) - Apply to ALL subjects in that state:

  • Texas: IPACC, SBOE, ELPS, compliance checklist
  • California: Adoption criteria, ELD Standards
  • Florida: Adoption criteria, ESOL/WIDA, statutory compliance

Subject-District Files (7 files) - State + Subject specific:

  • TEKS Math alignment (TX)
  • TEKS ELA alignment (TX)
  • CCSS-M alignment (CA)
  • MAFS alignment (FL)
  • Gap mitigation strategies (TX Math)
  • (ELA CA/FL and Science files - documented gaps)

Documentation: Complete Guide Suite

The system provides 4 specialized guides for different roles:

📝 AUTHOR_GUIDE.md - For Content Authors

  • Complete authoring workflows (lessons, assessments, activities)
  • Using the knowledge base to generate content
  • AI assistance with Professor Framework
  • Quality standards and checklists
  • 850 lines of comprehensive guidance

EDITOR_GUIDE.md - For Content Editors

  • Editorial workflow from assignment to approval
  • 8-section comprehensive review checklist
  • Effective feedback templates
  • Common issues and fixes
  • 450 lines of editorial guidance

📦 PRODUCTION_GUIDE.md - For Publishers/Production

  • Multi-format production (PDF, HTML, SCORM, accessible)
  • Asset management and optimization
  • QA checklists and delivery workflows
  • 450 lines of production guidance

⚙️ ENGINEER_GUIDE.md - For Knowledge Base Engineers

  • System architecture and hierarchical resolution
  • Adding new states/districts (2-6 files, 85-97% reuse)
  • Adding new subjects (8-15 files)
  • Creating curriculum configs
  • File creation best practices and templates
  • 1700 lines of engineering documentation

📋 USER_GUIDE.md - Overview and Navigation

  • Introduction to the 4 guides
  • Quick navigation by role
  • Complete content lifecycle
  • Documentation roadmap

For Engineers: How to Use This System

See: ENGINEER_GUIDE.md - Complete guide for knowledge base engineers

Quick Start:

  1. Choose or create a curriculum config (see config/curriculum/)
  2. Config defines resolution order (program → subject-district → subject-common → district → universal)
  3. System automatically finds and combines relevant knowledge files
  4. Generate content using combined knowledge with 85-97% reuse

Adding New States:

  • CCSS/NGSS state: 2-3 files (compliance, language, notes) → 90-95% reuse
  • State-specific standards: 5-6 files (standards per subject + compliance) → 85-90% reuse

Adding New Subjects:

  • 8-10 subject-common files (instructional routines, standards alignment)
  • 1-2 subject-district files per state
  • Reuse all 15 universal files and all district-wide files

Future Expansion Roadmap

See: .archive/HMH_Scaling_Roadmap_National_All_Subjects.md

Phase 2-7: Complete national coverage

  • Add 48 more states/districts (total: 51)
  • Add 7+ core subjects (Social
五维分析
清晰度8/10
创新性9/10
实用性9/10
完整性8/10
可维护性8/10
优缺点分析

优点

  • 跨课程的高知识重用
  • 支持多个州和学科
  • AI驱动的内容开发
  • 为用户提供全面的文档

缺点

  • 当前州覆盖有限
  • 层次结构复杂
  • 需要大量初始设置
  • 新用户可能面临学习曲线

相关技能

pytorch

S
toolCode Lib / 代码库
92/ 100

“它是深度学习的瑞士军刀,但祝你好运能从47种安装方法里找到那个不会搞崩你系统的那一个。”

agno

S
toolCode Lib / 代码库
90/ 100

“它承诺成为智能体领域的Kubernetes,但得看开发者有没有耐心学习又一个编排层。”

nuxt-skills

S
toolCo-Pilot / 辅助式
90/ 100

“这本质上是一份组织良好的小抄,能把你的 AI 助手变成一只 Nuxt 框架的复读机。”

免责声明:本内容来源于 GitHub 开源项目,仅供展示和评分分析使用。

版权归原作者所有 pauljbernard.