EduCanvas — AI skill for Claude Code
EduCanvas - A Multimodal AI Tutor for K–12 AI Literacy Education.
How to install EduCanvas
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open Timcai06/EduCanvas and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What EduCanvas does
EduCanvas - A Multimodal AI Tutor for K–12 AI Literacy Education.
Alternatives in AI
- Vllm Mlx — High-performance OpenAI and Anthropic compatible LLM inference server for Apple Silicon 1.5k ★
- AI Papers 2025 — 📚 2025 AI Engineering Reading List - 120+ curated AI research papers covering LLMs, Multimodal, RAG, Agents 101 ★
- Kid Tutor — 小学生 AI 家教。输入作文题/英语单词/数学题/错题,获得引导式辅导 31 ★
README
EduCanvas
以教育能力为核心的通用个人 Agent 平台。
用同一个 Agent 连接资料、对话、创作、互动学习与可信评价。
一次对话如何执行
一次对话由入口提供身份、Profile、传输方式与能力信息,再进入共享的 `TurnApplication` 和 `AgentLoopEngine`。模型负责提出回答或工具调用;工具的执行、预算、取消和事实记录由代码维护。

图中是逻辑系统的轴测拆解:入口、运行时、适配器与事实账本的空间分层不代表请求执行顺序。
- 循环可控。 模型重试同样计入预算;取消后不继续触发工具副作用;工具结果经过校验再送回模型。
- 入口共享实现。 Gateway 与 Web 的组合层使用同一个
createTurnApplication工厂和ToolKernelPort契约,保留各入口的身份与能力差异。 - 教育接入有边界。 教育能力通过 Profile、Skills、Tools 与可信领域服务接入。学习状态、判分与掌握度由确定性服务端代码维护。
实现入口:[`turn-composition.ts`](apps/gateway/src/turn-composition.ts)、[`agent-loop.ts`](packages/agent-runtime/src/agent-loop.ts)。跨入口能力与教学约束详见[智能体编排边界](docs/03-ai/01-智能体编排边界.md)。
Why EduCanvas
EduCanvas 不把教育做成独立的聊天机器人或第二套 Agent。通用协作与教育场景共享同一条 Agent Runtime:教育能力以 Profile、Skills、Tools 与可信领域服务按需接入;学习状态、判分和掌握度只由确定性服务端代码更新。
这意味着资料、对话、Canvas 产物和学习过程可以在同一 Notebook 边界内协作,同时保持 Provider、权限、数据事实与高风险行动的明确边界。
架构

apps/gateway是 `gatewa
Related Skills
Ainstain2025 Star
AI agent pipelines, Claude/MCP integrations, LLM evaluation frameworks, prompting systems and AInstAIn platfor
Edu Image Prompt
Generate student-friendly educational image prompts from knowledge points for AI literacy lessons, PPT slides,
AI Literacy Superpowers
A set of Claude Code and GitHub Copilot plugins providing the AI Literacy framework's complete development wor
Multimodal Voice Assistant
This project is a multi-modal AI voice assistant that uses LM Studio, OpenAI API or Claude Code, audio process
Multimodal RAG
Fully local multimodal RAG knowledge base for AI agents. Ingest videos, images, audio, docs into ChromaDB usin
Multimodal Vision Skill
One image-recognition engine for any AI agent — 3 protocols (OpenAI Chat/Responses, Anthropic Messages), multi
Related Agents
Gemini Search Agent
Use for multimodal research leveraging Gemini search patterns.
OCR Reader
Batch OCR re-reader for low-confidence book pages. Reads page screenshot images via multimodal Read tool and e
Multimodal Engineer Md
Multi-modal AI specialist for AutoBot's advanced AI capabilities. Use for computer vision, voice processing, s