magic
Magicrew. The first open-source all-in-one AI productivity platform (Generalist AI Agent + Workflow Engine + IM + Online collaborative office system)
- 5.1k stars
- 565 forks
- 23 open issues
- Language: TypeScript
- License: NOASSERTION
- updated
Install
git clone https://github.com/dtyq/magic && cp magic/*.md ~/.claude/agents/- Clone the repository and copy the agent .md definitions into ~/.claude/agents (or .claude/agents inside a project).
- Start a new Claude Code session to load the agents.
- Delegate work to them with the Task/Agent tool or by name.
Editor's note
Magic is an open-source enterprise AI agent platform built in TypeScript that combines a multi-agent workflow engine, an instant messaging layer, and an online collaborative office system into a single self-hosted deployment. It connects to Claude and other large language models through a configurable API layer and supports MCP for tool integration. The platform lets teams build autonomous agents that wrap internal systems such as ERP and CRM databases as reusable "digital employees," enforce per-department and per-user token budget caps, and route high-risk actions through a human approval gate before execution. A built-in rendering framework converts agent output directly into finished artifacts including PowerPoint files, Excel spreadsheets, data dashboards, and infinite canvases, bypassing the manual formatting step typical of chat-only tools. Every agent runs inside a sandboxed container isolated in a separate VPC with per-tenant resource separation. The platform targets both solo founders managing multiple business functions with minimal overhead and larger organizations that need centralized governance over AI usage and data residency.
24 items in this repository
Query the list of all agents (employees) available to the current user. Use when generating code that requires a real agentId, or when the user asks "which agents/employees do I have".
Migrate existing antd-style CSS-in-JS (createStyles) components to Tailwind CSS utility classes with full dark mode support, following the project's style system transition strategy. Use when the user asks to migrate, convert, or rewrite antd-style components to Tailwind, or when working on new components in files that already use antd-style.
Open and operate webpages in a real browser: navigate, read rendered content, inspect interactive elements, click, type, submit forms, press keys, select options, check boxes, hover, scroll, upload files, wait for page changes, manage pages, take and save screenshots, analyze visual layout, and inspect page errors and network activity.
Core canvas design skill covering project management, multimedia principles, AI image generation, web image search, and design marker processing. Load for any canvas design task. CRITICAL - When user message contains [@design_canvas_project:...] or [@design_marker:...] mentions, or when the user wants to generate video/animation/clip on a canvas project, you MUST load this skill first before any operations.
Use when details may have been lost after context compaction, or when the Agent needs to inspect earlier compacted messages or previously created sub-agents.
Manage persistence for user-installed third-party CLI tools in the Super Magic sandbox. Use when the user wants to install, configure, adopt, restore, inspect, or remove a CLI, or when an external skill, webpage, or installer asks to run CLI install commands such as curl|sh, npm install -g, pipx install, uv tool install, go install, cargo install, or brew install. Do not persist runtime-provided CLIs.
Expert code review of current git changes with a senior engineer lens. Detects SOLID violations, security risks, and proposes actionable improvements.
Summarize and compress the current conversation history into a structured context snapshot, then call compact_chat_history to save it. Read this skill only when the user explicitly asks to compact/summarize — system-triggered compaction injects the instructions directly without requiring a skill read.
Slide/PPT creation skill that provides complete slide creation, editing, and management capabilities. Use when users need to create slides, make presentations, edit slide content, or manage slide projects. CRITICAL - When user message contains [@slide_project:...] mention, you MUST load this skill first before any operations.
Data analysis dashboard (instrument panel) development skill. Use when users need to develop data dashboards, create/edit Dashboard projects, build large-screen data boards, or perform dashboard data cleaning. Includes dashboard project creation, card plan, data cleaning (data_cleaning.py), card management tools (create_dashboard_cards, update_dashboard_cards, delete_dashboard_cards, query_dashboard_cards), map download tool (download_dashboard_maps), dashboard development, and validation.
Use when the user wants to interact with DingTalk in any way — including but not limited to: reading, querying, searching, sending, replying to, forwarding, or recalling DingTalk chat messages and chat history; managing group chats and conversations; sending DING alerts; querying contacts, org structure, AI search, or coworkers; reading, searching, creating, or editing DingTalk docs, drive files, sheets, AI tables, wiki, mail, calendar events, meeting rooms, AI meeting minutes, attendance, OA approvals, todos, reports/logs, live sessions, AI apps, permissions, or open-platform docs.
Use for reading, analyzing, summarizing, and converting complex documents into Markdown artifacts or other supported document formats.
Use when external HTTP or HTTPS files must be downloaded reliably into the workspace, especially for multiple files, large videos or archives, resumable transfers, custom Referer or Origin headers, or downloads that may take a long time.
Migrate modules between shared `src/opensource/` code and enterprise overlays under `enterprise/src/opensource/`, while preserving the `src/opensource` boundary and existing component override points.
Use when the user provides API keys, tokens, or other configuration values that should persist across sessions, or when the user asks to query, list, or delete saved environment variables. Manages personal env by default and workspace env only when explicitly requested.
Complete API reference for window.Magic.* in SuperMagic HTML micro-apps (HTML 微应用). Read this skill when you need exact method signatures, parameters, return types, or usage examples for: fs (readFile/writeFile/listFiles/listDir/getFileUrl/deleteFile/deleteDir/moveFile/renameFile/watchFile/watchDir), llm (chat/stream/getModels), agent (getAgents/selectAgent), project (createTopicAndSend/sendMessage/uploadFiles/downloadFiles), user (getInfo with app.json userInfo scopes), getAppBasePath, setInputMessage, reload. Also covers file-per-record data storage, list projection file names, tiptap JSON message format, @file and @skill mention structures, model selector UI rules, user info authorization, error handling patterns, and backward compatibility table. Trigger phrases: 'window.Magic API', 'readFile writeFile', 'listDir watchDir', 'getFileUrl', 'get file url', '文件 URL', '获取文件链接', 'deleteFile deleteDir', 'moveFile renameFile', 'watchFile callback', 'watchDir callback', 'llm.stream', 'llm.chat', 'createTopicAndSend format', 'tiptap JSON mention', '@file mention structure', '@skill mention', 'getAppBasePath usage', 'model selector UI', 'user.getInfo', 'get user info', 'user avatar', 'userInfo scopes', 'app.json permissions', 'Magic API 用法', 'fs 读写文件 API', 'fs 获取文件链接', 'fs 删除文件', 'fs 移动重命名', '目录监听', '文件监听回调', '话题消息格式', 'mention 结构', '模型选择器', '用户信息', '用户授权', '获取头像'.
Use when connecting or configuring IM channel bots (WeChat, WeCom, DingTalk, Lark), or when you receive a user message that contains an <im> block indicating it was sent from an IM channel — read this skill to understand what the context means and how to handle it correctly.
Use when the user wants to interact with Lark/Feishu in any way — including but not limited to: reading, querying, searching, sending, replying to, forwarding, or downloading Lark/Feishu IM messages and chat history; managing group chats; listing, viewing, searching, creating, or editing cloud docs/files, Drive, Markdown, spreadsheets, Base tables, wiki, whiteboards, slides, apps, calendars/events/meeting rooms, contacts/org structure, tasks/todos, approvals, attendance, mail, minutes, VC notes, OKRs, real-time events, or custom Lark CLI skills. Load this skill first to find the right sub-skill.
Create and manage calendar projects for scheduling, content planning, and event management. Use when user wants to create a calendar, schedule events, or build a content/publishing calendar.
Subagents overview
What people ask about magic
What is dtyq/magic?
dtyq/magic is subagents for the Claude AI ecosystem. Magicrew. The first open-source all-in-one AI productivity platform (Generalist AI Agent + Workflow Engine + IM + Online collaborative office system) It has 5.1k GitHub stars and its last recorded update is dated 2026-08-12.
How do I install magic?
You can install magic by cloning the repository (https://github.com/dtyq/magic) or following the README instructions on GitHub. ClaudeWave also provides quick install blocks on this page.
Is dtyq/magic safe to use?
Our security agent has analyzed dtyq/magic and assigned a Trust Score of 82/100 (tier: Trusted). See the full breakdown of passed checks and flags on this page.
Who maintains dtyq/magic?
dtyq/magic is maintained by dtyq. The last recorded GitHub activity is dated 2026-08-12, with 23 open issues.
Are there alternatives to magic?
Yes. On ClaudeWave you can browse similar subagents at /categories/agents, sorted by popularity or recent activity.