登录

refly

refly-ai活跃维护

开源智能体技能构建器，用可视化工作流定义技能，可运行在 Claude Code、Cursor、Codex 等平台，并生成 API 或 Slack、飞书机器人。

The first open-source agent skills builder. Define skills by vibe workflow, run on Claude Code, Cursor, Codex & more. Build Clawdbot 🦞· APIs for Lovable · Bots for Slack & Lark/Feishu · Skills are infrastructure, not prompts.

v1.1.0发布于 2026-02-02 08:24 UTC7,551star +3 / 7 天726fork同步于 2026-11-10 11:32 UTC
智能体技能工作流编排技能构建器可视化编排机器人集成
对比 GitHub 官网agentagent-skillsautomationclaudeclawdbot

怎么上手

由 README 提炼，执行前请确认
原文 
docker 自部署，详见 https://docs.refly.ai/community-version/self-deploy/
硬件要求按 README 说明

README 未说明硬件要求，仅提到可用 Docker 自部署，默认访问 http://localhost:5700

下载：源码 zip·GitHub Releases（安装包）本站不镜像安装包

项目说明

由 README 自动翻译提炼
没人工点评？写下你的经验

开源 Agent 技能构建平台，把业务流程编译成可版本化、可导出的确定性技能

开源智能体技能构建器，用可视化工作流定义技能，可运行在 Claude Code、Cursor、Codex 等平台，并生成 API 或 Slack、飞书机器人。

核心能力

  • 用自然语言描述意图，由 Model-Native DSL 编译成可复用技能
  • 提供可干预运行时，支持中途暂停、审计和重新引导执行逻辑
  • 技能可导出为 API、Webhook 或 Claude Code、Cursor 等工具
  • 内置集中式技能注册表，支持版本管理与团队协作
  • 兼容 MCP 服务器，并接入 3000+ 原生工具集成

容易踩的坑

自部署后需在设置中自行配置模型提供商（如 OpenAI、Anthropic）并设为默认才能使用；Claude Code 与 Cursor 的原生导出标注为 coming soon，尚未可用；README 未给出具体 docker 命令，需按官方自部署文档操作

内容由模型根据 README 生成，可能不完整或不准确，以官方文档为准。看原文 README

最近发版

5 个版本v1.1.0
全部 release
  1. v1.1.0最新正式版2026-02-02 · 2026-02-02 08:24 UTC

    🚀 Refly 1.1.0 Release Notes

    A powerful update focused on developer productivity, workflow transparency, and advanced governance for agentic infrastructure.

    Refly is evolving from a creative engine into the first open-source platform for building stable, atomic, and versioned agent skills. With v1.1.0, we're bringing the power of natural language to your terminal and giving you unprecedented control over how your workflows execute and scale.

    ✨ What's New

    💻 CLI Evolution: Skills at Your Fingertips

    Manage your entire agent infrastructure from the command line using natural language. Create, run, and share skills without leaving your terminal.

    🔍 Enhanced Workflow Visibility

    Stop guessing what's happening inside your agents. With real-time execution logs and instant resource previews, your workflow's "black box" is now a glass box.

    ⚙️ Advanced Governance & Operations

    From custom credit expiration to multi-model routing for scheduled tasks, Refly now provides the enterprise-grade controls needed for production-ready agent skills.

    🚀 New Features

    💻 CLI Tools: The Developer's Powerhouse

    The Refly CLI has been completely reimagined to be your primary interface for skill development:

    • Natural Language Skill Management: Create, run, export, and share Skills using simple natural language commands.
    • File Upload Support: Seamlessly integrate local files into your remote workflows directly from the CLI.
    • Complex Variable Handling: Support for passing structured data and complex parameters into your workflows with 100% reliability.

    🔍 Workflow & Execution Control

    We've added several features to make building and debugging workflows faster and more intuitive:

    • Instant Resource Preview: Preview text documents and code files directly within the workflow output panel.
    • Detailed Running Logs: Track every step of your task execution with real-time logs to pinpoint exactly where things happen.
    • Granular Node Control: Run individual nodes independently for rapid testing and debugging without re-running the entire workflow.
    • Workflow Management: Easily rename workflows and query specific node details to keep your workspace organized.

    🌐 Ecosystem & Operations

    Refly continues to expand its reach and reliability:

    • ByteDance Image Generation: New native tool support for high-quality image generation.
    • Flexible Scheduling: Scheduled tasks now support hourly intervals for more frequent automations.
    • Multi-Model Routing: Implement random or weighted routing across multiple models for scheduled tasks to optimize for cost or performance.
    • Optimized Tool Authorization: A streamlined flow for checking and authorizing tools, making integration smoother than ever.

    🛠 Performance & Bug Fixes

    Fixed Issues

    • ✅ Fixed progress display inaccuracies (now correctly showing 100%) and expanded query limits to ensure transparent and reliable task tracking.
    • ✅ Resolved file conversion failures across documents and emails, and fixed intermittent email delivery bugs.
    • ✅ Refined the agent's citation mechanism to provide more accurate and stable result references.

    🌟 User Experience Improvements

    • More powerful and intuitive CLI-driven development flow.
    • Greater transparency into workflow execution and artifact generation.
    • More flexible and governed scheduling for automated tasks.
    • Enhanced performance for large-scale file operations.

    🧭 Get Started

    Refly 1.1.0 bridges the gap between raw APIs and intelligent agents, turning "vibe-coded" scripts into durable, versioned infrastructure.

    We're building the future of agentic work, one ski

  2. v0.10.02025-08-27 · 2025-08-27 07:45 UTC
    查看更新说明

    🚀 Refly 0.10.0 Release Notes

    A milestone update focused on seamless creativity, intelligent workflows, and uninterrupted creation.

    Ever had that moment when your perfect AI-generated image vanished into the canvas abyss? Or spent ages hunting for that reference material you know you uploaded somewhere? Or watched your Agent get stuck mid-workflow despite your detailed instructions?

    We get it. True creation should be fluid, connected, and never interrupted. Today, we're solving all of that with Refly v0.10.0.

    ✨ What's New

    💠 Resource Hub: Your Creative Command Center

    Think of it as a super-powered asset manager for your creative workspace.

    🤖 Next-Gen Agent: Smarter, Smoother, More Reliable

    Intelligent task breakdown, real-time information retrieval, and seamless execution.

    🧠 Deepseek-V3.1: Free Unlimited Access (Limited Time!)

    Hybrid reasoning architecture with enhanced efficiency and powerful Agent capabilities.

    🚀 New Features

    💠 Resource Hub: Your Creative Command Center

    The new Resource Hub is like having a brilliant assistant managing your entire creative workspace:

    • Unified Asset Management: All your generated images, videos, documents, and uploaded references in one organized panel
    • Smart Categorization: Instantly find what you need with intelligent type-based organization
    • Step-by-Step Tracking: Complete workflow history for easy backtracking and optimization
    • AI Context Integration: Soon, these resources will feed directly into your Agent's context for seamless cross-modal creation

    🤖 Next-Gen Agent: Three Revolutionary Upgrades

    🧠 Intelligent Task Breakdown
    • Automatically splits complex requests into executable steps
    • Dynamically adjusts execution plans based on review feedback
    🔍 Real-Time Information Retrieval
    • Proactively searches for the latest information when uncertain
    • Stay current with real-time data instead of outdated references
    ⚡ Uninterrupted Task Execution
    • Seamlessly handles long, multi-step workflows
    • Maintains context throughout the entire process
    • Never loses track from concept to final output

    🧠 Deepseek-V3.1 Model Upgrade

    Deepseek-V3.1 is now live on Refly with two weeks of free unlimited access!

    Key Features
    • Hybrid Reasoning Architecture: One model, two modes - detailed explanations or concise answers
    • Enhanced Efficiency: 20-50% less output while maintaining the same performance
    • Superior Agent Capabilities: Significantly improved tool usage and intelligent task handling
    • Extended Context: Text processing capacity increased from 64k to 128k tokens
    • Optimized Multilingual Support: Enhanced processing for over 100 languages

    🛠 Performance & Bug Fixes

    Fixed Issues

    • ✅ Resolved asset loss during creative workflows
    • ✅ Fixed Agent task execution interruptions
    • ✅ Improved resource discovery and organization
    • ✅ Enhanced long-chain task completion reliability

    🌟 User Experience Improvements

    • Smoother, more connected creative workflows
    • Intuitive and accessible resource management
    • More intelligent and efficient Agent interactions
    • Significantly enhanced model performance

    🧭 Get Started

    Refly 0.10.0 isn't just another version update — it's a complete reimagining of the creative experience, making creation more fluid, intelligent, and efficient than ever before.

    We can't wait for you to experience this new world of creation.

    As always, we're building Refly for creators like you.

  3. v0.9.12025-08-14 · 2025-08-14 05:04 UTC
    查看更新说明

    What's Changed

    New Contributors

    Full Changelog: https://github.com/refly-ai/refly/compare/v0.9.0...v0.9.1

  4. v0.9.02025-08-01 · 2025-08-01 05:15 UTC
    查看更新说明

    🚀 Refly 0.9.0 Release Note

    A milestone update focused on creative freedom, clarity, and control.

    We're excited to officially release Refly 0.9.0 — a major upgrade that redefines the way you interact with AI models.

    🌈 Highlights

    • 🧮 Credit-Based Usage: Model usage is now powered by a flexible credit system — no need to configure API keys.
    • 🎨 Redesigned Interface: From homepage to canvas, every touchpoint is clearer, faster, and more focused.
    • 🎥 Multimodal Access: Use text, image, audio, and video generation seamlessly — all in one workflow.

    🌟 New Features

    🧮 Credit System (Cloud version)

    Refly now unifies model usage under a credit-based billing system. All models consume credits transparently:

    • All model types (text/image/audio/video) are billed by credit
    • No more manual API key setup
    • Legacy users automatically receive monthly credits
    • Early unlimited members retain free access to major models (Kimi, GPT-4.1, Gemini Pro)

    🎛 New Interface: Familiar Layout, Reimagined Experience

    • Homepage now features centralized task entry, model selection, and community templates
    • Canvas nodes are smoother, smarter, and more responsive
    • Sidebars, preview panels, and settings now share a unified visual language
    • Dark/light theme toggle supported for long-session comfort

    ⚙️ Dual Model Configuration Modes

    • Global Mode: Automatically loads recommended models — zero configuration required
    • Custom Mode: Allows full control over model groups and custom providers, preserving past setups

    🛠 Performance & Bug Fixes

    Fixed Issues

    • ✅ Fixed: Task timeout misjudgment for long-running model calls
    • ✅ Fixed: Code generation node stuck in "processing" after completion
    • ✅ Fixed: Inactive tasks not correctly marked as failed
    • ✅ Fixed: Default media models failing to load in Agent mode
    • ✅ Fixed: Avatar setup not initializing correctly
    • ✅ Fixed: Duplicate provider install prompts & occasional install failures

    🌟 UX Improvements

    • Code blocks in AI responses now support folding/unfolding
    • Improved document typography: better spacing, font clarity, and layout width for reading

    🧭 Quick Start

    Refly 0.9.0 isn't just a version update — it's a reset on how AI tools should support creative work.

    We're excited for you to try it.

    As always, we're building Refly for creators like you.

  5. v0.8.02025-07-18 · 2025-07-18 09:20 UTC
    查看更新说明

    v0.8.0 Release Notes

    🌈 Highlights

    Refly v0.8.0 marks a major upgrade in both perception capabilities and workflow completeness:

    • Launch of the brand new Multimodal Engine — enabling seamless generation of images, videos, music, and voice
    • Agents now support full background execution, so your tasks can continue even if you close the browser
    • Introduction of a unified Model Supplier Store for managing 20+ AI model providers with one-click connectivity
    • Major improvements to canvas sync, performance, and user experience throughout the system

    🌟 New Features

    • 🎥 Multimodal Generation Fully Supported (Image / Video / Audio)
      • Generate 🖼️ images, 🎬 videos, 🎧 music, and 🔊 voice directly within your workflow
      • Visual/audio results can be previewed live in the canvas and written to context for downstream use
      • Integrated with lyria-2 for dynamic BGM generation
      • Integrated with chatterbox for lifelike multilingual voice synthesis
    • 🔁 End-to-End Multimodal Flow Orchestration
      • All generated media can be passed as structured variables to downstream nodes
      • Seamless integration with OCR, transcription, summarization, and rewriting nodes
      • Build a complete loop from one-line prompt to polished multimodal content
    • 🤖 Agent Background Execution
      • Agents now support full background task execution
      • You can safely close your browser — your skills will continue running on the server
      • Revisit Refly any time to check results or outputs
    • ⚙️ New: Model Supplier Store
      • A new unified panel to manage and configure 20+ model providers
      • Currently supports Replicate, OpenRouter, HuggingFace, and more
      • One-click connectivity testing for each provider after setup

    🛠️ System Optimizations & Bug Fixes

    🚀 Performance & Stability Improvements

    • Rebuilt canvas sync logic to significantly reduce node disappearance or connection issues
    • Fixed duplicate links in long-running AI conversations
    • Corrected MCP selection count and visibility bugs

    🌟 User Experience Enhancements

    • Auto-naming support for generated code widgets when using skill creation
    • Answer nodes now only display their current output — not full historical conversations

    Let your workflows not only think — but also see, speak, and perform.

    Refly v0.8.0 brings multimodal intelligence into every step of creation.

    Let your ideas move. 🧠🎥🎶

用户点评

登录用户的真实使用记录，可按版本和硬件筛选
共 0 条

    还没有人点评。第一个说出真实体验的人最有价值。

    README 原文

    未翻译
    在 GitHub 打开

    Refly — Agent Skills Builder Powered by Vibe Workflow

    APIs for Lovable · Webhooks for Lark/Feishu · Skills for Claude Code · Build Clawdbot

    Skills are not prompts. They are durable infrastructure.

    Refly is the first open-source platform for building stable, atomic, and versioned agent skills.

    Try it Now

    Refly Skills

    Refly Skills is the official executable skill registry for Refly.

    • ⚡ Run instantly: Execute skills in Refly with one click
    • 🧩 Reusable infrastructure: Versioned skills, not one-off prompts
    • 🔌 Export anywhere: Ship skills to Claude Code or deploy as APIs
    • 🌍 Community-powered: Import, fork, and publish your own skills

    Explore the registry: Refly Skills Repo

    Skills are deterministic agent capabilities—reusable across workflows, teams, and runtimes.

    TL;DR: Refly compiles your enterprise SOPs into executable agent skills. Built in 3 minutes. Shipped anywhere.

    Quick Start

    Deploy Refly

    • 📘 Self-Deployment Guide
      (Recommended for Developers) Step-by-step guide to deploying Refly on your own server using Docker.

    • 🔌 API Reference
      Complete API documentation for integrating Refly into your applications.

    [!TIP] Want to explore instantly? Open the hosted workspace: Refly Workspace

    What's Next?

    After deployment, choose your path based on your use case:

    I want to...Start hereTime
    🔧 Build my first workflowCreate a Workflow5 mins
    🔌 Call workflows via APIAPI Integration10 mins
    💬 Connect to Lark/FeishuWebhook Setup15 mins
    🤖 Export for Claude CodeExport Skills15 mins
    🦞 Build a ClawdbotBuild Clawdbot20 mins

    Why Refly?

    Most AI Agents fail in production because they rely on "Vibe-coded" scripts and fragile, black-box logic. As the ecosystem moves toward agentic frameworks like Claude Code, AutoGen, and MCP, the bottleneck is no longer the LLM—it's the lack of standardized, reliable actions.

    Refly bridges the gap between raw APIs and intelligent agents. We allow you to codify messy business logic into structured, version-controlled Agent skills that any agent can invoke with 100% reliability.

    Stop hard-coding tools. Build modular skills once in Refly's visual IDE and deploy them as MCP servers, standard APIs, or portable SDKs to any agent framework. Try it Now

    Core Capabilities

    🎯 Construct with Vibe (Copilot-led Builder)

    Describe your business logic in natural language, and Refly's Model-Native DSL compiles your intent into a high-performance skill.

    • Intent-Driven Construction: Describe the work once; Refly turns intent into deterministic, reusable, and composable skills.
    • Efficiency at Scale: Streamlined DSL optimized for LLMs, ensuring fast execution and significantly lower token costs.
    • 3-Minute Deployment: Transition from a static enterprise SOP to a production-ready agent skill in under 3 minutes.

    ⚡ Execute with Control (Intervenable Runtime)

    Break the "black box" of AI execution with a stateful runtime designed for deterministic reliability.

    • Intervenable Runtime: Pause, audit, and re-steer agent logic mid-run to ensure 100% operational compliance.
    • Deterministic Guarantees: Enforce strict business rules that minimize hallucinations and handle failure recovery.

    🚀 Ship to Production (Unified Agent Stack)

    Unify MCP integrations, tools, models, and reusable skills into a single execution layer.

    • Universal Delivery: Export as APIs for Lovable, webhooks for Slack or Lark/Feishu, or native tools for Claude Code and Cursor.
    • Stable Scheduling: Run workflows reliably on schedule with managed execution.

    🏛️ Govern as Assets (Skill Registry)

    Transform fragile scripts into governed, shared infrastructure across your organization.

    • Central Skill Registry: Securely manage, version, and share agent capabilities.
    • Team Workspace Collaboration: Build together with native version control and audit logs.

    Ecosystem

    Refly is designed to be the universal bridge between your existing enterprise toolchain and the next generation of agentic runtimes.

    Tooling & Protocols (Inputs)

    Bring your own data and logic into Refly with zero friction.

    • 3,000+ Native Tools: Seamlessly integrate with Stripe, Slack, Salesforce, GitHub, etc.

    A full list of supported model and tools providers can be found in provider-catalog.json.

    • MCP Support: Full compatibility with Model Context Protocol servers
    • Private Skill Connectors: Connect to your databases, scripts, and internal systems

    Agent Runtimes & Platforms (Outputs)

    Export your deterministic skills to any environment where work happens.

    • AI Coding Tools: Native export for Claude Code and Cursor (coming soon)
    • App Builders: Power Lovable or custom frontends via stateful APIs
    • Automation Hubs: Deploy as webhooks for Slack, Lark/Feishu, or Microsoft Teams
    • Agent Frameworks: Compatible with AutoGen, Manus, LangChain, and custom Python stacks

    Why Teams Choose Refly

    For Builders: From Vibe to Production

    Most agent tooling today falls into two categories:

    • Workflow builders (n8n, Dify): Great for orchestration, but workflows are fragile, trigger-only "black boxes," and hard to reuse.
    • Agent frameworks (LangChain): Powerful primitives, but require heavy engineering, manual boilerplate, and high maintenance to keep running. Refly eliminates the friction of manual configuration, giving you the fastest path from a "vibe" to a usable agent tool. By using our Streamlined DSL, you get the speed of a GUI with the precision of code.
    DimensionLegacy Automation (n8n, Dify)Code-First SDKs (LangChain)Refly Skills
    Interaction DepthTrigger-only Black boxProgrammatic Code changesIntervenable runtimeSteer logic mid-run
    ConstructionManual API wiring & JSONManual Python/TS boilerplateCopilot-ledDescribe intent → skills generated
    RecoveryFail = restart from scratchDebug → redeploy → rerunHot-fixRepair workflows during execution
    PortabilityHard to reuse across environmentsFramework-specificExport everywhereTo Claude Code, Cursor, Manus
    DeploymentLimited function toolsCustom microservicesProduction ReadyStateful, validated APIs

    For Enterprise: Scalable Skills Governance

    Workflow tools like n8n are great for basic connectivity, and frameworks like LangChain offer powerful primitives — but neither provides the governed, production-ready capability layer required for enterprise agent infrastructure.

    Refly acts as the Agent skills builder, providing the governance and reliability infrastructure required to deploy AI across the entire organization.

    Enterprise RequirementLegacy Tools (Workflow-first)SDKs (Code-first)Refly (Skill OS)
    Governance & ReuseTemplates are copied andreconfigured per instanceNo native registryfor sharing logicCentral skill registryVersioned, shareable capability assets
    Operational ReliabilityTrigger-basedlimited recoveryCustom handling requiredStateful runtimeWith validation + failure recovery
    SOP EnforcementWorkflows driftacross copiesDepends on manualengineering disciplineSOP-grade deterministic skillsWith controlled execution
    DeploymentInstance-bound workflowsCode maintained manuallyper teamLocal-first, on-prem readyOpen-source infrastructure
    Total Cost (TCO)Overhead grows withworkflow complexityHigh engineeringmaintenance costsMinimal DSLReduces token spend

    Create Your First Workflow

    [!NOTE] This section assumes you have completed self-deployment and can access Refly at http://localhost:5700

    Step 1: Register and Log In

    1. Open http://localhost:5700 in your browser
    2. Register with your email and password
    3. Configure your first model provider:
      • Click the account icon (top right) → Settings
      • Add a provider (e.g., OpenAI, Anthropic)
      • Add your first chat model
      • Set it as default

    📖 Detailed setup with screenshots: Self-Deployment Guide

    Step 2: Create a Workflow

    1. Click "New Workflow" on the home page
    2. Choose a template or start from scratch:
      • Blank Canvas: Build with visual nodes
      • Vibe Mode: Describe your workflow in natural language

    Example - Product Research Workflow:

    1. Add "Web Search" node - searches for product information
2. Add "LLM" node - analyzes search results
3. Add "Output" node - formats the report
4. Connect the nodes
5. Click "Save"

    [!TIP] If you want the fastest path, start with Vibe Mode and iterate from a template.

    Step 3: Test Your Workflow

    1. Click "Run" button
    2. Enter test input (e.g., a product URL)
    3. View execution results in real-time
    4. Check logs if something fails

    Use Cases

    Use Case 1: API Integration

    Goal: Call your workflow from your application via REST API

    Get Your API Credentials

    1. Go to Settings → API Keys
    2. Click "Generate New Key"
    3. Copy your API key (keep it secure!)

    Make Your First API Call

    curl -X POST https://your-refly-instance.com/api/v1/workflows/{WORKFLOW_ID}/execute \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "input": {
      "product_url": "https://example.com/product"
    }
  }'

    Response:

    {
  "execution_id": "exec_abc123",
  "status": "running"
}

    Check Execution Status

    curl https://your-refly-instance.com/api/v1/executions/{execution_id} \
  -H "Authorization: Bearer YOUR_API_KEY"

    📖 Full API Documentation: API Reference

    Use Case 2: Webhook for Lark (Feishu)

    Goal: Trigger your workflow when someone sends a message in Lark

    Prerequisites

    • A Lark workspace with admin access
    • A workflow created in Refly

    Setup Steps

    1. In Refly:

      • Open your workflow
      • Click "Settings" → "Triggers"
      • Enable "Webhook Trigger"
      • Copy the Webhook URL

    2. In Lark/Feishu:

      • Go to api.feishu.com/apps
      • Create a "Custom App"
      • Navigate to "Event Subscriptions"
      • Paste the Refly Webhook URL into "Request URL"
      • Click "Add Event" and select "Receive Message"
      • Go to "Version Management" and publish the app

    3. Test:

      • In Feishu, find your bot in the search bar and send a message (e.g., analyze report.pdf)
      • Your workflow executes and returns results via the webhook

    ⚠️ Note: Detailed Lark/Feishu integration guide coming soon. For now, see API Reference for webhook configuration.

    Use Case 3: Skills for Claude Code

    Goal: Publish your Refly workflows as Claude Code skills

    Quick Start

    1. Install Refly CLI
    npm install -g @powerformer/refly-cli
    1. Install a Skill
    # Via Refly CLI
refly skill install <skill-id>

#Via npx
npx skills add refly-ai/<skill-name>
    1. Publish a Skill
    refly skill publish <skill-id>

    The skill is now available in Claude Code, Cursor, and MCP-powered workflows. Agent can invoke your workflow as a tool!

    📖 Documentation: Refly Skills

    Use Case 4: Build Clawdbot

    📖 Tutorial: Build Clawdbot Tutorial

    Community & Support

    Join Us

    Join our community to get help, share your experience, and connect with other Refly users: Refly Community

    Contributing

    For those who'd like to contribute code, see our Contribution Guide. At the same time, please consider supporting Refly by sharing it on social media and at events and conferences.

    We are looking for contributors to help translate Refly into languages other than Mandarin or English. If you are interested in helping, please see the Contribution Guide for more information.

    Star History

    Star History Chart

    License

    This repository is licensed under the ReflyAI Open Source License, which is essentially the Apache 2.0 License with some additional restrictions.

    相关项目

    同分类里 star 最高的项目
    对比全部 →
    dify
    langgenius·Python

    可视化搭建 Agentic 工作流和 RAG 流水线，支持多种大模型与工具，可云端或私有化部署，帮团队从原型直接走到生产。

    Agent 编排+4Pipeline 编排
    160823+23425k1.19.02026-11-10 11:09 UTC
    airflow
    apache·Python

    用 Python 代码定义、调度和监控数据工作流，把 ETL/ELT 管道写成 DAG，告别手写 cron 和脚本拼接。

    工作流编排+4Pipeline 编排
    47435+3118k3.4.02026-11-10 09:53 UTC
    ray
    ray-project·Python

    分布式 AI 计算引擎，提供核心分布式运行时和一组 AI 库，用来加速机器学习负载、大模型推理与超参搜索。

    分布式计算+4Pipeline 编排
    44248+208.3kray-2.60.02026-11-10 10:22 UTC
    compose
    docker·Go

    用 YAML 文件定义多容器应用，一条命令就能把整套服务编排起来跑，省去逐个 docker run 的麻烦。

    容器编排+4Pipeline 编排
    38676+455.9kv5.7.02026-11-10 11:06 UTC