需 Docker 24+ 与 Docker Compose v2+，Kubernetes 路线需 Kubernetes 1.24+ 与 Helm 3+；部署依赖 deploy/env/.env 配置，缺失时回退到 docker/.env 或示例文件；infrastructure 组件为必选，application 与 data-process 默认选中，裁剪部署时需手动关闭。
nexent
ModelEngine-GroupMIT活跃维护零代码自动生成生产级 AI Agent 的平台，用统一工具、技能、记忆和编排，配合内置约束与反馈回路，让 Agent 稳定可控。
Nexent is a zero-code platform for auto-generating production-grade AI agents using Harness Engineering principles — unified tools, skills, memory, and orchestration with built-in constraints, feedback loops, and control planes.
v2.8.4发布于 2026-11-03 12:11 UTC6,174star +17 / 7 天778fork同步于 2026-11-10 05:34 UTC
git clone https://github.com/ModelEngine-Group/nexent.git
cd nexent
bash deploy.sh docker 硬件要求按 README 说明
README 未给出 GPU 或显存要求；Docker 部署最低 4 核 CPU、8 GiB 内存、40 GiB 磁盘，推荐 8 核、16 GiB、100 GiB，支持 x86_64 与 ARM64。Kubernetes 部署最低 16 GiB 内存、100 GiB 磁盘。
有 Docker 镜像
零代码平台，用自然语言自动生成生产级 AI Agent，省去编排和拖拽
零代码自动生成生产级 AI Agent 的平台，用统一工具、技能、记忆和编排，配合内置约束与反馈回路，让 Agent 稳定可控。
核心能力
- 用自然语言描述需求即可生成可执行 Agent
- 统一工具、技能、记忆与编排，内置约束和反馈回路
- 分层记忆分用户级与用户-Agent 级持久上下文
- 渐进式 Skill 披露按需加载以节省上下文窗口
- 支持 A2A 协议实现多 Agent 协作与分布式工作流
容易踩的坑
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
- v2.8.4最新正式版2026-11-03 · 2026-11-03 12:11 UTC
Bug 修复 / Bug Fixes
- 优化离线部署包的镜像导出，回滚按平台过滤的导出逻辑，旧版单平台镜像不再导致离线包构建中断。 / Improved image export for offline deployment packages by rolling back the platform-filtered export, so legacy single-platform images no longer abort the offline package build.
- 优化自动化消息来源的迁移逻辑，仅按显式运行消息引用归类，不再扫描全部消息单元；历史会话的活跃时间保持为空并沿用会话创建时间兜底。 / Improved the automation message-source migration to classify only explicit run message references instead of scanning all message units, and left legacy activity timestamps empty so conversation creation time remains the fallback.
- v2.8.32026-10-31 · 2026-10-31 06:08 UTC
查看更新说明
Bug 修复 / Bug Fixes
- 修复代理组内权限的历史数据问题：将旧表单遗留的 EDITABLE 值归一化为 EDIT，使非创建者、非管理员用户在界面上显示的编辑权限与实际行为一致。 / Fixed legacy agent in-group permissions by normalizing the EDITABLE value left by the previous deployment form to EDIT, so non-creator and non-admin users now actually get the edit access the UI already advertised.
- 提升北向 Chat API 的流式响应向前兼容性，新增 final_answer_mode 参数以在完整根答案与增量最终答案事件之间选择，默认行为保持兼容。 / Improved forward compatibility of the northbound Chat API streaming response with a new final_answer_mode parameter that selects between one complete root answer and incremental final-answer events, with the default keeping existing behaviour.
What's Changed
- feat(frontend): group MCP tools and unify description hover previews by @xuyaqist in https://github.com/ModelEngine-Group/nexent/pull/4261
- fix(sql): 归一化代理组内权限遗留 EDITABLE 值为 EDIT by @charmingchi1 in https://github.com/ModelEngine-Group/nexent/pull/4260
- [hotfix/v2.8.3] Backport #4255: separate application and navigation branding by @bernard1234 in https://github.com/ModelEngine-Group/nexent/pull/4262
- fix(agent): deduplicate execution history and add output compaction and recall by @JasonW404 in https://github.com/ModelEngine-Group/nexent/pull/4264
- 🐛 Bugfix: Forward-compatible streaming response for the northbound Chat API by @YehongPan in https://github.com/ModelEngine-Group/nexent/pull/4271
- feat(a2a): 将A2A代理调用超时时间调整为15分钟并统一配置 by @charmingchi1 in https://github.com/ModelEngine-Group/nexent/pull/4265
- v2.8.3 by @jeffwu-1999 in https://github.com/ModelEngine-Group/nexent/pull/4272
Full Changelog: https://github.com/ModelEngine-Group/nexent/compare/v2.8.2...v2.8.3
- v2.8.22026-10-28 · 2026-10-28 02:46 UTC
查看更新说明
Bug 修复 / Bug Fixes
- 修复对话来源信息在长列表中卡顿的问题：来源元数据改为两行展示并由样式裁剪，悬停与详情按需加载，来源与图片分页统一为每页 20 条，前台刷新做了合并处理，引用在手动翻页后仍能正确定位。 / Fixed chat source metadata stalling on long lists: metadata now renders as two clipped rows, hover and details load lazily, sources and images paginate at 20 items per page, foreground refresh is coalesced, and citations still relocate correctly after manual paging.
- 修复 AIDP 知识库目录分页过少导致四列布局残缺的问题：默认每页 12 条并提供 12/24/48 可选，切换卡片与表格视图时保留记录、页码与筛选条件且不重新请求。 / Fixed the AIDP knowledge catalog showing too few items to fill four-column rows: pages now default to 12 with 12/24/48 options, and switching between card and table views preserves records, page size, and filters without refetching.
- 修复个人信息栏复制 API 秘钥时结果带掩码星号的问题：秘钥列表默认仍返回掩码值，权限层在确认用户有权复制后才取回原文，避免其他调用方意外泄露秘钥。 / Fixed copied API keys in the profile panel carrying mask asterisks: the token list still returns masked values by default, and only the permission layer fetches the raw key after confirming the user may copy it, keeping other callers leak-safe.
What's Changed
- 🐛 Bugfix: 修复个人信息栏 API 秘钥复制结果带掩码星号的问题 (#4198) by @MoeexT in https://github.com/ModelEngine-Group/nexent/pull/4240
- 🐛 Fix(AIDP): backport source inspection and pagination fixes to v2.8.1 by @cj2026-bit in https://github.com/ModelEngine-Group/nexent/pull/4239
Full Changelog: https://github.com/ModelEngine-Group/nexent/compare/v2.8.1...v2.8.2
- v2.8.12026-10-27 · 2026-10-27 08:21 UTC
查看更新说明
🚀 Nexent：开源智能体平台 / Nexent: Open Source Intelligent Agent Platform 我们很高兴地宣布Nexent v2.8.1 正式发布！🎉 Nexent 是一个开源智能体平台，能够将流程的自然语言转化为完整的多模态智能体 —— 无需编排，无需复杂拖拉拽。基于 MCP 工具生态，Nexent 提供强大的模型集成、数据处理、知识库管理、零代码智能体开发能力。我们的目标很简单：将数据、模型和工具整合到一个智能中心中，使日常工作流程更智能、更互联。 We are excited to announce that Nexent v2.8.1 is released! 🎉 Nexent is an open-source agent platform that turns process-level natural language into complete multimodal agents — no diagrams, no wiring. Built on the MCP tool ecosystem, Nexent provides model integration, data processing, knowledge-base management, and zero-code agent development. Our goal is simple: to bring data, models, and tools together in one smart hub, making daily workflows smarter and more connected. 新功能 / New Features
- 智能体对话历史支持按分组分页加载，滚动到底部时按需续取，长会话的首屏与滚动性能明显改善。 / Agent conversation history now paginates by group with scoped infinite scrolling, so long sessions load faster and scroll smoothly.
- 智能体空间新增标签展示与筛选能力，我的智能体卡片统一了信息层级与操作入口。 / Agent Space now surfaces tags and filtering on agent cards, with a unified information hierarchy and action layout in the My Agents view.
- 智能体编辑页支持直接选用已配置的默认模型，无需每次手动重选。 / The agent editor now selects the configured default model automatically instead of asking you to pick it every time.
- 浏览器标签页标题与图标跟随产品品牌和当前语言同步更新。 / Browser tab titles and icons now follow the product branding and the active language. Bug 修复 / Bug Fixes
- 修复 SSE 连接中断后对话无法继续的问题，断线后可正常恢复后续回复。 / Fixed conversations getting stuck after an SSE disconnect; replies now resume correctly once the stream recovers.
- 修复外部 A2A 智能体名称含特殊字符时，智能体调用名称校验不通过的问题。 / Fixed agent invocation name validation rejecting externally registered A2A agents whose names contain special characters.
- 修复智能体对话页默认展示数量不当、以及查看对话详情时接口被重复调用的问题。 / Fixed an inappropriate default number of agents shown on the conversation page and duplicate calls to the conversation-detail API.
What's Changed
- 🐛 Bugfix: Adjusted the default number of agents displayed on the agent conversation page, and fixed an issue where the API for viewing agent conversation details was being called multiple times. by @YehongPan in https://github.com/ModelEngine-Group/nexent/pull/4209
- 🐛 Bugfix: Fixed an issue with agent invocation name validation regarding compatibility with special external A2A names. by @YehongPan in https://github.com/ModelEngine-Group/nexent/pull/4212
- Fix chat continuation after SSE disconnect and document verification by @hhhhsc701 in https://github.com/ModelEngine-Group/nexent/pull/4221
- fix(chat): paginate agent history by group with scoped infinite scrol… by @hzwllm in https://github.com/ModelEngine-Group/nexent/pull/4225
- fix(agent): select configured default model in agent editor by @xuyaqist in https://github.com/ModelEngine-Group/nexent/pull/4228
- [hotfix/v2.8.1] fix(frontend): 同步浏览器标签页的项目名称与 Logo by @bernard1234 in https://github.com/ModelEngine-Group/nexent/pull/4226
- Hotfix/v2.8.1 by @jeffwu-1999 in https://github.com/ModelEngine-Group/nexent/pull/4231
Full Changelog: https://github.com/ModelEngine-Group/nexent/compare/v2.8.0...v2.8.1
- v2.8.02026-10-23 · 2026-10-23 10:59 UTC
查看更新说明
🚀 Nexent：开源智能体平台 / Nexent: Open Source Intelligent Agent Platform
我们很高兴地宣布Nexent v2.8.0 正式发布！🎉
Nexent 是一个开源智能体平台，能够将流程的自然语言转化为完整的多模态智能体 —— 无需编排，无需复杂拖拉拽。基于 MCP 工具生态，Nexent 提供强大的模型集成、数据处理、知识库管理、零代码智能体开发能力。我们的目标很简单：将数据、模型和工具整合到一个智能中心中，使日常工作流程更智能、更互联。
We are excited to announce that Nexent v2.8.0 is released! 🎉
Nexent is an open-source agent platform that turns process-level natural language into complete multimodal agents — no diagrams, no wiring. Built on the MCP tool ecosystem, Nexent provides model integration, data processing, knowledge-base management, and zero-code agent development. Our goal is simple: to bring data, models, and tools together in one smart hub, making daily workflows smarter and more connected.
新功能 / New Features
- 新增 A2A 智能体协作协议支持，提供适配层、能力目录与校验模块，智能体可对外暴露可发现、可调用的服务端点。 / Added A2A agent collaboration protocol support with an adapter layer, capability catalog, and validation modules, so agents can expose discoverable and callable service endpoints.
- 新增智能体发布页签与访问详情，可查看发布状态、访问方式与 A2A 接入信息。 / Added an agent publication tab with access details, showing publication status, access methods, and A2A connection information.
- AIDP 知识库管理能力扩展：新增知识库详情页与完整的知识库创建、文件导入流程。 / Extended AIDP knowledge-base management with a knowledge-base detail page and complete creation and file-import flows.
- AIDP 检索支持图谱检索（graph search）与术语覆盖重排（jargon coverage），可在智能体与会话的检索配置中选择。 / AIDP retrieval now supports graph search and jargon coverage reranking, selectable in agent and conversation retrieval settings.
- 新增 DeepSeek 内置联网搜索工具，可在智能体中直接调用。 / Added a built-in DeepSeek web search tool that can be invoked directly from agents.
- 模型调用支持温度与 Top P 参数覆盖，并在参数弹窗中补齐两项参数的中英文说明与可聚焦提示。 / Model calls now support temperature and Top P overrides, with bilingual parameter guidance and keyboard-focusable help tooltips in the parameter dialog.
- 资源管理新增安全审计日志，覆盖后端管理类 API 的关键操作。 / Added security audit logging to backend management APIs.
- 新增官方智能体合集，内置 26 个开箱即用智能体，覆盖教育、金融、政务、人力资源、法律、营销、医疗七个行业，并配套技能包。 / Added an official agent collection with 26 ready-to-use agents across education, finance, government, human resources, legal, marketing, and medical domains, bundled with companion skills.
- 智能体仓库支持排序，并可自定义列表列宽与预置智能体头像。 / Agent repository now supports sorting, resizable list columns, and preset agent avatars.
- 本地工具支持不经过沙箱直接调用，官方工具动作可延迟启动 Docker，降低本地执行开销。 / Local tools can now be invoked without the sandbox, and official tool actions defer Docker startup to reduce local execution overhead.
- 自动化会话支持保留运行记录并统一活跃时间统计，标题改写采用独立版本号避免竞态。 / Automation conversations now preserve run records with unified activity tracking, and title renames use an independent version counter to avoid races.
- 新增可选的 Nginx HTTPS 终止配置，便于在部署时启用 TLS。 / Added optional Nginx HTTPS termination for enabling TLS at deployment time.
- 路由切换时展示骨架屏过渡，替代整页加载指示，导航期间保留源页面状态。 / Added shimmer skeletons during route navigation instead of a full-page loader, preserving source-page state while loading.
重构优化 / Refactoring & Improvements
- 重构后台服务目录结构，按业务模块重整 Management services 与公共工具，提升可维护性。 / Reorganized backend services by business domain, restructuring management services and shared utilities for better maintainability.
- 数据处理链路重构，以 Celery prefork 替代原有 Ray 实现，任务调度与恢复更稳定。 / Refactored the data-processing pipeline to use Celery prefork in place of Ray, making task scheduling and recovery more robust.
- 重构模型服务与模型/智能体契约和工具函数，统一调用入口与参数传递方式。 / Refactored the model service along with model/agent contracts and utilities, unifying entry points and parameter passing.
- 重构智能体提示词体系，统一提示词的生成、存储与注入路径。 / Refactored the agent prompt system, unifying how prompts are generated, stored, and injected.
- 全新改版智能体与对话工作流，统一工作台资源选择组件与推荐交互。 / Redesigned the agent and chat workflows, unifying workbench resource selection and recommendation interactions.
- 标签管理入口集中到资源管理页，同步优化知识库与资源配置交互。 / Consolidated tag management into the resource management page and refined knowledge-base
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。
Nexent is a zero-code platform for auto-generating production-grade AI agents, built on Harness Engineering principles. It provides unified tools, skills, memory, and orchestration with built-in constraints, feedback loops, and control planes — no orchestration, no complex drag-and-drop required, using pure language to develop any agent you want.
One prompt. Endless reach.
🚀 Get Started Now
⭐ Before you get started, please star us on GitHub — your support drives us forward!
Deploy on Your Own
If you need to run Nexent locally or in your private infrastructure, we offer two deployment options:
System Requirements
| Resource | Docker | Kubernetes |
|---|---|---|
| CPU | 4 cores (min) / 8 cores (rec.) | 4 cores (min) / 8 cores (rec.) |
| Memory | 8 GiB (min) / 16 GiB (rec.) | 16 GiB (min) / 64 GiB (rec.) |
| Disk | 40 GiB (min) / 100 GiB (rec.) | 100 GiB (min) / 200 GiB (rec.) |
| Architecture | x86_64 / ARM64 | x86_64 / ARM64 |
| Software | Docker 24+, Docker Compose v2+ | Kubernetes 1.24+, Helm 3+ |
Note: Recommended configurations ensure optimal performance in production environments.
Docker Deployment (Recommended for Individuals/Small Teams)
Quick and straightforward for most users. Prerequisites: Docker 24+ and Docker Compose v2+:
git clone https://github.com/ModelEngine-Group/nexent.git
cd nexent
bash deploy.sh docker
The root deploy.sh only forwards to the target deploy script; the native Docker implementation is bash deploy/docker/deploy.sh. The Docker and Kubernetes deploy scripts share the same deployment configuration model. Interactive runs show Bash TUI menus for component selection, port policy, and image source. infrastructure is required; application and data-process are selected by default and can be disabled when you want a smaller deployment. Use b/Backspace to return to the previous TUI step and q to quit. Use --defaults to skip the TUI and deploy with saved deploy.options or built-in defaults. Non-interactive runs can also pass the same choices with --version, --components, --port-policy development|production, and --image-source general|mainland|local-latest. Successful deployments save non-sensitive choices to each deploy directory's deploy.options for reuse on the next run.
Docker and Kubernetes both use deploy/env/.env as the runtime configuration file. Existing deploy/env/.env is kept as-is. If it does not exist, the deploy scripts first reuse docker/.env, then fall back to deploy/env/.env.example. Monitoring-specific settings are generated from deploy/env/monitoring.env.example into deploy/env/monitoring.env.
Docker uninstall is handled by bash uninstall.sh docker. It can preserve or delete data volumes: run it interactively, pass --delete-volumes true|false, or use bash uninstall.sh docker delete-all to remove containers and persistent data.
Offline image packages can be built with bash build.sh --package --target docker --compress true or bash deploy/offline/build_offline_package.sh --target docker --compress true. The package includes image tar files, load-images.sh, push-images.sh, root deploy/uninstall entrypoints, deployment scripts, SQL files, manifest.yaml, and checksums.txt. Package deploys use saved deploy.options or built-in defaults without opening the TUI; add --config to configure interactively. Deploy with bash deploy.sh --load-images docker ... on the target host, or use bash deploy.sh --push-images --image-registry-prefix registry.example.com/nexent docker ... to push loaded images to an internal registry and deploy with that image prefix. When --push-images is used without a prefix, deploy.sh asks for it before push-images.sh prompts for the registry username and password.
For detailed deployment instructions, see Docker Installation.
Kubernetes Deployment (For Enterprise Production)
Ideal for enterprise scenarios requiring high availability and elastic scaling. Prerequisites: Kubernetes 1.24+ and Helm 3+:
git clone https://github.com/ModelEngine-Group/nexent.git
cd nexent
bash deploy.sh k8s
The native Kubernetes implementation is bash deploy/k8s/deploy.sh. It installs two independent Helm releases: nexent-infrastructure for Elasticsearch, PostgreSQL, Redis, and MinIO, then nexent for application services after infrastructure and the Elasticsearch API key are ready. Use --release-scope all|infrastructure|nexent to operate both releases or either side independently. The script reads deploy/env/.env and renders explicit Helm ConfigMap and Secret overrides. Use --persistence-mode local|dynamic|existing, --storage-class/--sc, --local-path, --local-node-name, and --existing-claim-prefix to control PVC behavior.
Kubernetes uninstall is handled by bash uninstall.sh k8s. By default it removes nexent first and nexent-infrastructure second. The same --release-scope option supports application-only removal and dependency-protected infrastructure removal.
Kubernetes offline packages use the same builder with --target k8s or --target all. Run load-images.sh on every cluster node that needs the images, or use --push-images --image-registry-prefix registry.example.com/nexent to push the images to an internal registry before deploying with the same version, image source, and image registry prefix.
For detailed deployment instructions, see Kubernetes Installation.
✨ Core Features
Nexent provides a comprehensive feature set for building powerful AI agents:
| Feature | Description |
|---|---|
| ⚙️ Multi-Model Integration | OpenAI-compatible with any provider, full LLM/Embedding/VLM/STT/TTS coverage, supports domestic model switching |
| 🤖 Zero-Code Agent Generation | Describe requirements in natural language, generate executable agents instantly, what you think is what you get |
| 🤝 A2A Agent Collaboration | Agent-to-Agent protocol enables seamless multi-agent cooperation and distributed workflows |
| 🧠 Layered Memory Mechanism | Two-tier memory (user-level + user-agent-level) for persistent context across conversations |
| 📝 Progressive Skill Disclosure | Dynamically loads Skill into context, maximizing context window efficiency |
| 🗄️ Personal-Grade Knowledge Base | Real-time import and intelligent retrieval for 20+ document formats, auto summaries, fine-grained access control |
| 🔧 MCP Tool Ecosystem | Plug-and-play extension system with custom development and third-party MCP service support |
| 🌐 Internet Knowledge Integration | Multi-source search blending real-time information with private data |
| 🔍 Knowledge-Level Traceability | Precise citations and source verification, full transparency for every fact |
| 🎭 Multimodal Interaction | Voice, text, images, files — comprehensive natural dialogue |
| 🔢 Agent Version Management | Version iteration and history rollback, safe and controllable |
| 🏪 Agent Marketplace | Official and community curated agents, one-click install and use |
| 👥 Multi-Tenancy & RBAC | Multi-tenant isolation, role-based access control, fine-grained resource management |
🤝 Join Our Community
If you want to go fast, go alone; if you want to go far, go together.
We have released Nexent v2.0! A comprehensive upgrade from v1.0, featuring A2A protocol support, progressive Skill disclosure, layered memory mechanism, user management with multi-tenancy, agent version management, agent marketplace, and more.
- 🗺️ Check our Feature Map to explore current and upcoming features.
- 🔍 Try the current build and leave ideas or bugs in the Issues tab.
Rome wasn't built in a day.
If our vision speaks to you, jump in via the Contribution Guide and shape Nexent with us.
Early contributors won't go unnoticed: from special badges and swag to other tangible rewards, we're committed to thanking the pioneers who help bring Nexent to life.
Most of all, we need visibility. Star ⭐ and watch the repo, share it with friends, and help more developers discover Nexent — your click brings new hands to the project and keeps the momentum growing.
📖 What's Next
Ready to dive deeper? Here are the main documentation entry points:
- Quick Start — System requirements and deployment guide
- Core Features — Comprehensive feature documentation
- User Guide — Agent development and usage
- Developer Guide — Build from source and customization
- FAQ — Common questions and troubleshooting
📄 License
Nexent is licensed under the MIT License.
TradingAgents
TauricResearch·Python
用多个 LLM 智能体分工协作模拟金融交易决策，把研究、分析、下单等环节拆给不同角色，适合做量化交易策略的智能体实验。
多智能体+4多智能体协作
121689+62723kv0.7.12026-11-04 10:13 UTC
MetaGPT
FoundationAgents·Python
用自然语言描述需求，就能让多个 LLM 智能体扮演产品经理、架构师、工程师等角色协作，自动产出需求文档、设计稿和代码。
多智能体协作+4多智能体协作
71233+289.1kv0.8.22026-01-21 10:12 UTC
camel
camel-ai·Python
多智能体框架，让多个大模型智能体通过角色扮演和对话自主协作，用于研究智能体规模化规律并构建协作式 AI 系统。
多智能体+4多智能体协作
18024+62.1kv0.2.91a72026-11-09 10:25 UTC
ekko-studio
EKKOLearnAI·TypeScript
本地优先的多智能体工作台，支持多智能体对话、编码和可视化工作流，桌面端和网页端都能用。
多智能体+4多智能体协作
11736+211.5kv0.8.202026-11-10 04:35 UTC