🚀 Nexent:开源智能体平台 / Nexent: Open Source Intelligent Agent Platform
我们很高兴地宣布Nexent v2.6.0 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 v2.6.0 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
- 重构模型管理页面:统一模型编辑对话框,新增模型容量自动推荐(内置 LiteLLM 模型目录、离线可用)、自定义参数校验与推理参数配置。 / Revamped the model management page: unified model edit dialog, automatic model capacity suggestions (bundled LiteLLM catalog, works offline), custom parameter validation, and inference parameter configuration.
- 新增完整会话线程(Thread)管理能力,覆盖创建、切换与并发控制,并配套遥测指标。 / Added full conversation thread management, covering creation, switching, and concurrency control, with telemetry support.
- 上线外部记忆体系:智能体可在最终回答后自动提取记忆,支持外部记忆提供方检索(top_k、超时可配置),并提供 AIDP Memory 插件与插件热加载。 / Introduced the external memory system: agents automatically extract memories after final answers, with configurable external provider search (top_k, timeout), an AIDP Memory plugin, and hot-reloadable plugins.
- 优化 Agent 创建体验:NL2Agent 生成流程与落地卡片打磨,新增 NL2Agent 优化建议,提示词同步升级。 / Improved the agent creation experience: polished NL2Agent generation flow and landing cards, with new NL2Agent optimization suggestions and upgraded prompts.
- 检索引用交互升级:回答中的引用支持统一悬浮卡片与点击定位高亮。 / Upgraded retrieval citations: references in answers now support unified hover cards with click-to-focus highlighting.
- 重构统一标签(Label)系统,支持标签库权限、资源打标与筛选,并为多版本 Agent 提供协作版本标签。 / Refactored the unified label system with tag library permissions, resource tagging, and filtering, plus collaborative version labels for multi-version agents.
- 支持模型优先级排序,可按优先级为智能体选择模型。 / Added model priority sorting so agents can select models by priority.
- 沙箱适配 Anthropic Skills,可直接运行 Anthropic 技能格式。 / Adapted the sandbox to run Anthropic skills natively.
- S3 上传/下载工具支持用户自定义配置。 / The upload_to_s3 and download_from_s3 tools now support user configuration.
- 日志支持落盘为文件,便于排查与归档;服务启动时自动恢复中断的后台任务。 / Logs are now written to disk as files for easier troubleshooting, and interrupted background tasks are automatically recovered on service startup.
Bug 修复 / Bug Fixes
- 完成上下文预算契约迁移,增加有界压缩与真实溢出恢复,长对话不再丢失上下文。 / Completed the context budget contract migration with bounded compaction and real overflow recovery, preventing context loss in long conversations.
- 修复会话链路多处问题:返回后恢复会话、线程切换保留智能体、会话标题使用会话模型生成、正确渲染执行代码与对比模式下的历史保留。 / Fixed several conversation-flow issues: conversation restore after returning, agent retention across thread switches, session-model-based title generation, correct execution-code rendering, and history retention in compare mode.
- 修复多模态网关适配器缓存键缺陷,模型配置变更后不再使用过期适配器。 / Fixed the multimodal gateway adapter cache keying defect so model configuration changes no longer serve stale adapters.
- 修复模型健康管理问题:部分绑定模型健康检查失败时智能体仍可用,健康检查改为内联图标入口,embedding 连通性探测通过的 base_url 正确持久化,被删除模型给出明确提示。 / Fixed model health management: agents stay available when some bound models fail health checks, the health check moved to an inline icon, the embedding base_url validated by probing is persisted correctly, and deleted models now show clear notices.
- 修复知识库问题:删除中的知识文件/知识库处理、AIDP 知识库保存与会话向量模型提示、搜索范围提示准确性,以及损坏数据跳过处理。 / Fixed knowledge-base issues: in-flight knowledge file and knowledge base deletion, AIDP knowledge base saving and session vector model prompts, accurate search scope notices, and corrupted data skipping.
- 修复评估模块的页签统计过期与流式关键字重复问题。 / Fixed stale tab totals and duplicate stream keywords in the evaluation module.
- 修复多租户存储配额、复制智能体名称冲突及旧版 Docker 沙箱容器启动失败等问题。 / Fixed tenant storage quota restoration, copied agent name conflicts, and sandbox container startup failures on older Docker versions.
文档与部署变化 / Docs & Deployment Changes
- 新增 nexent-infrastructure Helm Chart(Elasticsearch、PostgreSQL、Redis、MinIO)并配置健康探针;项目配置支持在 Docker 与 Kubernetes 部署间持久化,并修复 K8s 离线部署的镜像加载、共享存储 PV 渲染与离线技能资产问题。 / Added the nexent-infrastructure Helm chart (Elasticsearch, PostgreSQL, Redis, MinIO) with health probes; project configuration now persists across Docker and Kubernetes deployments, with fixes for K8s offline image loading, shared-storage PV rendering, and offline skill assets.
- 数据库初始化改为幂等,统一标签迁移整合为 v2.5.2 版本化脚本,新增 v2.6.0 推理参数等增量迁移。 / Made database bootstrap idempotent, consolidated unified tag migrations into the versioned v2.5.2 script, and added incremental migrations including v2.6.0 inference parameters.
- 完善中英文文档体系(Mermaid 图表、侧边栏本地搜索),新增智能体评估用户指南与第三方资源集成文档,并更新自动化测试与离线包构建流水线。 / Improved the bilingual documentation system (Mermaid diagrams, sidebar local search), added the agent evaluation user guide and third-party resource integration docs, and updated the automated testing and offline package build pipelines.
What's Changed
- fix: update tests for merged v2.5.0 migrations by @JasonW404 in #3823
- ✨ Feature: Add NL2Agent optimization suggestions by @RedShakespeare in #3821
- 🐛 Fix: avoid duplicate stream keyword in evaluator LLM calls by @cj2026-bit in #3829
- Bugfix: 当Agent配置的模型被删除,应该有对应的提示 by @xuyaqist in #3832
- ♻️ Remove agent market url by @WMC001 in #3840
- ci: build offline package after images by @hhhhsc701 in #3826
- Add nexent-infrastructure Helm chart with health endpoints by @hhhhsc701 in #3795
- ✨ Feature: Sandbox adaptation for Anthropic's skills. by @YehongPan in #3839
- fix: restore unlimited tenant storage quota by @EDDIWARD in #3844
- fix(agent-repository): resolve copied agent name conflicts by @MoeexT in #3803
- ♻️ Refactor: remove deprecated app config stuff by @Jasonxia007 in #3841
- 📝 Add docs to illustrate how to integrate third-party resources (in&out) by @SimengBian in #3838
- ♻️ Refactor phase 1: agent_service / vectordatabase_service / skill_service by @Jasonxia007 in #3843
- Feat: update v2.5 documentation and synchronize Chinese and English content by @DoYX in #3853
- Recover interrupted tasks on service startup by @hhhhsc701 in #3847
- 🐛 Bugfix: conflict causing nexent-runtime failure by @Jasonxia007 in #3854
- feat: refactor label system by @MoeexT in #3809
- 🐛 Bugfix: Fixed an issue where sandbox containers failed to start correctly in Docker version 18.09 and earlier versions. by @YehongPan in #3856
- 🐛 Bugfix: Fixed an issue where sandbox containers failed to start correctly in Docker version 18.09 and earlier versions. by @YehongPan in #3857
- Update offline package OBS upload destination by @hhhhsc701 in #3855
- 🐛 Fix: calculate usage from source file size and show ES physical usage separately by @cj2026-bit in #3845
- ✨ Feature: agent evaluation user guide by @cj2026-bit in #3834
- Bugfix: 调整Agent-Config的顺序 by @xuyaqist in #3846
- ✨Feature: Logs are written to disk as files by @WMC001 in #3859
- feat(memory): extract memories from final answers by @JasonW404 in #3860
- feat(memory): complete external memory provider integration by @JasonW404 in #3690
- docs: 完善 v2.5.0 文档体系 — 中英文同步、Mermaid 图表支持、侧边栏本地搜索 by @charmingchi1 in #3858
- fix: move unified tag SQL out of init.sql and consolidate #3809 migrations into v2.5.2 by @MoeexT in #3861
- ✨feat: Add and update GitHub workflow for automated testing by @DongJiBao2001 in #3865
- fix(deploy): include official skills in K8s offline packages by @JasonW404 in #3875
- fix(k8s): correct shared storage PV rendering by @JasonW404 in #3876
- fix(database): make bootstrap and suadmin initialization idempotent by @JasonW404 in #3878
- fix(deploy): load K8s images into containerd first by @JasonW404 in #3877
- feat: interactive retrieval citations with unified hover cards by @EDDIWARD in #3654
- 🐛 Bugfix: The upload_to_s3 and download_from_s3 tools do not support user configuration. by @YehongPan in #3879
- fix(memory): allow external retrieval without embedding by @JasonW404 in #3882
- fix: allow tag migration for versioned agents by @MoeexT in #3867
- fix(memory): let enabled providers drive external search by @JasonW404 in #3883
- 🐛 Fix: prevent stale totals when switching result tabs by @cj2026-bit in #3866
- 🐛 Fix: align AIDP knowledge base name validation by @cj2026-bit in #3864
- 🐛 Fix:In-Flight Knowledge-File and Knowledge-Base Deletion by @cj2026-bit in #3842
- ♻️ Refactor: extract reusable ResourceCard components by @ayiya12 in #3862
- bugfix: remove identity-keyed multimodal gateway adapter cache by @jeffwu-1999 in #3887
- bugfix(agent-debug): preserve chat history across compare-mode toggle by @bernard1234 in #3885
- ✨ feat: add product documentation builder by @WMC001 in #3889
- 🐛Fix:Persist the embedding base_url that connectivity probing validated by @jeffwu-1999 in #3891
- Support alternate reasoning stream fields by @hhhhsc701 in #3863
- fix(memory): 保留并热加载记忆插件 by @JasonW404 in #3890
- feat(memory-plugin): add AIDP Memory plugin by @JasonW404 in #3894
- 🐛 Fix: remove redundant parallel_executor imports from generated code by @cj2026-bit in #3892
- bugfix(newchat): render parsed execution code by @xuyaqist in #3886
- Bugfix: retain agent during thread switches by @xuyaqist in #3884
- fix: skip processing corrupted data by @MoeexT in #3903
- Fix: keep agents available when some bound models fail health checks by @jeffwu-1999 in #3900
- fix(agents): show related agent version labels by @xuyaqist in #3906
- fix(context): add bounded compaction and overflow recovery by @JasonW404 in #3907
- fix(context): 完成 W2 预算契约迁移与真实溢出恢复 by @JasonW404 in #3910
- ♻️ Refactor: Refactored ReAct recoverable exceptions by @YehongPan in #3915
- 🐛bugfix:Fix AIDP knowledge base saving and session vector model prompts by @DongJiBao2001 in #3918
- bugfix: User displays username by @hzwllm in #3911
- 🐛Fix: preserve configured KB scope for accurate search notices by @cj2026-bit in #3904
- Bugfix: polish agent landing cards and NL2Agent generation prompts by @bernard1234 in #3925
- Persist project configuration across Docker and Kubernetes deployments by @hhhhsc701 in #3923
- ♻️ Refactor: Optimization of the Agent Creation Experience by @YehongPan in #3926
- ✨ Support full thread management by @Jasonxia007 in #3917
- 🐛Bugfix: move health check out of dropdown into inline icon button by @ayiya12 in #3913
- fix(agent): preserve completed action speaker role by @JasonW404 in #3922
- feat(agents): support model priority sorting by @xuyaqist in #3914
- fix: generate conversation titles with session model (cherry-pick #3898 to develop) by @JasonW404 in #3919
- fix(newchat): restore conversation after returning by @xuyaqist in #3908
- Feat hzw 20260810 by @lijiayang619 in #3920
- Configure monitoring dashboard role access by @hhhhsc701 in #3928
- 🐛 Bugfix: httpx2 timeout by @Jasonxia007 in #3929
- feat: human-in-the-loop (HITL) functionality and support by @Dallas98 in #3927
- fix: stabilize unified tag migration for large datasets by @MoeexT in #3931
- Document backup, upgrade, and rollback procedures by @hhhhsc701 in #3924
- Fix/connectivity probe reasoning models by @lijiayang619 in #3932
- feat: Enhance HITL schema with conditional table creation and trigger management by @Dallas98 in #3933
- 🐛 Bugfix: sandbox-tool-bridge thread leak by @Jasonxia007 in #3934
- feat(sql): consolidate migrations after v2.5.1 into v2.6.0_merged_mig… by @jeffwu-1999 in #3936
- Fix/model edit resets connectivity by @lijiayang619 in #3941
- bugfix: 模型配置问题修改 by @hzwllm in #3940
- fix(mcp): unbind deleted tools from agent drafts by @xuyaqist in #3937
- fix(memory): match AIDP plugin entry filename case by @JasonW404 in #3939
- merge(develop): sync conflict resolutions from release/v2.6.0-merge by @jeffwu-1999 in #3942
- Revert "merge(develop): sync conflict resolutions from release/v2.6.0-merge" by @jeffwu-1999 in #3944
- Fix/override form consistency by @lijiayang619 in #3943
- merge(develop): sync conflict resolutions from release/v2.6.0-merge by @jeffwu-1999 in #3945
- merge(main): merge v2.6.0 release from develop by @jeffwu-1999 in #3946
New Contributors
- @lijiayang619 made their first contribution in #3920
Full Changelog: v2.5.1...v2.6.0