Commit Graph

411 Commits

Author SHA1 Message Date
朱潮
ee65d208a8 Merge branch 'prod' into staging 2026-03-24 10:54:59 +08:00
朱潮
2d82d7c8b6 merge from developing 2026-03-24 10:54:54 +08:00
朱潮
e13405ba29 shell_env support 2026-03-24 00:12:19 +08:00
朱潮
29da20fa22 soffice sharp 支持 2026-03-18 23:54:41 +08:00
shuirong
822589eab6 ci: 重新调整CI 2026-03-18 16:16:53 +09:00
朱潮
e2e0a7d985 跨语言场景 - 即使用户用其他语言提问,也必须用指定语言回复 2026-03-17 22:03:23 +08:00
朱潮
380764d5ed add chromium 2026-03-17 14:48:50 +08:00
朱潮
95e34ed172 add chromium 2026-03-17 14:27:16 +08:00
朱潮
f24c3ff78f playwright support 2026-03-16 23:07:00 +08:00
朱潮
c27270588f 增加取消推理 2026-03-16 22:22:39 +08:00
shuirong
0a6ec95518 feat(CI): 添加 onprem-dev 环境的构建和部署配置
在 CircleCI 配置中新增 onprem-dev 环境的 build-and-push 和 deploy 任务,部署到 cluster-for-B 的 onprem-dev 命名空间

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-03-16 16:18:50 +09:00
朱潮
32fd8c8656 shell_env支持 2026-03-16 13:31:59 +08:00
朱潮
b8368068ae 心跳发送增加 15 秒间隔判断,发送后也更新 last_yield_time 2026-03-15 16:53:18 +08:00
朱潮
8cab0180e3 update novare 2026-03-13 21:50:57 +08:00
朱潮
7a058065c1 在 routes/skill_manager.py:669-672,解压完成后立即检测并删除 __MACOSX 目录,这样后续的 skill
验证流程就不会受到干扰。
2026-03-13 10:08:38 +08:00
朱潮
5de9803d0c Merge branch 'feature/agentic_rag' 2026-03-12 19:54:31 +08:00
朱潮
9f669e1b6d dataset支持dataset_id软连接 2026-03-12 19:54:09 +08:00
朱潮
0469ea2ecd 心跳输出格式改成json 2026-03-12 19:32:37 +08:00
朱潮
94471c90d8 添加心跳机制 2026-03-12 16:57:44 +08:00
朱潮
082630e281 table_rag_retrieve 和 rag_retrieve 增加调用策略 2026-03-12 11:59:38 +08:00
朱潮
8640f99674 table_rag 超时时间改成300 2026-03-11 22:05:37 +08:00
朱潮
e3c400f663 邮箱列表大幅扩展 2026-03-11 16:08:41 +08:00
朱潮
8360e1467c 已更新 description,移除了具体存储服务(Cloudflare R2、AWS S3、S3-compatible storage)的描述,只保留了文件上传操作相关的说明。 2026-03-11 12:50:54 +08:00
朱潮
21fb6c924f remove excel-analysis 2026-03-11 12:47:37 +08:00
朱潮
c1d2d48979 add table_rag 2026-03-11 12:25:31 +08:00
朱潮
4adf62afb7 add document skill 2026-03-11 09:27:08 +08:00
朱潮
67290f7e5e add config/s3-upload-sparticle.yaml 2026-03-11 09:21:43 +08:00
朱潮
25272015a0 add public 2026-03-11 09:21:18 +08:00
朱潮
611bd23a7c add skill 2026-03-11 09:02:05 +08:00
朱潮
f830b8bb08 add table_rag_retrieve 2026-03-10 16:13:48 +08:00
朱潮
38c5d1e622 修复方案:在 routes/webdav.py 中添加了一个 WSGI 中间件包装函数,在请求到达 WsgiDAV 前:
1. 读取 X-Forwarded-Proto 设置正确的 wsgi.url_scheme
  2. 将 Destination 头中的外部 scheme/host 重写为内部实际值
2026-03-09 18:51:08 +08:00
朱潮
5d97be9557 routes/webdav.py — 完全重写,从手写 WebDAV 协议改为使用 WsgiDAV 开源库 2026-03-09 18:33:05 +08:00
朱潮
4ad1c96bf3 ALLOWED_RESOURCE_TYPES 从 {robot, dataset} 改为 {robot, docs}。现在 WebDAV 路径为: 2026-03-09 16:00:07 +08:00
朱潮
9fd4a099da remove resource_id 2026-03-09 15:48:35 +08:00
朱潮
7039bec61a add webdav support 2026-03-09 12:54:38 +08:00
朱潮
49034bc571 add webdav support 2026-03-09 12:31:07 +08:00
朱潮
b277c9bbff add webdav support 2026-03-09 11:56:17 +08:00
朱潮
85cc57ce1a aupdate from_v2_request 新增 model_name、model_server、api_key 三个可选参数 2026-03-06 12:53:16 +08:00
朱潮
8264257ab6 现在 from_v2_request 的 model_name 和 model_server 优先级逻辑为:
1. 最高优先级:generate_cfg 中的 model / model_server(需不为空且不等于 whatever)
  2. 回退:bot_config 中的值(原有逻辑)
2026-03-06 12:47:33 +08:00
朱潮
49151916c6 Merge branch 'master' of https://github.com/sparticleinc/catalog-agent 2026-03-05 15:08:34 +08:00
朱潮
4c74acaf18 chat.py:730 的 v2 接口 exclude_fields 现在与 v1 接口保持一致,补上了 'model', 'model_server', 'dataset_ids', 'system_prompt', 'mcp_settings',
'robot_type', 'enable_thinking', 'skills', 'enable_memory' 这些字段。
2026-03-05 15:08:09 +08:00
autobee-sparticle
6309a53860
feat(novare): 添加人员在離判定规则,利用 last_communication_time 字段判断人员是否在 NOVARE 楼内 (#13)
EB API 始终返回人员最后已知坐标,不区分在馆/离馆。通过检查 last_communication_time
与当前时刻的差值(阈值5分钟/24小时),实现三级在離判定并提供对应的日语回答模板。

Closes sparticleinc/mygpt-frontend#1963

Co-authored-by: zhuchao <zhuchaowe@163.com>
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-04 17:21:16 +09:00
朱潮
b6b821c9bb bug fix 2026-03-04 14:30:44 +08:00
朱潮
f7773a93e1 Merge branch 'upgrade/deepagents-0.4.4' 2026-03-04 14:08:51 +08:00
朱潮
52d8161b0e 修改确保 max_length 永远不会是 None 2026-03-04 14:08:26 +08:00
朱潮
9215f8236d 🐛 fix: 修复 Mem0 连接池耗尽和 LLM 参数透传问题
- mem0_manager: 添加 _cleanup_mem0_instance 方法,在缓存移除实例时显式释放数据库连接,避免等待 GC 导致连接池耗尽
- deep_assistant: 根据 model_provider 过滤不支持的参数,Anthropic 不支持 OpenAI 特有参数如 n、presence_penalty 等

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2026-03-04 10:11:44 +08:00
朱潮
b3b6dd85d9 排查n参数 2026-03-04 09:57:36 +08:00
autobee-sparticle
a6f071119f
Merge pull request #12 from sparticleinc/bugfix/autobee-20260302-memory-results-fix
fix(memory): handle Mem0 get_all response format with results key
2026-03-02 22:29:13 +09:00
86b38262fa fix(memory): handle Mem0 get_all response format with results key
Mem0's get_all() returns data in format {"results": [...]}, not a direct list.
Also, agent_id is at the top level of each memory, not in metadata.

Changes:
- Add _extract_memories_from_response() to handle both new and old response formats
- Add _check_agent_id_match() to check agent_id at both top-level and in metadata
- Update get_all_memories, delete_memory, delete_all_memories to use new helpers

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-02 13:28:39 +00:00
autobee-sparticle
572374b297
fix(memory): handle Mem0 get_all returning string or dict items (#11)
Mem0's get_all() method may return a list containing either strings
(older version format) or dictionaries (newer version format). This
fix adds proper type checking to handle both cases gracefully.

- get_all_memories: Skip string items, only process dict items
- delete_memory: Add type checking before accessing dict methods
- delete_all_memories: Add type checking before accessing dict methods

Fixes: 'str' object has no attribute 'get' error

Co-authored-by: zhuchao <zhuchaowe@163.com>
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-02 22:02:59 +09:00