朝夕
a1adbd00af
* Refactor tool loading to use static database assignment - Add `tools` field to `WorkerIndividual` and `SystemNodeConfig` models - Update frontend WorkerIndividualSettings to support multi-select tools and select `bound_skill` from available backend API - Deprecate dynamic `get_tool` method during agent execution, replacing it with `load_tools_from_list` via absolute paths during agent initialization - Fix deprecated tool loading usage across `SupervisoryNode`, `ConsciousnessNode`, `ControlNode`, and `BaseIndividual` - Provide `reload` API route to clear `GlobalIndividualManager` state to rebuild nodes cleanly upon next request - Fix `user_test.py` related to unhashed password assignments in testing mock object Co-authored-by: zhaoxi826 <198742034+zhaoxi826@users.noreply.github.com> * Refactor tool loading to use static database assignment part 2 - Restore BaseIndividual's `_init_agent` method to inject parsed `tools_list` via absolute paths dynamically using `load_tools_from_list` - Restore React component multi-select support for tools dropdown via `/api/v1/resource/tool` backend response integration - Restore `user_test.py` compatibility related to access hasher assignments - Passed 75 tests with no errors Co-authored-by: zhaoxi826 <198742034+zhaoxi826@users.noreply.github.com> * Refactor tool loading to use static database assignment - Add `tools` field to `WorkerIndividual` and `SystemNodeConfig` models - Update frontend WorkerIndividualSettings to support multi-select tools and select `bound_skill` from available backend API - Deprecate dynamic `get_tool` method during agent execution, replacing it with `load_tools_from_list` via absolute paths during agent initialization - Fix deprecated tool loading usage across `SupervisoryNode`, `ConsciousnessNode`, `ControlNode`, and `BaseIndividual` - Provide `reload` API route to clear `GlobalIndividualManager` state to rebuild nodes cleanly upon next request - Fix `user_test.py` related to unhashed password assignments in testing mock object Co-authored-by: zhaoxi826 <198742034+zhaoxi826@users.noreply.github.com> * Finalize all tool loading refactoring with complete changes - Merged database schema enhancements for `tools` multi-select list - Completed the `load_tools_from_list` via absolute paths utility update - Refactored `AgentFactory` and corresponding initialization calls in `BaseIndividual`, `SupervisoryNode`, `ConsciousnessNode`, and `ControlNode` to inject statically via constructor - Updated frontend UI `WorkerIndividualSettings` with UI for `bound_skill` selection and dynamic multi-tool selector array - Removed all legacy usages of dynamic `tools=tool` invocation - Provided backend `/reload` API and automatic clearing of instance cache upon individual configuration update - Repaired mock password assignment logic in testing suite Co-authored-by: zhaoxi826 <198742034+zhaoxi826@users.noreply.github.com> * Restore complete tool loading refactor logic - Re-apply BaseIndividual tools resolution logic - Correctly patch API, UI files, system nodes and testing scripts avoiding checkout loss - Passed all 75 integration and unit tests successfully Co-authored-by: zhaoxi826 <198742034+zhaoxi826@users.noreply.github.com> --------- Co-authored-by: google-labs-jules[bot] <161369871+google-labs-jules[bot]@users.noreply.github.com> Co-authored-by: zhaoxi826 <198742034+zhaoxi826@users.noreply.github.com> |
||
|---|---|---|
| config | ||
| docs | ||
| frontend | ||
| pretor | ||
| tests | ||
| .dockerignore | ||
| .env | ||
| .gitignore | ||
| .python-version | ||
| Dockerfile | ||
| LICENSE | ||
| Makefile | ||
| README.md | ||
| docker-compose.yml | ||
| main.py | ||
| pyproject.toml | ||
| uv.lock | ||
README.md
Pretor (执政官)
一款基于 Python 的分布式多 Agent 协作系统
Pretor 是一款基于 Ray 构建的下一代分布式多 Agent 协作系统。项目采用“中心监管 + 边缘执行”的异构集群模式，通过大参数 MoE 模型进行高层逻辑推理，并协同微调后的轻量化模型高效完成具体任务。借助 Pydantic-AI 提供的强类型约束与 FastAPI 异步网关，Pretor 实现了任务从需求拆解、资源调度到自动化执行的全链路闭环，为个人提供可靠的人工智能助手服务。
特别之处？
- 本项目通过 Ray 和 vllm 实现个人个性化助手的创建，你可以通过收集符合你自己偏好的数据，构建独属于自己的风格的人工智能助手。
- 本项目通过多 Agent 协作，实现比起单 Agent 系统更强的性能，从而完成更复杂的任务。
- 本项目通过调用本地模型，从而减少对于昂贵的API使用和一定程度上的安全保护。如果你是创作者，可以通过用自己的作品去训练属于自己的模型，提高工作效率。
- （暂未实现）本项目适配多种消息平台，实现在外可通过多种方式给 Pretor 下达指令完成工作。
- （暂未实现）本项目内置 growth_node(生长节点) ，实现傻瓜式微调模型操作，让你的 Pretor 自己学会一些独特的技能。
快速开始
本项目正在开发中...
项目子项目
pretor-viceroy:
项目名称 : 总督
github网址 :https://github.com/zhaoxi826/viceroy
功能 :pretor的资源管理工具
目前实现 :
- 对于skill的安装与解析
pretor-domain
项目名称 : 域
github网址 :（暂无）
功能 : 轻量化沙箱
目前实现 :
（规划中）
pretor-explorer
项目名称 : 探索者
github网址 :（暂无）
功能 : 自动爬虫
目前实现 :
（规划中）
pretor-pioneer
项目名称 : 先驱者
github网址 :（暂无）
功能 : RAG检索增强生成
目前实现 :
（规划中）