Soulter
f88031b0c9
* Refactor: using sqlmodel(sqlchemy+pydantic) as ORM framework and switch to async-based sqlite operation (#2294) * stage * stage * refactor: using sqlchemy as ORM framework, switch to async-based sqlite operation - using sqlmodel as ORM(based on sqlchemy and pydantic) - add Persona, Preference, PlatformMessageHistory table * fix: conversation * fix: remove redundant explicit session.commit, and fix some type error * fix: conversation context issue * chore: remove comments * chore: remove exclude_content param * Fix: 当多个相同消息平台实例部署时上下文可能混乱（共享） (#2298) * perf: update astrbot event session format, using platfrom id to ensure uniqueness fixes: #1000 * fix: 更新 MessageSession 类以使用 platform_id 作为唯一标识符，并调整相关方法以确保一致性 * fix: 更新 MessageSession 文档以明确 platform_id 的赋值规则，并调整 get_platform 和 get_platform_inst 方法的返回类型 * Improve: 引入全新的人格管理模式以及重构函数工具管理器 (#2305) * feat: add persona management * refactor: 重构函数工具管理器，引入 ToolSet，并让 Persona 支持绑定 Tools * feat: 更新 Persona 工具选择逻辑，支持全选和指定工具的切换 * feat: 更新 BaseDatabase 中的 persona 方法返回类型，支持返回 None * fix: platform id * feat: add support to sync mcp servers from ModelScope (#2313) * fix: 修复访问令牌的空格问题 * chore: 移除 MCP 市场相关逻辑 (#2314) * chore: 移除 MCP 市场相关路由 * Refactor: 重构配置文件管理，以支持更灵活的、会话粒度的（基于 umo part）配置文件隔离 (#2328) * refactor: 重构配置文件管理，以支持更灵活的、基于 umo part 的配置文件隔离 * Refactor: 重构配置前端页面，新增数个配置项 (#2331) * refactor: 重构配置前端页面，新增数个配置项 * feat: 完善多配置文件结构 * perf: 系统配置入口 * fix: normal config item list not display * fix: 修复 axios 请求中的上下文引用问题 * chore: remove status checking in chat page * fix: 修复 stage 在不同 pipeline 中被重复使用的问题和 persona 相关问题 * Feature: 增加图片转述提供商配置、支持用户自定义模型模态能力 (#2422) * feat: 增加图片转述提供商配置、支持用户自定义模型模态能力 * fix: 修复 LLMRequestSubStage 中会话管理方法参数不一致的问题，简化方法调用 * Feature: 优化 WebSearch 的爬取网页速度并且支持使用 Tavily 作为搜索引擎 (#2427) * feat: 优化了 websearch 的速度；支持 Tavily 作为搜索引擎 * fix: 优化日志记录格式，修复搜索结果处理中的索引和内容显示问题 * feat: 添加对话选中状态管理，优化默认对话加载逻辑 * feat: 支持通过解析URL 的方式导入网页数据到知识库 (#2280) * feat:为webchat页面添加一个手动上传文件按钮（目前只处理图片） * fix:上传后清空value，允许触发change事件以多次上传同一张图片 * perf:webchat页面消息发送后清空图片预览缩略图，维持与文本信息行为一致 * perf:将文件输入的值重置为空字符串以提升浏览器兼容性 * feat:webchat文件上传按钮支持多选文件上传 * fix:释放blob URL以防止内存泄漏 * perf:并行化sendMessage中的图片获取逻辑 * feat:完成从url获取部分的UI * feat: 添加从URL导入功能的组件 * fix: 优化导入结果处理，添加整体摘要和主题摘要的文件命名 * perf: 更新url导入选项添加默认值 * perf: 在导入url的部分配置项未启用时隐藏暂不使用的下拉框选项 * feat: 添加上传前提提示信息至导入url至知识库功能 * feat: 更新导入功能提示信息，添加上传状态通知 * fix: 优化url转知识库错误处理 * feat: 合并知识库的上传文件和 URL 标签页 * feat: 删除导入URL至知识库功能的相关组件 --------- Co-authored-by: Soulter <905617992@qq.com> * feat: 添加条件显示逻辑以优化插件配置项的可见性管理 (#2433) * Feature: 支持在 WebUI 配置文件页中配置默认知识库 (#2437) * feat: 支持配置默认知识库 * chore: clean code * refactor: 重构 Function Tool 管理并初步引入 Multi Agent 及 Agent Handsoff 机制 (#2454) * stage * refactor: 重构 Function Tool 管理并引入 multi agent handsoff 机制 - Updated `star_request.py` to use the global `call_handler` instead of context-specific calls. - Modified `entities.py` to remove the dependency on `FunctionToolManager` and streamline the function tool handling. - Refactored `func_tool_manager.py` to simplify the `FunctionTool` class and its methods, removing deprecated code and enhancing clarity. - Adjusted `provider.py` to align with the new function tool structure, removing unnecessary type unions. - Enhanced `star_handler.py` to support agent registration and tool association, introducing `RegisteringAgent` for better encapsulation. - Updated `star_manager.py` to handle tool registration for agents, ensuring proper binding of handlers. - Revised `main.py` in the web searcher package to utilize the new agent registration system for web search tools. * chore: websearch * perf: 减少嵌套 * chore: 移除未使用的 mcp 导入 * feat: 添加 WebUI 迁移助手以及相关迁移方法 (#2477) * fix: 修复迁移对话时的一些问题 * feat: 增加工具使用模型能力选项 * feat: 添加知识库插件更新检查和更新功能 * perf: 调整 WebUI sidebar 顺序 * refactor: 重构 SharedPreference 类并采用数据库存储替换 json 存储 (#2482) * perf: 使用 run_coroutine_threadsafe Co-authored-by: Raven95676 <raven95676@gmail.com> * Feature: 支持配置重排序模型（vLLM API 格式）用于 score 任务 (#2496) * feat: 支持添加重排序模型（vLLM API 格式）用于 score 任务 * fix: update rerank API base URL to use localhost * feat: 知识库支持配置重排序模型 * fix: remove debug print statement for reranked results in FaissVecDB * fix: 移除知识库中的提示文本 * Feature: 支持在配置文件配置可用的插件组 (#2505) * feat: 增加可用插件集合配置项 * remove: 旧版平台可用性配置 已经基于多配置文件实现。 * feat: 应用配置文件插件可用性配置 * perf: hoist if from if * feat: llm_tool 装饰器返回值支持返回 mcp 库中 tool 的返回值类型(mcp.type.CallToolResult) (#2507) * fix: add type definition for migrationDialog and ensure open method exists before calling * chore: update project version to 4.0.0 * feat: 多 t2i 服务的随机负载均衡 (#2529) * fix: bugfixes * Improve: 扩大配置文件生效范围的自定义程度到会话粒度 (#2532) * feat: 扩大配置文件生效范围的自定义程度 * perf: 冲突检测 * refactor: simplify config form validation and improve conflict message clarity * chore: clean code * feat: 插件配置支持多个快捷魔法配置项 * chore: 修复当自动更新 webchat title 时，history 被重置的问题 * bugfixes * feat: add custom T2I template editor (#2581) * perf: add option to clear provider selection in ProviderSelector component * 📦 release: bump verstion to v4.0.0-beta.1 * chore: delete uv.lock --------- Co-authored-by: RC-CHN <67079377+RC-CHN@users.noreply.github.com> Co-authored-by: Raven95676 <raven95676@gmail.com>
AstrBot is a loosely coupled, asynchronous chatbot and development framework that supports multi-platform deployment, featuring an easy-to-use plugin system and comprehensive Large Language Model (LLM) integration capabilities.
✨ Key Features
- LLM Conversations - Supports various LLMs including OpenAI API, Google Gemini, Llama, Deepseek, ChatGLM, etc. Enables local model deployment via Ollama/LLMTuner. Features multi-turn dialogues, personality contexts, multimodal capabilities (image understanding), and speech-to-text (Whisper).
- Multi-platform Integration - Supports QQ (OneBot), QQ Channels, WeChat (Gewechat), Feishu, and Telegram. Planned support for DingTalk, Discord, WhatsApp, and Xiaomi Smart Speakers. Includes rate limiting, whitelisting, keyword filtering, and Baidu content moderation.
- Agent Capabilities - Native support for code execution, natural language TODO lists, web search. Integrates with Dify Platform for easy access to Dify assistants/knowledge bases/workflows.
- Plugin System - Optimized plugin mechanism with minimal development effort. Supports multiple installed plugins.
- Web Dashboard - Visual configuration management, plugin controls, logging, and WebChat interface for direct LLM interaction.
- High Stability & Modularity - Event bus and pipeline architecture ensures high modularization and loose coupling.
Tip
Dashboard Demo: https://demo.astrbot.app/
Username:astrbot, Password:astrbot(LLM not configured for chat page)
✨ Deployment
Docker Deployment
See docs: Deploy with Docker
Windows Installer
Requires Python (>3.10). See docs: Windows Installer Guide
Replit Deployment
CasaOS Deployment
Community-contributed method.
See docs: CasaOS Deployment
Manual Deployment
See docs: Source Code Deployment
⚡ Platform Support
| Platform | Status | Details | Message Types |
|---|---|---|---|
| QQ (Official Bot) | ✔ | Private/Group chats | Text, Images |
| QQ (OneBot) | ✔ | Private/Group chats | Text, Images, Voice |
| WeChat (Personal) | ✔ | Private/Group chats | Text, Images, Voice |
| Telegram | ✔ | Private/Group chats | Text, Images |
| WeChat Work | ✔ | Private chats | Text, Images, Voice |
| Feishu | ✔ | Group chats | Text, Images |
| WeChat Open Platform | 🚧 | Planned | - |
| Discord | 🚧 | Planned | - |
| 🚧 | Planned | - | |
| Xiaomi Speakers | 🚧 | Planned | - |
Provider Support Status
| Name | Support | Type | Notes |
|---|---|---|---|
| OpenAI API | ✔ | Text Generation | Supports all OpenAI API-compatible services including DeepSeek, Google Gemini, GLM, Moonshot, Alibaba Cloud Bailian, Silicon Flow, xAI, etc. |
| Claude API | ✔ | Text Generation | |
| Google Gemini API | ✔ | Text Generation | |
| Dify | ✔ | LLMOps | |
| DashScope (Alibaba Cloud) | ✔ | LLMOps | |
| Ollama | ✔ | Model Loader | Local deployment for open-source LLMs (DeepSeek, Llama, etc.) |
| LM Studio | ✔ | Model Loader | Local deployment for open-source LLMs (DeepSeek, Llama, etc.) |
| LLMTuner | ✔ | Model Loader | Local loading of fine-tuned models (e.g. LoRA) |
| OneAPI | ✔ | LLM Distribution | |
| Whisper | ✔ | Speech-to-Text | Supports API and local deployment |
| SenseVoice | ✔ | Speech-to-Text | Local deployment |
| OpenAI TTS API | ✔ | Text-to-Speech | |
| Fishaudio | ✔ | Text-to-Speech | Project involving GPT-Sovits author |
🦌 Roadmap
Tip
Suggestions welcome via Issues <3
- Ensure feature parity across all platform adapters
- Optimize plugin APIs
- Add default TTS services (e.g., GPT-Sovits)
- Enhance chat features with persistent memory
- i18n Planning
❤️ Contributions
All Issues/PRs welcome! Simply submit your changes to this project :)
For major features, please discuss via Issues first.
🌟 Support
✨ Demos
Note
Code executor file I/O currently tested with Napcat(QQ)/Lagrange(QQ)
✨ Docker-based Sandboxed Code Executor (Beta) ✨
✨ Multimodal Input, Web Search, Text-to-Image ✨
✨ Natural Language TODO Lists ✨
✨ Plugin System Showcase ✨
✨ Web Dashboard ✨
✨ Built-in Web Chat Interface ✨
⭐ Star History
Tip
If this project helps you, please give it a star <3
Disclaimer
- Licensed under
AGPL-v3. - WeChat integration uses Gewechat. Use at your own risk with non-critical accounts.
- Users must comply with local laws and regulations.
私は、高性能ですから!