zhayujie/bot-on-anything
Bot On Anything
This project is a multi-channel AI agent and chatbot framework that allows a single AI intelligence to be deployed across various messaging platforms, web interfaces, and email accounts. It functions as a cross-model AI gateway, providing a unified interface to route requests between different large language model providers.
The system is distinguished by its autonomous task planning and knowledge management capabilities. It can decompose complex goals into sequential execution steps using external tools and a headless browser, while simultaneously extracting information from conversations to build a structured knowledge base with visual relationship graphs.
The platform includes a tiered memory system for long-term context retrieval, a dynamic skill plugin engine for extending functionality, and the ability to execute remote system commands. It also supports multimodal workflows, enabling the exchange of images, voice messages, and files across different communication channels.
The instance can be managed through a web console, a terminal-based command interface, or via in-chat slash commands.
Features
- Chatbot Frameworks - Provides a framework for building and deploying LLM-powered conversational agents across multiple messaging services and web interfaces.
- Task Decomposition Orchestrators - Decomposes complex user goals into sequential execution steps that are iteratively processed by autonomous agents.
- Multi-Channel AI Deployments - An agent that deploys a single AI intelligence across Telegram, Slack, Gmail, and other communication platforms.
- Long-term Memory Stores - Implements persistent storage for retaining context and information across multiple user sessions.
- Hybrid Short-and-Long Term Memory - Implements a memory system combining immediate session context with long-term persistent archives using vector and keyword retrieval.
- Skill Installation - Allows the installation of new agent capabilities from remote URLs, community hubs, or natural language.
- Extensible CLI Assistants - Ships an extensible CLI assistant for managing process lifecycles, system status, and configurations.
- AI Backend Integrations - Provides a unified connectivity layer to route requests to various local AI models and third-party proxy backends.
- AI Gateways - Functions as a unified interface for managing and routing requests between diverse large language model providers.
- AI Knowledge Management - Organizes conversation history and documents into structured knowledge bases and visual graphs to support AI query responses.
- Automated Knowledge Extraction - Automatically extracts key information from documents and conversations to create a structured, interlinked knowledge base.
- Automated Skill Loading Systems - Dynamically loads functional modules and custom skills installed from remote hubs or generated via natural language.
- Autonomous Agent Orchestrators - Provides a runtime environment that decomposes complex goals into multi-step plans using external tools and memory.
- Autonomous System Controllers - Implements an LLM-driven controller that manages local files, terminal commands, and web browsing via action-observation loops.
- External Tool Integrations - Connects the AI agent to system utilities, web search, and file storage using standardized context protocols.
- Language Model Backend Switching - Provides a unified interface to switch between different LLM providers and backends via a central console.
- LLM Provider Integrations - Implements authentication adapters and configurations for connecting to multiple external large language model providers.
- Multi-Platform Chatbot Development - Provides development patterns for a single bot backend that operates across various messaging services like Telegram and Slack.
- Task Planners - Implements logic to break down complex user objectives into sequential, executable steps for autonomous agents.
- Knowledge Structuring - Organizes distilled information into structured knowledge pages with indexes and cross-references.
- Agent Memory Persistence - Uses keyword and vector retrieval to persist important information and ensure context continuity.
- Knowledge Page Distillations - Periodically summarizes conversation history to build a structured knowledge base with visual relationship graphs.
- Conversational Knowledge Generators - Extracts information from conversations into a structured wiki and visual knowledge graph for long-term memory.
- Autonomous Planning Engines - Autonomously decomposes high-level objectives into executable action sequences using adaptive planning and external tools.
- Messaging Platform Integrations - Deploys a single AI instance across multiple messaging platforms like Telegram, Slack, and other social apps.
- Multi-Channel Message Abstractions - Standardizes incoming and outgoing data formats to ensure a single AI instance operates consistently across diverse messaging channels.
- Agent Skill Extenders - Features a dynamic engine that loads isolated tools and skills at runtime to extend agent capabilities.
- Remote Command Execution - Enables the AI agent to execute terminal commands and perform file operations on a remote operating system.
- Provider-Agnostic LLM Routing - Provides a routing layer that translates standardized requests into provider-specific API calls for seamless LLM switching.
- Group Chat Routing - Supports multi-user group chat interactions across various messaging platforms via intelligent routing.
- Session Context Managers - Manages and clears immediate conversation history to control how the AI remembers current interactions.
- Multimodal Capabilities - Enables the chatbot to process and generate multiple media types, including images, voice messages, and files.
- Context Summarizations - Summarizes conversation history into long-term storage when context limits are reached or on a fixed schedule.
- Multimodal AI Orchestrators - Coordinates multiple AI model types to process and exchange text, images, and voice across communication channels.
- Real-Time Web Search Integrations - Integrates real-time internet searching to ground AI reasoning with current external data.
- Conversational Memory Consolidation - Provides autonomous logic to consolidate memories and refine agent skills by reviewing idle conversations.
- Tool-Protocol Standardizations - Standardizes the interface between the AI model and system utilities like terminals and file storage for agentic operations.
- Knowledge Base Interfacing - Provides interfaces for administering the knowledge base, including directory views and access toggles.
- Semantic Knowledge Base Search - Provides semantic search and index lookups to retrieve specific information from the internal knowledge base.
- Context Search Retrievers - Uses keyword and vector-based semantic matching to retrieve historical context and inject it into active conversations.
- Headless Browser Automation - Controls a headless browser to interact with web elements, fill forms, and extract real-time data.
- Management Consoles - Includes a built-in web-based management console for chatting and configuring system settings.
- Slash Command Interfaces - Implements a slash-command interface within chats to modify runtime settings and manage agent skills.
- Knowledge Relationship Graphs - Visualizes interlinked knowledge base documents through an interactive relationship graph.
- Document Content Extraction - Extracts text and data from documents, images, and PDFs to be processed by the AI.
- Multimodal Message Exchange - Supports the exchange of images, files, and voice messages across multiple supported communication channels.
- Memory Activation Triggers - Triggers the condensation and distillation of long-term memory to optimize information retrieval.
- Headless Browser Controllers - Integrates a remote headless browser to navigate dynamic web content and extract data for the AI model.
