<picture> <img src="./docs/images/logo.png" alt="WeKnora Logo" height="120"/> </picture> <picture> <a href="https://trendshift.io/repositories/15289" target="_blank"> <img src="https://trendshift.io/api/badge/repositories/15289" alt="Tencent%2FWeKnora | Trendshift" style="width: 250px; height: 55px;" width="250" height="55"/> </a> </picture> <a href="https://weknora.weixin.qq.com" target="_blank"> <img alt="官方网站" src="https://img.shields.io/badge/官方网站-WeKnora-4e6b99"> </a> <a href="https://chatbot.weixin.qq.com" target="_blank"> <img alt="微信对话开放平台" src="https://img.shields.io/badge/微信对话开放平台-5ac725"> </a> <a href="https://github.com/Tencent/WeKnora/blob/main/LICENSE"> <img src="https://img.shields.io/badge/License-MIT-ffffff?labelColor=d4eaf7&color=2e6cc4" alt="License"> </a> <a href="./CHANGELOG.md"> <img alt="Version" src="https://img.shields.io/badge/version-0.3.4-2e6cc4?labelColor=d4eaf7"> </a> | English | <a href="./README_CN.md">简体中文</a> | <a href="./README_JA.md">日本語</a> | <h4 align="center">

Overview • Architecture • Key Features • Getting Started • API Reference • Developer Guide

</h4>

💡 WeKnora - LLM-Powered Document Understanding & Retrieval Framework

📌 Overview

WeKnora is an LLM-powered framework designed for deep document understanding and semantic retrieval, especially for handling complex, heterogeneous documents.

It adopts a modular architecture that combines multimodal preprocessing, semantic vector indexing, intelligent retrieval, and large language model inference. At its core, WeKnora follows the RAG (Retrieval-Augmented Generation) paradigm, enabling high-quality, context-aware answers by combining relevant document chunks with model reasoning.

Website: https://weknora.weixin.qq.com

✨ Latest Updates

v0.3.4 Highlights:

IM Bot Integration: WeCom, Feishu, and Slack IM channel support with WebSocket/Webhook modes, streaming, and knowledge base integration
Multimodal Image Support: Image upload and multimodal image processing with enhanced session management
Manual Knowledge Download: Download manual knowledge content as files with proper filename sanitization
NVIDIA Model API: Support NVIDIA chat model API with custom endpoint and VLM model configuration
Weaviate Vector DB: Added Weaviate as a new vector database backend for knowledge retrieval
AWS S3 Storage: Integrated AWS S3 storage adapter with configuration UI and database migrations
AES-256-GCM Encryption: API keys encrypted at rest with AES-256-GCM for enhanced security
Built-in MCP Service: Built-in MCP service support for extending agent capabilities
Agent Streaming Panel: Optimized AgentStreamDisplay with auto-scrolling, improved styling, and loading indicators
Hybrid Search Optimization: Grouped targets and reused query embeddings for better retrieval performance
Final Answer Tool: New final_answer tool with agent duration tracking for improved agent workflows

v0.3.3 Highlights:

🧩 Parent-Child Chunking: Hierarchical parent-child chunking strategy for enhanced context management and more accurate retrieval
📌 Knowledge Base Pinning: Pin frequently-used knowledge bases for quick access
🔄 Fallback Response: Fallback response handling with UI indicators when no relevant results are found
🖼️ Image Icon Detection: Automatic image icon detection and filtering in document processing
🧹 Passage Cleaning for Rerank: Passage cleaning for rerank model to improve relevance scoring accuracy
🐳 Docker & Skill Management: Enhanced Docker setup with entrypoint script and skill management
🗄️ Storage Auto-Creation: Storage engine connectivity check with auto-creation of buckets
🎨 UI Consistency: Standardized border styles, updated theme and component styles across the application
⚡ Chunk Size Tuning: Updated chunk size configurations for knowledge base processing

<details> <summary>Earlier Releases</summary>

v0.3.2 Highlights:

🔍 Knowledge Search: New "Knowledge Search" entry point with semantic retrieval, supporting bringing search results directly into the conversation window
⚙️ Parser & Storage Engine Configuration: Configure document parser engines and storage engines for different sources in settings, with per-file-type parser selection in knowledge base
🖼️ Image Rendering in Local Storage: Support image rendering during conversations in local storage mode, with optimized streaming image placeholders
📄 Document Preview: Embedded document preview component for previewing user-uploaded original files
🎨 UI Optimization: Knowledge base, agent, and shared space list page interaction redesign
🗄️ Milvus Support: Added Milvus as a new vector database backend for knowledge retrieval
🌋 Volcengine TOS: Added Volcengine TOS object storage support
📊 Mermaid Rendering: Support mermaid diagram rendering in chat with fullscreen viewer, zoom, pan, toolbar and export
💬 Batch Conversation Management: Batch management and delete all sessions functionality
🔗 Remote URL Knowledge: Support creating knowledge entries from remote file URLs
🧠 Memory Graph Preview: Preview of user-level memory graph visualization
🔄 Async Re-parse: Async API for re-processing existing knowledge documents

v0.3.0 Highlights:

🏢 Shared Space: Shared space with member invitations, shared knowledge bases and agents across members, tenant-isolated retrieval
🧩 Agent Skills: Agent skills system with preloaded skills for smart-reasoning agent, sandboxed execution environment for security isolation
🤖 Custom Agents: Support for creating, configuring, and selecting custom agents with knowledge base selection modes (all/specified/disabled)
📊 Data Analyst Agent: Built-in Data Analyst agent with DataSchema tool for CSV/Excel analysis
🧠 Thinking Mode: Support thinking mode for LLM and agents, intelligent filtering of thinking content
🔍 Web Search Providers: Added Bing and Google search providers alongside DuckDuckGo
📋 Enhanced FAQ: Batch import dry run, similar questions, matched question in search results, large imports offloaded to object storage
🔑 API Key Auth: API Key authentication mechanism with Swagger documentation security
📎 In-Input Selection: Select knowledge bases and files directly in the input box with @mention display
☸️ Helm Chart: Complete Helm chart for Kubernetes deployment with Neo4j GraphRAG support
🌍 i18n: Added Korean (한국어) language support
🔒 Security Hardening: SSRF-safe HTTP client, enhanced SQL validation, MCP stdio transport security, sandbox-based execution
⚡ Infrastructure: Qdrant vector DB support, Redis ACL, configurable log level, Ollama embedding optimization, DISABLE_REGISTRATION control

v0.2.0 Highlights:

🤖 Agent Mode: New ReACT Agent mode that can call built-in tools, MCP tools, and web search, providing comprehensive summary reports through multiple iterations and reflection
📚 Multi-Type Knowledge Bases: Support for FAQ and document knowledge base types, with new features including folder import, URL import, tag management, and online entry
⚙️ Conversation Strategy: Support for configuring Agent models, normal mode models, retrieval thresholds, and Prompts, with precise control over multi-turn conversation behavior
🌐 Web Search: Support for extensible web search engines with built-in DuckDuckGo search engine
🔌 MCP Tool Integration: Support for extending Agent capabilities through MCP, with built-in uvx and npx launchers, supporting multiple transport methods
🎨 New UI: Optimized conversation interface with Agent mode/normal mode switching, tool call process display, and comprehensive knowledge base management interface upgrade
⚡ Infrastructure Upgrade: Introduced MQ async task management, support for automatic database migration, and fast development mode

</details>

🔒 Security Notice

Important: Starting from v0.1.3, WeKnora includes login authentication functionality to enhance system security. For production deployments, we strongly recommend:

Deploy WeKnora services in internal/private network environments rather than public internet
Avoid exposing the service directly to public networks to prevent potential information leakage
Configure proper firewall rules and access controls for your deployment environment
Regularly update to the latest version for security patches and improvements

🏗️ Architecture

WeKnora employs a modern modular design to build a complete document understanding and retrieval pipeline. The system primarily includes document parsing, vector processing, retrieval engine, and large model inference as core modules, with each component being flexibly configurable and extendable.

🎯 Key Features

🤖 Agent Mode: Support for ReACT Agent mode that can use built-in tools to retrieve knowledge bases, MCP tools, and web search tools to access external services, providing comprehensive summary reports through multiple iterations and reflection
🔍 Precise Understanding: Structured content extraction from PDFs, Word documents, images and more into unified semantic views
🧠 Intelligent Reasoning: Leverages LLMs to understand document context and user intent for accurate Q&A and multi-turn conversations
📚 Multi-Type Knowledge Bases: Support for FAQ and document knowle

WeKnora

Install / Use