A professional knowledge base assistant featuring semantic search, multimodal document understanding, and Retrieval-Augmented Generation (RAG) capabilities.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install knowledge-chat
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install knowledge-chat using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Knowledge Chat Assistant?
Knowledge Chat is a robust skill designed for the Openclaw Skills ecosystem, serving as a specialized bridge between large language models and external data repositories. It leverages advanced Retrieval-Augmented Generation (RAG) to provide context-aware responses based on private or enterprise knowledge bases. By integrating semantic search rather than simple keyword matching, it ensures higher precision and relevance in information retrieval.
Built with a modern tech stack including Next.js 14 and Dashscope API, this skill supports multimodal inputs, allowing users to upload images and complex documents like PDFs for instant analysis. It provides a polished user experience with real-time thinking indicators and comprehensive Markdown rendering, making it an essential component for any developer building intelligent, data-driven Openclaw Skills.
Knowledge Chat Assistant Use Cases
- Enterprise knowledge base integration for internal documentation access.
- Intelligent Q&A systems for automated customer support.
- Advanced document retrieval assistants for research and academic purposes.
- RAG-driven application development requiring high-precision data fetching.
- Multimodal content analysis for processing images and attachments in chat.
How Knowledge Chat Assistant Works
- The user initiates a query through the Openclaw Skills interface or dedicated chat endpoint.
- The system triggers a semantic search across the connected knowledge base to identify relevant document chunks.
- If attachments (PDFs, images) are provided, the skill parses them using dedicated processors like pdf-parse or vision models.
- Contextual data is fed into the LLM (qwen-plus/qwen-vl-max) along with the user's prompt.
- The AI generates a detailed response, including citations and source references, while providing real-time feedback through a thinking progress bar.
- Follow-up questions are automatically suggested to guide the user toward deeper exploration of the topic.
Knowledge Chat Assistant Setup
To deploy this skill within your environment, follow these steps:
- Deploy the Next.js project to your preferred server or cloud provider.
- Configure your environment variables in a
.env.localfile:
DASHSCOPE_API_KEY=your_api_key_here
DATABASE_URL=your_sqlite_path
- Install dependencies and build the project:
npm install
npm run build
- Start the service using a process manager like PM2:
pm2 start npm --name "knowledge-chat" -- start
- Access the dashboard at
/dashboard/chatto begin managing your knowledge base and Openclaw Skills integration.
Knowledge Chat Assistant Data Schema & Taxonomy
Knowledge Chat organizes data using a structured SQLite schema to ensure fast retrieval and session management:
| Table Name | Description |
|---|---|
chat_sessions |
Stores metadata for individual chat conversations. |
chat_messages |
Contains the history of messages, including system, user, and AI roles. |
chat_files |
Tracks uploaded attachments, images, and processed document content. |
knowledge_base |
Manages configurations and API endpoints for external data sources. |
knowledge_files |
Stores vector indices and metadata for documents within the knowledge base. |
Knowledge Chat Assistant Advanced Features
- Manual vector index construction for optimized RAG performance.
- Multimodal understanding supporting JPG, PNG, GIF, and WEBP formats.
- Advanced PDF parsing with specialized extraction logic for complex layouts.
- Real-time ThinkingIndicator UI component for improved user feedback.
- Automated follow-up question generation for interactive discovery.
- Seamless integration with Dashscope API batch processing (batch size 10) for efficiency.
SKILL.md
Loading
Related Openclaw Skills
Classical Chinese Token Compression
A sophisticated token compression skill that uses Classical Chinese grammar to dramatically reduce AI output length while preserving technical accuracy.
hbdc: State Grid Bidding Document Analysis Skill
A professional automation tool designed to parse State Grid bidding documents and extract structured project data into Excel.
Amazon Inventory Management
A specialized AI agent skill for optimizing FBA inventory planning, restock cycles, and IPI score recovery across global Amazon marketplaces.
Olostep Web Data API
Olostep is a comprehensive Web Data API that enables AI agents to scrape, crawl, and search the live internet for real-time information retrieval.
Classical Chinese Compression
A specialized skill that utilizes the high information density of Classical Chinese to compress AI outputs and save significant token costs.
Pilot Network Map
A visualization utility for mapping Pilot Protocol network structures, trust relationships, and connection performance.
Featured*
