ragflow

What This Skill Does

The Ragflow skill serves as a robust bridge between the OpenClaw agent and self-hosted Ragflow instances. It provides a complete interface for Retrieval-Augmented Generation (RAG) workflows, enabling the agent to manage knowledge bases, upload documentation, and perform context-aware queries. By centralizing the management of datasets and parsing logic, the skill allows users to transform unstructured data—such as technical manuals, PDFs, or research articles—into interactive chat-ready knowledge bases without manual API configuration or complex middleware. It acts as an abstraction layer for the Ragflow REST API, allowing both CLI-based management and programmatic control via Node.js.

Installation

To integrate this skill into your environment, run the following command via the terminal: clawhub install openclaw/skills/skills/angusthefuzz/ragflow Once installed, ensure your .env file is configured with the RAGFLOW_URL and RAGFLOW_API_KEY associated with your Ragflow server instance. This is required for authentication.

Use Cases

1. Corporate Knowledge Bases: Automate the indexing of company wikis, HR policies, or legal documents to allow employees to query information naturally.
2. Technical Documentation: Maintain a searchable repository of software documentation where the AI can look up implementation details or API parameters.
3. Research Analysis: Upload long-form scientific papers or datasets to a specific dataset and perform cross-document queries to synthesize findings.

Example Prompts

1. "Create a new dataset named 'Project Alpha' and upload the 'requirements.md' file from my current directory."
2. "Search the 'Project Alpha' knowledge base for any information regarding the current authentication protocols."
3. "List all documents currently inside the 'General Knowledge' dataset and trigger a re-parse for the 'manual.pdf' document."

Tips & Limitations

• Parsing Latency: Remember that triggering a document upload does not immediately make the content searchable. Always call the triggerParsing function or check the status of the parsing job to ensure content is fully indexed.
• Authentication: Keep your RAGFLOW_API_KEY secure. Do not hardcode this in scripts; always use environment variables.
• File Limits: Ragflow performance depends on your server resources. For massive document libraries, consider increasing the server's dedicated hardware allocation to handle high-concurrency parsing tasks.
• Networking: Ensure your host server can reach the Ragflow API endpoint, as the skill relies on direct network communication with the remote instance.