U2 Doc Parser

What This Skill Does

The U2 Doc Parser is a powerful document conversion utility provided by aaiccee that integrates the UniDoc API into your OpenClaw workflow. It is designed to take complex document formats—including PDF, DOC, DOCX, and various image formats like PNG and JPG—and convert them into clean, structured Markdown or JSON. By leveraging a cloud-based backend, the skill handles the heavy lifting of OCR and text extraction, allowing you to easily transform static documents into machine-readable data or editable text files. The skill supports both synchronous processing, where the agent waits for the file to be processed, and asynchronous processing, which uses a polling mechanism to check for completion, making it highly robust for handling various file sizes and network conditions.

Installation

To integrate this utility into your environment, use the OpenClaw package manager. Open your terminal and run the following command: clawhub install openclaw/skills/skills/aaiccee/u2-doc-parser Ensure that you have an active internet connection, as the installer will pull the necessary scripts from the openclaw/skills repository to enable the functionality.

Use Cases

This skill is perfect for developers and data analysts who need to automate the ingestion of legacy documents. For example, you can use it to convert scanned PDF invoices into JSON objects for automated bookkeeping, or transform dense technical manuals into Markdown files to be included in a static site generator or a knowledge base. It is also an excellent tool for converting academic papers or image-based notes into accessible formats that can be easily parsed or summarized by other AI agents within your ecosystem.

Example Prompts

1. "Use the U2 Doc Parser to convert the file 'manual.pdf' in my downloads folder to a Markdown file named 'manual.md'."
2. "Extract the data from 'report.docx' and save it as a JSON file, using the async mode to ensure it handles the conversion process reliably."
3. "Convert my scanned image 'handwritten_notes.jpg' to Markdown and print the output directly to my terminal so I can verify the OCR quality."

Tips & Limitations

• Privacy First: Always remember that this skill uploads your files to https://unidoc.uat.hivoice.cn. Never process sensitive or confidential data.
• Performance: For large documents, always prefer --mode async to prevent timeouts.
• Monitoring: If you encounter issues, refer to references/unidoc-notes.md to troubleshoot API connectivity.
• Formatting: Default to Markdown for reading convenience and JSON for data integration workflows.