A professional-grade document parser for extracting and searching text within modern and legacy Microsoft Word documents without requiring Office installation.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install read-word
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 read-word 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 read-word?
The read-word skill is a versatile utility designed for Openclaw Skills users who need to programmatically access content within Microsoft Word files. It provides robust support for both modern .docx files (Word 2007+) and legacy .doc files (Word 97-2003) via OLE parsing.
Developed with a focus on cross-language compatibility, this skill features full Chinese language support, ensuring that multi-byte characters are handled correctly during extraction and conversion. By operating independently of the Microsoft Word application, it serves as a lightweight, high-performance solution for document analysis and automated data extraction workflows.
read-word Use Cases
- Automated extraction of text from large batches of Word documents for archival purposes.
- Searching for specific keywords across multiple .docx and .doc reports without opening each file manually.
- Converting legacy Word documents into clean, UTF-8 encoded text files for use in LLM training or data pipelines.
- Analyzing document statistics such as paragraph counts and character totals for project management.
How read-word Works
- The user provides a file path to a .docx or .doc file via the CLI or Python API.
- The skill automatically detects the file format to determine the appropriate parsing engine (python-docx for modern files or OLE parsing for legacy files).
- It extracts the raw text content while maintaining paragraph integrity and handling specific encodings like Chinese characters.
- Optionally, the tool filters content based on user-defined keywords or limits the number of paragraphs retrieved.
- The final output is either displayed in the terminal, returned as a Python list, or saved as a UTF-8 text file.
read-word Setup
Prerequisites
Ensure you have the necessary parsing libraries installed:
pip install python-docx olefile
Install Skill
Clone or copy the skill into your local Openclaw Skills directory:
# Copy to your Openclaw Skills directory
cp -r read-word ~/.openclaw/skills/
read-word Data Schema & Taxonomy
The skill processes document metadata and text content according to the following schema:
| Feature | Description | Output Format |
|---|---|---|
| Paragraph List | Array of all text blocks extracted from the document | Python List (Strings) |
| Keyword Matches | Paragraphs containing specific search terms with index headers | String with [Paragraph N] prefix |
| Document Info | Metadata including filename, size, and character count | Dictionary / JSON |
| Export File | Standardized plain text conversion | UTF-8 .txt file |
read-word Advanced Features
- Multi-format auto-detection: Seamlessly switches between modern XML-based parsing and legacy OLE binary parsing.
- High-performance keyword filtering: Search across massive documents using a list of multiple keywords simultaneously.
- Native Python integration: Can be imported directly into other Openclaw Skills to provide document-reading capabilities to larger AI agent workflows.
- CLI-driven analysis: Extract document statistics and metadata without generating any intermediate files.
SKILL.md
Loading
Related Openclaw Skills
Number One Cultural Insights
A bilingual expert skill exploring the philosophical, religious, and cultural significance of the number 1 across Eastern and Western traditions.
Stock Price Query
A powerful Python-based tool for fetching real-time market data across A-share, Hong Kong, and US markets using natural language queries.
Number Two Cultural Symbolism
A bilingual technical skill for exploring the philosophical, cultural, and modern symbolism of the number two.
minimax-use
A versatile AI skill providing web search, image understanding, and multi-model LLM chat capabilities via the MiniMax API.
ZeeLin Deep Research
ZeeLin Deep Research is an AI-driven professional research platform that transforms simple queries into comprehensive, multi-step expert reports and systematic analyses.
wsl-path-converter
A utility skill that bridges the gap between Windows and WSL by automatically translating file paths for Linux environment compatibility.
Featured*
