A specialized tool that parses local Word documents and PDF files into structured, report-first knowledge artifacts without directly altering core memory files.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install docx-pdf-knowledge-parser
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 docx-pdf-knowledge-parser 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 Docx PDF Knowledge Parser?
The Docx PDF Knowledge Parser is an automated utility engineered to turn raw, local, or uploaded .docx and .pdf files into clean, structured, and highly reviewable knowledge outputs. Instead of immediately writing to your core database or modifying your main memory files directly, this skill prioritizes a report-first philosophy. It produces structured intermediate reports, candidate files, and logs that allow for human verification and validation.
By leveraging this tool among your Openclaw Skills, you can prevent AI agents from hallucinating or claiming they have learned document contents without executing successful text extraction. It provides developers and AI agents with a deterministic, safe, and highly structured framework to digest batch attachments while preserving details about low-confidence or failed files.
Docx PDF Knowledge Parser Use Cases
- Processing local batches of mixed .docx and .pdf files stored within a project directory.
- Generating reviewable knowledge artifacts before updating the core memory or knowledge base of an AI agent.
- Auditing large archives of documentation to identify unparseable scans, corrupted files, or unsupported formats.
- Extracting structured topic summaries and metadata from technical reports and manuals safely.
How Docx PDF Knowledge Parser Works
- Resolve Input Type: The skill determines whether the provided path is a single file or a directory containing multiple files.
- Create Batch Record: It establishes a unique batch_id and records the starting timestamp.
- Build Manifest: It indexes the discovered files, logging their name, path, type, and routing decision.
- Attempt Extraction: It routes .docx files to the Docx parser and .pdf files to the PDF parser to extract the underlying text.
- Produce Structured Outputs: Successfully extracted content is logged to a success registry, while failed files are routed to a failure log with detailed error contexts.
- Summarize Batch: It automatically compiles an ingestion report and a candidate memory file for human-in-the-loop review.
Docx PDF Knowledge Parser Setup
Prerequisites
Make sure you have python and the necessary document processing libraries installed in your environment.
Running the Parser
You can run the batch process locally using the included test runner:
python run.py --input /path/to/your/documents
Docx PDF Knowledge Parser Data Schema & Taxonomy
The skill organizes processed data into distinct, structured file types to maintain separation of concerns:
Output Files Schema
| File Name | Format | Purpose |
|---|---|---|
kb-items.jsonl |
JSON Lines | Contains successfully extracted knowledge items, topics, content types, and confidence scores. |
failed-items.jsonl |
JSON Lines | Records every failed document alongside the failure reason, error details, and suggested actions. |
MEMORY.candidate.md |
Markdown | A draft memory file featuring batch headers, grouped knowledge summaries, and source references. |
ingest-report.md |
Markdown | A human-readable execution summary containing file routing counts, risk analysis, and next steps. |
Docx PDF Knowledge Parser Advanced Features
- Robust Failure Routing: Automatically separates unsupported files (like .pptx, images, or archives) into manual-review tracks instead of breaking execution.
- Report-First Philosophy: Ensures safety by writing to a candidate memory file (
MEMORY.candidate.md) rather than auto-writing directly to production memory. - Uncertainty Preservation: Prevents AI hallucinations by recording explicit error logs and refusing to invent text when extraction confidence is low.
SKILL.md
Loading
Related Openclaw Skills
Social Content Pillars
A strategic content framework generator that develops 5 core pillars and 50+ high-performance post ideas for creators and brands.
Apple Store Deals
A specialized commerce agent skill designed to track, compare, and rank the best prices for Apple products across authorized resellers and refurbished outlets.
flow-hugo-publisher
A managed publishing workflow for Hugo sites that automates environment setup, Git version control, and GitHub Actions deployment.
Sandbox Reference Tool
A comprehensive SysOps reference tool for managing, troubleshooting, and optimizing sandbox environments through a streamlined CLI.
imglink
A seamless utility for generating AI-powered images through customizable URL parameters for instant embedding in web and document projects.
Social Bio Optimizer
A specialized AI skill designed to craft high-converting, platform-specific social media bios that turn profile visitors into engaged followers.
Featured*
