A local-first reading and distillation engine designed to transform long-form text and documents into actionable insights without cloud dependencies.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install reader
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 reader 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 Reader?
Reader is a powerful, privacy-centric distillation engine built to help developers and researchers process information faster. As a core component of the Openclaw Skills ecosystem, it operates entirely locally, ensuring that sensitive documents, reports, and notes never leave your machine. It specializes in normalizing diverse text formats—including Markdown, HTML, and JSON—to provide clear, concise briefs.
By prioritizing clarity over verbosity, Reader goes beyond simple summarization. It is designed to find the signal in the noise, specifically targeting decisions, action items, and open questions buried within dense text. This makes it an essential tool for anyone needing to maintain a low-friction, high-context workflow.
Reader Use Cases
- Summarizing lengthy technical documents or pasted logs into executive summaries.
- Extracting specific action items and unanswered questions from meeting notes or reports.
- Performing side-by-side comparisons between document drafts to identify key changes.
- Generating concise reading briefs to quickly get up to speed on new topics.
- Processing local JSON or HTML data into human-readable plain text formats.
How Reader Works
- Initialization: The skill sets up a dedicated local directory structure to track reading history and metadata.
- File Normalization: Reader ingests local text-based files (TXT, MD, HTML, JSON) and normalizes them for consistent analysis.
- Analysis Engine: Depending on the workflow selected, the skill runs specific Python scripts to parse the content for summaries, briefs, or comparisons.
- Insight Extraction: The engine identifies linguistic patterns to isolate key points, decisions, and required actions.
- Local Persistence: Results and metadata are logged to a local history file, providing a transparent audit trail of processed documents.
Reader Setup
To integrate Reader into your Openclaw Skills environment, ensure you have Python 3 installed. No external dependencies or API keys are required.
First, initialize the local storage:
python3 scripts/init_storage.py
To read a specific file:
python3 scripts/read_text.py --file path/to/your/document.txt
To generate an executive summary:
python3 scripts/summarize.py --file path/to/file --style executive
Reader Data Schema & Taxonomy
Reader maintains a strict local-only data policy. All information is organized within the following structure:
| Component | Path | Description |
|---|---|---|
| History Log | ~/.openclaw/workspace/memory/reader/history.json |
Stores a JSON array of past reading tasks and document metadata. |
| Workspace | ~/.openclaw/workspace/ |
The root directory for all processed text assets and temporary buffers. |
No data is ever synchronized to external cloud providers, maintaining total privacy for your Openclaw Skills workspace.
Reader Advanced Features
- Custom Summary Styles: Support for multiple output styles including executive, technical, and bulleted briefs.
- Multi-Document Comparison: High-fidelity diffing between two separate text files to track evolving reports or drafts.
- Format Agnostic Parsing: Seamlessly handles various text-like file extensions including JSON and HTML without manual conversion.
- Zero-Dependency Runtime: Built using standard Python libraries to ensure maximum portability and security across different environments.
- Action-Oriented Extraction: Specifically tuned to detect and list 'open questions' and 'decisions' to facilitate project management.
SKILL.md
Loading
Related Openclaw Skills
backupclaw
A robust configuration management utility for OpenClaw that automates versioned backups with change detection and restoration capabilities.
tomtat-video
A high-performance utility to extract YouTube transcripts and generate structured AI summaries automatically.
Giggle Image Generation
A professional AI image generation skill that connects to giggle.pro to create high-quality visuals using multiple state-of-the-art models.
WeChat Article Search API
A professional tool for extracting full-text WeChat articles and structured metadata via the BrowserAct API without manual scraping or CAPTCHA hurdles.
Giggle Generation Scripts
A specialized AI engine for generating high-conflict, subtext-driven script content with a focus on black humor and narrative reversals.
MaoMing City Knowledge Base
A comprehensive data repository and research skill covering MaoMing's industrial economy, agricultural heritage, and coastal tourism.
Featured*
