A multi-dimensional article evaluator that filters low-quality content and detects AI-generated text to save you valuable reading time.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install article-taster
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 article-taster 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 Article Taster?
Article Taster is an advanced content analysis tool designed to act as a digital sommelier for your reading material. By integrating with Openclaw Skills, it provides a comprehensive framework for evaluating the technical depth, structural clarity, and original value of any given text. It specifically addresses the modern challenge of content saturation by distinguishing between high-quality human writing and hollow, template-driven AI outputs.
The tool classifies content into specific genres like technical articles, emotional essays, or novels, applying specialized scoring weights to each. This ensures that a technical guide is judged on practicality while a novel is evaluated on narrative structure and emotional resonance, all while maintaining a spoiler-free experience for the reader.
Article Taster Use Cases
- Filtering high-signal technical articles from low-quality web scrapes.
- Detecting AI-generated garbage content in automated content feeds.
- Verifying the originality of classical poetry and literature using specialized exemption rules.
- Prioritizing reading lists by batch-analyzing document directories for quality scores.
- Evaluating the emotional depth and writing style of creative essays before committing time to read.
How Article Taster Works
- Genre Classification: The skill identifies the article type to determine which specialized analysis module (Tech, Creative, Novel, or Generic) to trigger.
- Specialized Analysis: Content is processed through specific analyzers to measure industry-relevant metrics like technical depth or narrative arc.
- AI Detection Engine: The system performs statistical analysis on perplexity, vocabulary richness, and paragraph uniformity to identify AI-generated patterns.
- Originality Validation: It applies exemption logic for classical literature and poetry to ensure high-quality traditional works aren't misclassified as AI.
- Report Synthesis: A final score and grade (A+ to D) are generated along with actionable reading advice and time estimations.
Article Taster Setup
To get started with this skill, ensure you have Python 3.10 or higher installed. Follow these steps for installation:
# Install required Python dependencies
pip install jieba scikit-learn
# Clone the repository and navigate to the directory
cd article-taster
# Run a test analysis on a text file
python -m article_taster analyze --file sample_article.txt
Note: You can optionally configure OpenAI or Anthropic API keys in the config directory to enable LLM-assisted depth analysis within the Openclaw Skills environment.
Article Taster Data Schema & Taxonomy
The skill organizes its findings into a structured JSON report and a human-readable Markdown summary. Below is the metadata taxonomy used:
| Data Field | Description |
|---|---|
| overall_score | A weighted final score from 0-100 |
| ai_probability | Probability that the text was generated by an LLM |
| ai_flavor_level | Qualitative assessment of AI patterns (e.g., Lightly Suspected, Garbage Content) |
| dimension_scores | Individual scores for technical depth, structure, practicality, and originality |
| reading_advice | Target audience identification and key benefit summaries |
| exemption_type | Flags for classical poetry or classic literature to prevent false positives |
Article Taster Advanced Features
- Batch Processing: Analyze entire directories of articles in one command for massive content curation.
- Exemption Engine: Custom rules to prevent high-quality classical or unique literary styles from being flagged as AI.
- Spoiler-Free Analysis: Specialized logic for novels that evaluates structure and pacing without revealing plot points.
- Customizable Weights: Adjust the scoring importance of technical depth versus readability via JSON configuration files.
- Multi-Agent Integration: Designed to work within Openclaw Skills to pass quality reports to other automated research agents.
SKILL.md
Loading
Related Openclaw Skills
Movi Review-First Bundle
A safety-oriented skill for AI agents to manage local MCP servers with a review-first approach to data mutation and batch analysis.
Switchyard Runtime Diagnostics
A specialized skill for installing and utilizing Switchyard's read-only MCP runtime diagnostics surface for precise AI agent debugging.
Shopflow Read-only Packet
A technical integration packet designed to teach AI agents how to interface with Shopflow as a read-only surface using MCP servers.
Polymarket Value Trader
A disciplined prediction market trading skill for AI agents that utilizes value-based analysis to trade on Simmer and Polymarket venues.
AI Collaborative Writing Methodology
A sophisticated multi-agent methodology designed to amplify human judgment for high-quality document drafting and reporting.
ProofTrail MCP
A governed browser-evidence and recovery layer that enables AI agents to perform verifiable web automation using stdio-based MCP tools.
Featured*
