Collect and clean Arabic web content into JSONL training data.
npx skills add https://github.com/Moshe-ship/mkhlab --skill safha
Please help me install this Agent Skill. Skill: safha Source: https://github.com/Moshe-ship/mkhlab/tree/main/hermes-skills/safha Command: npx skills add https://github.com/Moshe-ship/mkhlab --skill safha
This tool helps gather and clean Arabic content from the web to prepare training data, saving time and improving data quality.
Scrape a URL and save the collected Arabic content as JSONL.
High-intent search queries and answers about installing and using this skill.
To build Arabic NLP training data, you can crawl websites and extract text while handling Arabic-specific normalization. This tool collects raw content and cleans it into structured JSONL files suitable for NLP datasets.
Arabic dialect detection classifies scraped web text by regional dialect. This tool tags cleaned content with dialect labels, enabling the creation of dialect-aware corpora for specialized NLP applications.
Cleaning Arabic text for machine learning involves normalizing dialect, removing diacritics, and pruning URLs. This process standardizes scraped web content into high-quality training data.
Yes, you can use a simple CLI to crawl and clean Arabic datasets. This tool supports straightforward command-line operations to scrape URLs and save the collected Arabic content directly as JSONL files.
The best way to prepare Arabic content moderation datasets is to scrape web content and apply text normalization. This tool cleans collected data and tags dialects, providing structured data for moderation model training.
vibewatchFetch URLs and extract readable text from HTML pages and PDFs.
Moshe-shipDetect encoding problems, hidden content, and dialect mixing in Arabic JSONL datasets.
hochoa13Train custom BPE, WordPiece, and Unigram tokenizers on text corpora.
jackquelinunpredictable827Detect sentiment and toxicity scores in Arabic text across five dialects.
