Extract BoW, TF-IDF, FinBERT embeddings, and sentiment scores from financial text.
npx skills add https://github.com/IgorGanapolsky/trading --skill text-analyzer
Please help me install this Agent Skill. Skill: Text Analyzer Source: https://github.com/IgorGanapolsky/trading/tree/main/.claude/skills/text_analyzer Command: npx skills add https://github.com/IgorGanapolsky/trading --skill text-analyzer
This Skill extracts numerical features from financial text (news, earnings reports, social media) to feed ML-based trading decisions, enabling more informed decisions with less manual effort.
Run the Text Analyzer on a sample piece of financial text to produce a feature vector and sentiment signals, then feed it into your ML model.
High-intent search queries and answers about installing and using this skill.
Text Analyzer processes financial text—headlines, articles, earnings reports, social media—into ML-ready features: bag-of-words, TF-IDF, FinBERT embeddings, and sentiment scores. Feed the output directly into your trading decision model.
Yes. Text Analyzer supports batch processing of multiple texts to generate BoW, TF-IDF, and FinBERT embeddings with bullish, bearish, and uncertainty signals in a single run.
Text Analyzer applies domain-specific sentiment scoring tuned for financial context, using FinBERT embeddings to detect market mood and produce interpretable bullish/bearish/uncertainty classifications.
BoW captures raw word frequency for quick baseline signals; TF-IDF weights term importance to isolate meaningful keywords. Text Analyzer generates both to let you choose the feature richness your model needs.
Yes. Text Analyzer outputs 768-dimensional FinBERT vectors that capture semantic context and integrate directly into downstream ML models for trading decisions.
dgriffin831Normalize whitespace, strip HTML tags, and handle Markdown in raw text.
santoosaraujoExtract financial sentiment signals from Twitter, Telegram, Discord, and Reddit.
lucifertrjClean, tokenize, and normalize English text with NLTK for NLP pipelines.
AJBcodingBuild ML preprocessing pipelines transforming raw tabular data into model-ready features.
