sentiment-analysis

Quantify market sentiment into 0-100 scores with position and hedge recommendations.

Updated Apr 10, 2026
One-click install
npx skills add https://github.com/ebrahim-sani/trading-automation --skill sentiment-analysis-ebrahim-sani
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: sentiment-analysis
Source: https://github.com/ebrahim-sani/trading-automation/tree/main/vibe-trading/agent/src/skills/sentiment-analysis
Command: npx skills add https://github.com/ebrahim-sani/trading-automation --skill sentiment-analysis-ebrahim-sani

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Converts subjective market mood into quantitative, actionable signals by combining fear-and-greed indices, options put-call ratios, margin financing, northbound capital flows, and social media sentiment into a single, comparable framework for trading and risk decisions.

Core Features & Use Cases

  • Multi-dimension Sentiment Scoring: Standardizes each dimension to a 0–100 scale and constructs a weighted composite sentiment score for crypto, A-share, and major equity markets.
  • Signal Interpretation & Position Guidance: Maps composite and per-dimension scores to concrete position size recommendations, suggested hedges, and alert thresholds (e.g., inverse signals at extreme ranges).
  • Data-aware Thresholds & Cross-validation: Uses market-specific thresholds (A-share vs US vs crypto), cross-validates PCR with VIX or volatility, and recommends ignoring noisy single-day spikes in favor of weekly/monthly trends.
  • Use Case Example: Run the framework to detect an over-levered A-share market where financing monthly growth >10% and PCR <0.5 produce a composite greed score >80, then automatically recommend reducing equity exposure and buying protective options.

Quick Start

Use the sentiment-analysis skill to evaluate current crypto and A-share market sentiment, return a 0-100 composite score per market dimension, and provide position sizing and hedging recommendations.

Frequently Asked Questions about sentiment-analysis

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I quantify market sentiment for position sizing across multiple asset classes?

You can quantify market sentiment by standardizing fear-and-greed indices, put-call ratios, margin financing, northbound flows, and social media signals into a 0-100 composite score, which maps directly to concrete position sizing and hedging recommendations.

What is the best way to combine put-call ratio and VIX data for trend confirmation?

The best way to combine put-call ratio and VIX data is through cross-validation within a multi-dimension framework, filtering out noisy single-day spikes and focusing on weekly or monthly trends for reliable signal confirmation.

How do I calculate a composite sentiment score for A-share and crypto markets?

To calculate a composite sentiment score, standardize each market-specific dimension to a 0-100 scale using local thresholds, then apply a weighted construction to aggregate metrics like northbound flows for A-shares or social media signals for crypto.

Can I use margin financing growth rates to detect over-leveraged markets and generate alerts?

Yes, you can use margin financing growth rates to detect over-leveraged markets. For example, if A-share monthly financing growth exceeds 10% and PCR drops below 0.5, the composite greed score triggers threshold-based alerts.

Does multi-dimension sentiment scoring work for automated trading workflows?

Yes, multi-dimension sentiment scoring works for automated trading workflows by providing normalized dimension scores, threshold-based alerts, and explicit position and hedge recommendations suitable for research or automated execution.

When should I ignore social media sentiment signals in market analysis?

You should ignore social media sentiment signals when they manifest as noisy single-day spikes, instead relying on cross-validated weekly or monthly trends to confirm market direction and avoid false signals.