alphaear-sentiment

1# AlphaEar Sentiment Skill
2 
3## Overview
4 
5This skill provides sentiment analysis capabilities tailored for financial texts, supporting both FinBERT (local model) and LLM-based analysis modes.
6 
7## Capabilities
8 
9## Capabilities
10 
11### 1. Analyze Sentiment (FinBERT / Local)
12 
13Use `scripts/sentiment_tools.py` for high-speed, local sentiment analysis using FinBERT.
14 
15**Key Methods:**
16 
17-   `analyze_sentiment(text)`: Get sentiment score and label using localized FinBERT model.
18    -   **Returns**: `{'score': float, 'label': str, 'reason': str}`.
19    -   **Score Range**: -1.0 (Negative) to 1.0 (Positive).
20-   `batch_update_news_sentiment(source, limit)`: Batch process unanalyzed news in the database (FinBERT only).
21 
22### 2. Analyze Sentiment (LLM / Agentic)
23 
24For higher accuracy or reasoning capabilities, **YOU (the Agent)** should perform the analysis using the Prompt below, calling the LLM directly, and then update the database if necessary.
25 
26#### Sentiment Analysis Prompt
27 
28Use this prompt to analyze financial texts if the local tool is insufficient or if reasoning is required.
29 
30```markdown
31请分析以下金融/新闻文本的情绪极性。
32返回严格的 JSON 格式:
33{"score": <float: -1.0到1.0>, "label": "<positive/negative/neutral>", "reason": "<简短理由>"}
34 
35文本: {text}
36```
37 
38**Scoring Guide:**
39- **Positive (0.1 to 1.0)**: Optimistic news, profit growth, policy support, etc.
40- **Negative (-1.0 to -0.1)**: Losses, sanctions, price drops, pessimism.
41- **Neutral (-0.1 to 0.1)**: Factual reporting, sideways movement, ambiguous impact.
42…