Agent Skills
› NeverSight/learn-skills.dev
› xai-sentiment
xai-sentiment
GitHub基于Grok的X/Twitter实时情感分析工具。用于追踪市场情绪、公众舆论或特定话题的社会倾向，输出包含情感标签、置信度及详细统计的结构化JSON数据。
Trigger Scenarios
分析社交媒体上的公众情绪
追踪金融市场情绪波动
评估特定品牌或事件的网络舆论
Install
npx skills add NeverSight/learn-skills.dev --skill xai-sentiment -g -y
SKILL.md
Frontmatter
{
"name": "xai-sentiment",
"version": "1.0.0",
"description": "Real-time sentiment analysis on Twitter\/X using Grok. Use when analyzing social sentiment, tracking market mood, or measuring public opinion on topics."
}
xAI Sentiment Analysis
Real-time sentiment analysis on Twitter/X content using Grok's native integration and built-in NLP capabilities.
Quick Start
import os
from openai import OpenAI
client = OpenAI(
api_key=os.getenv("XAI_API_KEY"),
base_url="https://api.x.ai/v1"
)
def analyze_sentiment(topic: str) -> dict:
"""Analyze sentiment for a topic on X."""
response = client.chat.completions.create(
model="grok-4-1-fast",
messages=[{
"role": "user",
"content": f"""Analyze sentiment on X for: {topic}
Search recent posts and return JSON:
{{
"topic": "{topic}",
"sentiment": "bullish" | "bearish" | "neutral",
"score": -1.0 to 1.0,
"confidence": 0.0 to 1.0,
"positive_percent": 0-100,
"negative_percent": 0-100,
"neutral_percent": 0-100,
"sample_size": number,
"key_themes": ["theme1", "theme2"],
"notable_posts": [
{{"author": "@handle", "summary": "...", "sentiment": "..."}}
]
}}"""
}]
)
return response.choices[0].message.content
# Example
result = analyze_sentiment("$AAPL stock")
print(result)
Sentiment Score Scale
| Score Range | Label | Description |
|---|---|---|
| 0.6 to 1.0 | Very Bullish | Strong positive sentiment |
| 0.2 to 0.6 | Bullish | Moderately positive |
| -0.2 to 0.2 | Neutral | Mixed or balanced |
| -0.6 to -0.2 | Bearish | Moderately negative |
| -1.0 to -0.6 | Very Bearish | Strong negative sentiment |
Sentiment Analysis Functions
Basic Sentiment
def get_basic_sentiment(query: str) -> dict:
"""Get simple sentiment score."""
response = client.chat.completions.create(
model="grok-4-1-fast",
messages=[{
"role": "user",
"content": f"""Search X for "{query}" and analyze sentiment.
Return only JSON:
{{"positive": 0-100, "neutral": 0-100, "negative": 0-100, "score": -1 to 1}}"""
}]
)
return response.choices[0].message.content
Detailed Sentiment Analysis
def get_detailed_sentiment(topic: str, timeframe: str = "24h") -> dict:
"""Get comprehensive sentiment analysis."""
response = client.chat.completions.create(
model="grok-4-1-fast",
messages=[{
"role": "user",
"content": f"""Perform detailed sentiment analysis on X for: {topic}
Timeframe: Last {timeframe}
Return JSON:
{{
"overall_sentiment": {{
"label": "bullish/bearish/neutral",
"score": -1 to 1,
"confidence": 0 to 1
}},
"breakdown": {{
"positive": {{"percent": 0-100, "count": n}},
"negative": {{"percent": 0-100, "count": n}},
"neutral": {{"percent": 0-100, "count": n}}
}},
"themes": [
{{"theme": "...", "sentiment": "...", "frequency": n}}
],
"influencer_sentiment": [
{{"handle": "@...", "sentiment": "...", "followers": n}}
],
"trending_hashtags": ["#tag1", "#tag2"],
"sentiment_drivers": {{
"positive_factors": ["..."],
"negative_factors": ["..."]
}}
}}"""
}]
)
return response.choices[0].message.content
Comparative Sentiment
def compare_sentiment(topics: list) -> dict:
"""Compare sentiment across multiple topics."""
topics_str = ", ".join(topics)
response = client.chat.completions.create(
model="grok-4-1-fast",
messages=[{
"role": "user",
"content": f"""Compare X sentiment for: {topics_str}
Return JSON:
{{
"comparison": [
{{
"topic": "...",
"sentiment_score": -1 to 1,
"volume": "high/medium/low",
"trend": "improving/declining/stable"
}}
],
"winner": "most positive topic",
"loser": "most negative topic",
"insights": ["..."]
}}"""
}]
)
return response.choices[0].message.content
Sentiment Over Time
def sentiment_timeline(topic: str, periods: list) -> dict:
"""Track sentiment changes over time."""
response = client.chat.completions.create(
model="grok-4-1-fast",
messages=[{
"role": "user",
"content": f"""Analyze how sentiment for "{topic}" has changed on X.
Return JSON with sentiment for different time periods:
{{
"topic": "{topic}",
"timeline": [
{{"period": "last hour", "score": -1 to 1}},
{{"period": "last 24 hours", "score": -1 to 1}},
{{"period": "last week", "score": -1 to 1}}
],
"trend": "improving/declining/stable",
"momentum": "accelerating/decelerating/steady",
"key_events": [
{{"time": "...", "event": "...", "impact": "..."}}
]
}}"""
}]
)
return response.choices[0].message.content
Financial Sentiment Analysis
Stock Sentiment
def stock_sentiment(ticker: str) -> dict:
"""Analyze stock sentiment with financial context."""
response = client.chat.completions.create(
model="grok-4-1-fast",
messages=[{
"role": "user",
"content": f"""Analyze X sentiment for ${ticker} stock.
Return JSON:
{{
"ticker": "{ticker}",
"sentiment": {{
"overall": "bullish/bearish/neutral",
"score": -1 to 1,
"strength": "strong/moderate/weak"
}},
"trading_signals": {{
"retail_sentiment": "...",
"smart_money_mentions": "...",
"options_chatter": "..."
}},
"catalysts_mentioned": ["earnings", "product", "macro"],
"price_predictions": {{
"bullish_targets": [...],
"bearish_targets": [...]
}},
"risk_factors": ["..."],
"recommendation": "..."
}}"""
}]
)
return response.choices[0].message.content
Crypto Sentiment
def crypto_sentiment(coin: str) -> dict:
"""Analyze cryptocurrency sentiment."""
response = client.chat.completions.create(
model="grok-4-1-fast",
messages=[{
"role": "user",
"content": f"""Analyze X sentiment for {coin} cryptocurrency.
Return JSON:
{{
"coin": "{coin}",
"sentiment_score": -1 to 1,
"fear_greed_indicator": "extreme fear/fear/neutral/greed/extreme greed",
"whale_mentions": "high/medium/low",
"influencer_sentiment": [...],
"trending_narratives": [...],
"fud_detection": {{
"level": "high/medium/low",
"sources": [...]
}},
"fomo_detection": {{
"level": "high/medium/low",
"triggers": [...]
}}
}}"""
}]
)
return response.choices[0].message.content
Batch Sentiment Analysis
def batch_sentiment(topics: list) -> list:
"""Analyze sentiment for multiple topics efficiently."""
topics_formatted = "\n".join([f"- {t}" for t in topics])
response = client.chat.completions.create(
model="grok-4-1-fast",
messages=[{
"role": "user",
"content": f"""Analyze X sentiment for each:
{topics_formatted}
Return JSON array:
[
{{"topic": "...", "score": -1 to 1, "label": "...", "volume": "high/med/low"}}
]"""
}]
)
return response.choices[0].message.content
Sentiment Alerts
def check_sentiment_alert(topic: str, threshold: float = 0.5) -> dict:
"""Check if sentiment has crossed alert threshold."""
response = client.chat.completions.create(
model="grok-4-1-fast",
messages=[{
"role": "user",
"content": f"""Check X sentiment for {topic}.
Alert threshold: {threshold} (positive) or {-threshold} (negative)
Return JSON:
{{
"topic": "{topic}",
"current_score": -1 to 1,
"alert_triggered": true/false,
"alert_type": "bullish/bearish/none",
"reason": "...",
"recommended_action": "..."
}}"""
}]
)
return response.choices[0].message.content
Best Practices
1. Request Confidence Scores
Always ask for confidence levels to gauge reliability.
2. Specify Sample Size
Request the number of posts analyzed for context.
3. Account for Sarcasm
Grok may misinterpret sarcasm - request explicit sarcasm detection:
"Note: Flag any potentially sarcastic posts separately"
4. Filter by Quality
Combine with handle filtering for higher-quality signals:
"Focus on verified accounts and accounts with >10k followers"
5. Combine with Price Data
Sentiment is most valuable when combined with price action.
Limitations
| Limitation | Mitigation |
|---|---|
| Sarcasm detection | Request explicit flagging |
| Bot content | Ask to filter suspicious patterns |
| Historical accuracy | Focus on recent data |
| Sample size | Request volume metrics |
Related Skills
xai-x-search- X search functionalityxai-stock-sentiment- Stock-specific analysisxai-crypto-sentiment- Crypto analysisxai-financial-integration- Combine with price data
References
Version History
- e0220ca Current 2026-07-05 23:15
