alphaear-predictor
npx machina-cli add skill RKiding/Awesome-finance-skills/alphaear-predictor --openclawAlphaEar Predictor Skill
Overview
This skill utilizes the Kronos model (via KronosPredictorUtility) to perform time-series forecasting and adjust predictions based on news sentiment.
Capabilities
1. Forecast Market Trends
1. Forecast Market Trends
Workflow:
- Generate Base Forecast: Use
scripts/kronos_predictor.py(viaKronosPredictorUtility) to generate the technical/quantitative forecast. - Adjust Forecast (Agentic): Use the Forecast Adjustment Prompt in
references/PROMPTS.mdto subjectively adjust the numbers based on latest news/logic.
Key Tools:
KronosPredictorUtility.get_base_forecast(df, lookback, pred_len, news_text): ReturnsList[KLinePoint].
Example Usage (Python):
from scripts.utils.kronos_predictor import KronosPredictorUtility
from scripts.utils.database_manager import DatabaseManager
db = DatabaseManager()
predictor = KronosPredictorUtility()
# Forecast
forecast = predictor.predict("600519", horizon="7d")
print(forecast)
Configuration
This skill requires the Kronos model and an embedding model.
-
Kronos Model:
- Ensure
exports/modelsdirectory exists in the project root. - Place trained news projector weights (e.g.,
kronos_news_v1.pt) inexports/models/. - Or depend on the base model (automatically downloaded).
- Ensure
-
Environment Variables:
EMBEDDING_MODEL: Path or name of the embedding model (default:sentence-transformers/all-MiniLM-L6-v2).KRONOS_MODEL_PATH: Optional path to override model loading.
Dependencies
torchtransformerssentence-transformerspandasnumpyscikit-learn
Source
git clone https://github.com/RKiding/Awesome-finance-skills/blob/main/skills/alphaear-predictor/SKILL.mdView on GitHub Overview
AlphaEar Predictor uses the Kronos model via KronosPredictorUtility to deliver time-series forecasts for financial markets. It also adjusts forecasts based on the latest news sentiment using the Forecast Adjustment Prompt for a more context-aware projection.
How This Skill Works
It generates a base forecast by calling KronosPredictorUtility.get_base_forecast with your data, lookback, pred_len, and news_text. It then applies a subjective adjustment using the Forecast Adjustment Prompt from references/PROMPTS.md to reflect current news and logic. The resulting forecast is a List of KLinePoint values that can be used for short-to-mid horizon planning. Example usage shows constructing KronosPredictorUtility and calling predictor.predict('600519', horizon='7d').
When to Use It
- When you need time-series forecasts for equities, commodities, or FX with a 1- to 2-week horizon
- When you want forecast adjustments that reflect the latest news or market logic
- When you need a reproducible base forecast plus subjective tuning for events (earnings, macro data)
- When your workflow uses Kronos-backed predictions integrated into a Python pipeline
- When you require embedding-powered model loading and configurable Kronos path
Quick Start
- Step 1: Ensure Kronos model and embedding model are available (exports/models and embeddings).
- Step 2: Prepare your data and call the predictor, e.g. predictor.predict("ticker", horizon="7d"), with news_text if required.
- Step 3: Apply the Forecast Adjustment Prompt to align the base forecast with the latest news, then use or store the final forecast.
Best Practices
- Ensure Kronos model and embedding model are installed and accessible (exports/models, correct embeddings).
- Choose lookback and pred_len that match your forecasting horizon and data frequency.
- Supply up-to-date news_text so the adjustment prompt reflects current sentiment.
- Validate forecasts with backtesting to quantify accuracy before production use.
- Log and version forecast outputs and model configuration for traceability.
Example Use Cases
- Forecast a stock ticker like 600519 over a 7-day horizon using base Kronos forecast plus news adjustment.
- Adjust forecasts for a commodity after a major shipment or political event.
- Forecast a currency pair trend and tune it with latest macro news sentiment.
- Use AlphaEar to drive short-term trading signals around earnings announcements.
- Incorporate forecasts into a portfolio rebalancing plan with horizon-aligned predictions.
