Generate financial market forecasts with Kronos using time-series data, meaning observations recorded over time. An agent then adjusts the numerical forecast using recent news sentiment and reasoning, combining a quantitative baseline with a subjective assessment.
Market prediction skill using Kronos. Use when user needs finance market time-series forecasting or news-aware finance market adjustments.
Run score Not tested
See how to installReview the instructions the agent will receive before you install.
This skill utilizes the Kronos model (via KronosPredictorUtility) to perform time-series forecasting and adjust predictions based on news sentiment.
Workflow:
scripts/kronos_predictor.py (via KronosPredictorUtility) to generate the technical/quantitative forecast.references/PROMPTS.md to subjectively adjust the numbers based on latest news/logic.Key Tools:
KronosPredictorUtility.get_base_forecast(df, lookback, pred_len, news_text): Returns List[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)
This skill requires the Kronos model and an embedding model.
exports/models directory exists in the project root.kronos_news_v1.pt) in exports/models/.[!CAUTION] Model Security: This skill loads model weights from
exports/models. We useweights_only=Trueand only scan for thekronos_news_*.ptpattern. Ensure you only place trusted checkpoints in this directory.
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.torchtransformerssentence-transformerspandasnumpyscikit-learnNo reviews yet. Tell others what you thought.
No login needed. Share how it worked for you, ask a question, or note where it did not run.
No comments yet. Be the first to write one.
Retrieve the latest financial signals from DeepEar Lite as a formatted report. Review titles, summaries, sentiment indicators, confidence scores, and source links to understand reported market trends and factors affecting stock performance.
Turn investment theses and chains of financial market effects into Draw.io-compatible diagrams. The skill generates diagram XML and saves it in a viewable HTML file, helping readers follow complex relationships and reasoning.
Fetch current financial headlines and combine trending stories from multiple sources, including Weibo and WallstreetCN. You can also retrieve a summary of active Polymarket prediction markets to review alongside the news.
Turn financial analysis and investment signals into a structured report. Group related signals, draft sections, assemble and edit the report, and create chart configurations to support the written analysis when needed.
Search for financial information on the web or in a local news database. Use Jina, DDG, or Baidu, combine results from multiple engines, and assess whether cached results can answer a query before searching again.
Analyze financial text with local FinBERT or an LLM to identify positive, negative, or neutral sentiment. Return a numeric score and short reason, with support for batch processing unanalyzed news using FinBERT.