Score news and macro updates with an NLP model into event CSVs.
npx skills add https://github.com/HKUDS/Vibe-Trading --skill event-driven-hkuds
Please help me install this Agent Skill. Skill: event-driven Source: https://github.com/HKUDS/Vibe-Trading/tree/main/agent/src/skills/event-driven Command: npx skills add https://github.com/HKUDS/Vibe-Trading --skill event-driven-hkuds
Trading teams struggle to translate scattered news, policy updates, and sentiment swings into consistent signals, leaving technical models blind to real-world catalysts. This skill centralizes event scoring, enforces an event CSV schema, and feeds a time-decayed sentiment view so that the agent can account for news-based conviction alongside technical indicators.
signal_engine.py reads the CSV, filters by freshness and score threshold, applies exponential decay, and clips the output to [-1, 1] before merging with technical signals.Use the event-driven skill to score today's news, append each sentiment row to the event CSV, and run signal_engine.py to blend the results with your technical signal.
High-intent search queries and answers about installing and using this skill.
To generate event-driven trading signals, you score news announcements with an NLP model into sentiment rows, append them to a standardized event CSV, and apply exponential time-decay to produce a clipped [-1, 1] signal.
Time-decay aggregation applies exponential decay to sentiment scores filtered by freshness and threshold, clipping the weighted output to [-1, 1] before merging with technical signals to prevent look-ahead bias.
The best way to combine news sentiment with technical signals is appending each news event to a standardized CSV, running decay processing via pandas and numpy, and merging the clipped output with your trend-following data.
Yes, you can use pandas and numpy to process event-driven signals for macro assets by reading the event CSV, applying configurable aggregation parameters, and merging the time-decayed sentiment with technical indicators.
Yes, you need a consistent event CSV schema to store scored news rows so the signal engine can filter by freshness, apply exponential decay, and merge the weighted output with technical signals.
Standardized LLM scoring ensures repeatable sentiment inputs by rating every announcement from extremely bearish to extremely bullish, centralizing event scoring so technical models can account for news-based conviction.
ebrahim-saniConvert news into time-decayed event signals for algorithmic trading.
GGwujunConvert event CSV data into quantitative trading signals with time decay.
thanhtai040805Generate event-driven trading signals from news and macro events using sentiment analysis and technical indicators.
IhsanDanish25Analyze news article sentiment for stocks using NLP scoring.