Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
Version History
Version 1.2.0 - Introduced subskills for configuration optimization and daily recommendation workflows, with dedicated SKILL.md documentation and Python scripts. - Added support for recurring optimize–then–recommend flows, with instructions and commands included in SKILL.md. - Updated SKILL.md: clarified runtime commands (using python3), and documented integration of new subskills and data storage structure.
What This Skill Does
Analyzes Chinese A-share markets for short-term (1-5 day) trading decisions. Runs signal scoring, logs predictions, and compares forecasts against actual market closes. Outputs daily reports with stock candidates or explicit no-trade notices.
Combines signal scoring, prediction logging, and post-hoc accuracy comparison in a single automated pipeline rather than requiring manual cross-referencing of each step.
When to use it
- Scanning for momentum stocks before the CN market opens
- Logging daily predictions to track forecast accuracy
- Comparing yesterday's picks against actual closing prices
- Running sector rotation checks on a specific date
- Generating a structured daily report for A-share watchlists
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: Give me a short-term trading recommendation for 2026-02-12
- 1Run short_term_signal_engine('2026-02-12') to compute weighted score and candidate list
- 2If candidates exist, call run_prediction_for_date('2026-02-12') to persist snapshot to data/decision_log.jsonl
- 3Call generate_daily_report('2026-02-12') to produce structured output
- 4If next-day data is available, call compare_prediction_with_market('2026-02-12', '2026-02-13') to evaluate accuracy
Scored candidate stocks with buy signals, or explicit '当前暂无可执行短线买入标的' message with reason and next action