quantskills/skill-quant-factor-skill-factory
Developer toolsThis page was translated automatically. Technical details may differ from the original. View the original
QuantSkills factor Skill repository
Overview
Not the factor library itself, but the tools for continuing to produce it: batch generation, validation, and packaging framework-neutral OHLCV quantization factor skills. skill-quant-factor-skill-factory is a factor production tool Skill provided by the QuantSkills organization. It is used for batch generation, validation, and packaging of framework-neutral OHLCV quantization factor skills. QuantSkills GitHub organization: https://github.com/quantskills It is not a factor library itself, but a tool for continuing to produce the factor library. Currently, QuantSkills publicly discloses three types of factor libraries: - skill-quant-factor-directional-alpha - skill-quant-factor-risk-pattern-alpha - skill-quant-factor-volume-stat-alpha When you want to organize a new set of OHLCV factor ideas into a Skill package that can be installed, verified, and called by AI Agents, you can use this repository. - Batch generation factor Skill folder - Generates SKILL.md, user README, formula descriptions, and agent metadata for each factor - Generates framework-neutral scripts/factor.py - Generates standalone scripts/validate.py - Computes validation metrics based on real market panels - Maintains factor_index.json and merged indexes - Generates batch evaluation reports - Available sample coverage - 5-day Rank IC mean - 5-day ICIR - Five-group Q5-Q1 earnings spread - Top group turnover rate - No future function check
Overview (machine translated)
Chinese Simplified | English
is not the factor library itself, but a tool for continuing to produce it: batch generation, validation, and packaging framework-neutral OHLCV quantization factor skills.
skill-quant-factor-skill-factory is a factor production tool Skill provided by the QuantSkills organization. It is used for batch generation, validation, and packaging of framework-neutral OHLCV quantization factor skills.
QuantSkills GitHub organization: https://github.com/quantskills
It is not a factor library itself, but a tool for continuing to produce the factor library. Currently, QuantSkills publicly discloses three types of factor libraries:
skill-quant-factor-directional-alphaskill-quant-factor-risk-pattern-alphaskill-quant-factor-volume-stat-alpha
🎯 What problem does this Skill solve?
When you want to organize a new set of OHLCV factor ideas into a Skill package that can be installed, verified, and called by AI Agents, you can use this repository.
It will automatically complete:
- Batch generate the Skill folder for the factor
- Generates
SKILL.md, user README, formula description, and agent metadata for each factor - Generate frame-neutral
scripts/factor.py - Generate standalone
scripts/validate.py - Validation metrics are calculated based on real market panels
- Maintain
factor_index.jsonand merge indexes - Generate batch evaluation reports
⚡ Production Line
🗃️ Enter data requirements
Target projects require preparing a real market data panel:
Market field must include:
Recommended extra reservations:
📦 The generated factor structure
Each factor generates an independent skill:
The generated factors can be called individually by an agent environment supported by the SKILL.md specification.
🚀 Get Started Quickly
Operating in target projects:
Parameter Description:
| Parameters | Description |
|---|---|
--count |
Number of factors generated in this batch |
--existing-index |
Existing factor indexes are used to avoid duplicate |
--output-root |
The output directory for this batch is |
--combined-index |
The generated merged index |
--report-name |
Factor Evaluation Report File Name |
🧪 Verification metrics
During batch generation, each factor is calculated:
- Available sample coverage
- 5-day Rank IC mean
- 5th ICIR
- Five groups: Q5-Q1 earnings spread
- Top group turnover rate
This page was translated automatically. Technical details may differ from the original.
Recommended Tools
Try a different keyword or remove a filter.
Install
npx skillfish add quantskills/skill-quant-factor-skill-factory