qwen-dianjin
AI skills for banking credit review, insurance, and investment research
All Skills in This Repository (129)
Pure Emerald Level Indicatorscredit-collateral-risk-mgmt
Validate collateral admissibility, appraisal, registration, monitoring, and disposal compliance.
pre-visit-credit-analysis
Integrate registry, judicial, opinion, lending, and credit signals into a structured pre-visit risk brief.
credit-related-party-detection
Detect related parties and abnormal transactions for corporate credit risk assessment.
reviewer-visit-memo
Structure oral credit investigation observations into a six-dimension visit memo.
credit-case-intake-check
Validate credit application intake materials with an I1–I6 one-veto gate.
admission-rules-scan
Scan customer profiles against product rule libraries and output three-state admission assessments.
ai-risk-planning
Generate structured credit risk analysis and data collection task plans from loan application inputs.
univariate-analysis
Compute univariate distribution binning and IV-based predictive power for tabular data.
xgb-tuning
Optimize XGBoost hyperparameters via diagnostic-driven Optuna TPE search.
model-comparison
Run XGBoost, LR+WoE, and DNN on a shared split and aggregate OOT metrics.
model-explanation
Generate SHAP-based explainability reports for trained XGBoost classification models.
xgb-modeling
Train and evaluate XGBoost binary classification models with standardized metrics and portable artifacts.
Frequently Asked Questions
FAQPage SchemaHow to install qwen-dianjin?▼
Run `npx skills add aliyun/qwen-dianjin --all -g -y` in your terminal to install all skills in this suite globally.
What financial tasks can DianJin-SKILLS automate?▼
It covers bank credit review (intake checks, collateral risk, related-party detection), insurance agent work (policy explanation, gap analysis, renewal reminders), and investment research (company analysis, macro briefings, bond risk checks).
Does it support credit risk modeling with XGBoost?▼
Yes. The financial-engineering-expert skills provide end-to-end modeling: data profiling, feature analysis, XGBoost/LR/DNN training, hyperparameter tuning, and model comparison.
Do these skills work with Claude Code and other AI agents?▼
Yes. All skills follow the standard SKILL.md format and can be loaded by any agent framework that supports the skills protocol.
Are the skills compliant with financial regulations?▼
Each skill embeds regulatory constraints, audit trails, disclaimers, and one-vote-veto red lines based on Chinese banking and insurance rules, but final decisions always require licensed human review.
Related Repositories in Finance & Accounting
View All in Finance & Accounting→ccxt
Trade and pull market data from 100+ crypto exchanges
financial-services
AI agents for banking, research, fund admin, and wealth workflows
financial-services-plugins
AI agents for banking, research, fund admin, and wealth workflows