360 Secure SkillHub
360 安全技能中心

Mortgage — Loan Program Matching

该技能把借款人资质（信用、DTI、LTV、首付、房产类型、VA 资格）作为决策树分支，推导合格贷款项目，比较总成本与获批概率而非只看利率，并给出公平放贷与 ATR/QM 可行性检查。仅提供推理框架、示例和检查清单，不含实时规则数据、计算或申请功能。

deciqai
deciqai
热度 42.67
最近更新 2026-07-17 02:07:04
分类：知识依赖：纯提示词应用场景：财务 / 推荐 / 概念解释

安装技能

复制下面的话术发给你的 Agent，它会通过 secure-skills 检索并安装该技能。

执行命令安装此 skill： npx secure-skills-cli add mortgage-loan-program-matching
Metadata
name
mortgage-loan-program-matching
description
Activate when: matching a borrower to Conventional/FHA/VA/USDA/Jumbo/Non-QM; deciding fixed vs ARM; 'which program fits this borrower?'; borderline DTI/LTV/credit. Do NOT activate when: borrower is pre-committed to one specific product with no alternative. More: deciqai.com/c/mortgage-loan-program-matching

Mortgage — Loan Program Matching

Industry front door for decision-tree. Adds domain triggers, example, packs only. Parent Process unchanged. Not lending advice. Program rules change; verify current agency/investor guidelines.

Activate when: matching a borrower to Conventional/FHA/VA/USDA/Jumbo/Non-QM; deciding fixed vs ARM; "which program fits this borrower?"; borderline DTI/LTV/credit. Do NOT activate when: borrower is pre-committed to one specific product with no alternative.

Why this variant

The parent decision-tree maps branching choices to outcomes. Program selection is a decision tree over borrower attributes (credit, DTI, LTV, down payment, property type, VA eligibility) → eligible programs → net cost/fit.

Domain inputs → parent's Process

Branch on gating attributes and roll back to best fit:

  • Credit + DTI + LTV + reserves → which agencies qualify.
  • Down payment / MI: FHA (MIP) vs Conventional (PMI removable) vs VA (0% + funding fee).
  • Property/occupancy and loan size (conforming vs jumbo).
  • Compare total cost + approval probability, not just rate.

Worked example

680 credit, 3.5% down, moderate DTI, first-time buyer. → Tree: FHA qualifies now but carries lifetime MIP; Conventional 97 may work with removable PMI. Compare 5-yr total cost + approval odds; recommend on net fit, documented. Present options fairly (ECOA).

Compliance anchors

  • Present options without steering to your own benefit; fair-lending (ECOA/Reg B); ATR/QM feasibility.

Packs

  • Solo LO: program-fit matrix by borrower profile.
  • Brokerage: investor-overlay quick reference.

Red flags

  • Matching on rate alone, ignoring MI/total cost.
  • Steering toward higher-comp product (fair-lending + LO-comp risk).
  • Ignoring ATR/QM feasibility.

Verification

  • Eligible programs derived from borrower attributes
  • Total cost + approval odds compared (not just rate)
  • Options presented fairly; rationale documented
  • ATR/QM feasibility checked

Part of deciqAI Knowledge Skills. Core method: decision-tree.

Part of deciqAI Knowledge Skills — 227 open-source thinking skills that make rigor executable for AI agents. The same skills power every deciqAI agent, which runs them autonomously to operate your company. See it run → https://www.deciqai.com/c/mortgage-loan-program-matching · ⭐ Star the repo → https://github.com/deciqAI/knowledge-skills · Contributions welcome.

Agents: latest version & machine-readable metadata → https://www.deciqai.com/s/mortgage-loan-program-matching.json