MortgageCheck

A Next.js application that evaluates mortgage feasibility against European underwriting practice. Users enter their financial details (in EUR) and get an instant assessment — approved, conditionally approved, or denied — with a per-criterion breakdown (DTI ratio, LTV ratio, employment contract type, and a +3% rate stress test).

Tech stack

• Next.js 15 (App Router) with React 19 and TypeScript (strict)
• Tailwind CSS 3
• Vitest for unit and API tests

Quick start

npm install
npm run dev

Open http://localhost:3000 and submit a sample analysis.

To run the test suite:

npm test

Gemini CLI assets

This repository ships a feature-delivery pipeline built on Gemini CLI subagents and skills. The relevant files:

• GEMINI.md — project context auto-loaded by Gemini CLI. Describes the app, tech stack, structure, the subagent pipeline, and the skills shipped with the project.
• .gemini/agents/ — subagent definitions (Markdown files with YAML frontmatter):
  • architect — designs implementation plans (read + write plans only).
  • feature-dev — implements features (full access).
  • security-auditor — security review (read-only, writes SECURITY-REPORT.md).
  • test-writer — writes and runs Vitest tests (writes only inside __tests__/).
  • doc-writer — generates README.md and docs.
  • code-reviewer — reviews against the plan, writes CODE-REVIEW.md.
  • release-manager — branch, commit, push, open MR.
• .gemini/skills/ — agent skills following the Agent Skills standard:
  • eu-mortgage-rules — canonical European mortgage rule spec (DTI/LTV thresholds, employment contract handling, +3% stress test). Activated by architect, feature-dev, and test-writer.
  • code-review-rubric — severity-tag system and required CODE-REVIEW.md structure. Activated by code-reviewer.
• .gemini/settings.json — enables the built-in browser_agent (locked to localhost) for live-app verification.
• FEATURE-REQUEST.md — sample feature request the pipeline runs against (amortization schedule + scenario comparison).

To reproduce the pipeline run, install Gemini CLI, open this repo, and follow the prompts described in the companion blog post.