# AI Apps That Work

> Make your AI work first time, for less. A Next.js starter for apps with AI features. It records which AI results users had to redo and what each cost, so Claude Code or Cursor can fix what fails.

Free tools for builders of AI apps: real cost per AI result, model prices, switch costs and shutdown dates, LLM judge checks and A/B test sample sizes. Every page has a Markdown version: add .md to its URL (https://aiappsthatwork.com/index.md for the home page) or send Accept: text/markdown.

## For agents
- [Connect your agent](https://aiappsthatwork.com/connect.md): MCP setup for Claude Code, Cursor, Claude and ChatGPT, curl examples and the skill
- [Skill](https://aiappsthatwork.com/skill/SKILL.md): SKILL.md describing every tool and its curl call
- [MCP server](https://aiappsthatwork.com/mcp): Streamable HTTP, no login, 8 read-only tools that return numbers and a permalink
- [OpenAPI description](https://aiappsthatwork.com/openapi.json): POST /api/tools/<slug> with JSON returns the numbers and a permalink; no key, inputs not stored
- [Model prices and shutdowns](https://aiappsthatwork.com/models.json): JSON, with sources and the date checked

## Free tools
- [Judge agreement checker](https://aiappsthatwork.com/tools/llm-judge-agreement.md): Check how often your AI judge agrees with a person, and whether to trust it.
- [Judge prompt builder](https://aiappsthatwork.com/tools/judge-prompt-builder.md): A pass/fail judge prompt for any AI step, from support replies to invoice fields.
- [Model switch test](https://aiappsthatwork.com/tools/model-switch-test.md): Run your prompt on two models side by side and compare what worked and what it really costs.
- [Real cost per result calculator](https://aiappsthatwork.com/tools/real-cost-per-ai-result.md): What each AI result really costs, what you waste a month, and what to charge in credits.
- [Prompt A/B test sample size](https://aiappsthatwork.com/tools/prompt-ab-test-sample-size.md): How many results you need before a prompt or model change is proven.
- [AI launch checklist](https://aiappsthatwork.com/tools/ai-launch-checklist.md): The AI-specific checks that generic vibe-coding checklists skip.
- [Model shutdown calendar](https://aiappsthatwork.com/tools/model-shutdown-calendar.md): Announced API shutdown dates and replacements, with sources.

## Pages
- [Home](https://aiappsthatwork.com/index.md)
- [The template](https://aiappsthatwork.com/template.md): what it is, its building blocks and status
- [Model prices and switch costs](https://aiappsthatwork.com/models.md): every priced model and the model-switch pages