FormLoom + v0
Prompt + llms.txtv0 generates React/Next UI. Even without MCP, our llms.txt means v0 emits correct FormLoom code; with the SDK it's typed.
1. Just ask
Type a prompt like:
“Generate a contact form that submits to FormLoom (api.formloom.ai) and shows a success state.”
What you get back
v0 calls get_snippet + provision_endpoint and writes framework-correct code with a live access key — it works on first run:
app/contact/page.tsx
// app/contact/page.tsx — Next.js 15 App Router, Server Action
export default function ContactPage() {
async function submit(formData: FormData) {
"use server";
const res = await fetch("https://formloom.vercel.app/api/submit/fl_provisioned_key", {
method: "POST",
headers: { "Content-Type": "application/json", Accept: "application/json" },
body: JSON.stringify(Object.fromEntries(formData)),
});
// FormLoom scores spam, stores the row, and emails you — no backend to wire.
return res.json();
}
return (
<form action={submit}>
<label>
Name
<input type="text" name="name" required />
</label>
<label>
Email
<input type="email" name="email" required />
</label>
<label>
Message
<textarea name="message" required placeholder="How can we help?"></textarea>
</label>
<!-- honeypot: bots fill this, humans don't see it -->
<input type="checkbox" name="botcheck" style="display:none" tabindex="-1" autocomplete="off" />
<button type="submit">Send</button>
</form>
);
}Steps
- 1Prompt v0 to build a form that posts to FormLoom.
- 2Because FormLoom publishes llms.txt, the model knows the correct endpoint shape and emits working code.
- 3Paste your real access key in place of the placeholder, or install `@formloom/client` for types.
FAQ
v0 is prompt-driven rather than MCP-native, but FormLoom's llms.txt means the model already knows how to emit correct FormLoom code.