Quick Start

Get up and running with ForPrompt in under 5 minutes.

How it works: ForPrompt stores your prompts in the cloud with versioning and AI editing. When you're ready, run forprompt deploy to generate local TypeScript files. Your code imports these files directly — no runtime API calls needed.

1. Install the SDK

bash
npm install @forprompt/sdk

2. Create a prompt in the dashboard

Sign in to your dashboard and create a new prompt. Give it a unique key like onboarding or customer-support.

3. Initialize and deploy

Run these commands in your project directory:

bash
# Initialize ForPrompt (creates .forprompt config)
npx forprompt init

# Deploy prompts to local files
npx forprompt deploy

This creates a forprompt/ directory with your prompts as TypeScript files.

4. Import and use

Import prompts directly — no API key or network calls needed at runtime:

typescript
// Import your prompt (it's just a string!)
import { onboarding } from "./forprompt";

// Use with any LLM provider
import Anthropic from "@anthropic-ai/sdk";

const anthropic = new Anthropic();
const message = await anthropic.messages.create({
  model: "claude-sonnet-4-20250514",
  max_tokens: 1024,
  system: onboarding,  // Your prompt from ForPrompt
  messages: [{ role: "user", content: "Hello!" }]
});

5. Update prompts

When you update prompts in the dashboard, sync them to your codebase:

bash
# Sync latest prompts
npx forprompt deploy

# Commit to version control
git add forprompt/
git commit -m "Update prompts"

That's it!

Your prompts are now managed in ForPrompt (with AI editing, versioning, testing) but your app uses simple local imports with zero latency. Check out the for more details.