Integration · Anthropic

Anthropic Prompt Management

Your Claude Opus 4.6 system prompts deserve version control. Store, iterate, and serve them via API without touching your application code.

Start Free Trial Anthropic API Docs

Integration Example

PromptForge + Anthropic in one file

Version, template, and serve Anthropic prompts via REST API. Keep Claude Opus 4.6's system prompts in PromptForge, iterate freely, track every change, and roll back instantly without redeploying your app.

1
Fetch your versioned, interpolated prompt from the PromptForge REST API with a single fetch() call.
2
Pass the returned content string directly to the Anthropic SDK as the system prompt, no transformation needed.
3
Update the prompt in the PromptForge dashboard anytime, running applications pick up the change on the next request. No redeployment required.

integration.ts

TypeScript

import Anthropic from "@anthropic-ai/sdk";

const anthropic = new Anthropic({ apiKey: process.env.ANTHROPIC_API_KEY });

async function fetchPrompt(persona: string, language: string) {
  const res = await fetch(
    "https://www.promptforge-app.com/api/v1/prompts/your-prompt-id",
    {
      method: "POST",
      headers: {
        Authorization: "Bearer pfk_your_api_key",
        "Content-Type": "application/json",
      },
      body: JSON.stringify({
        version: "latest",
        variables: { persona, language },
      }),
    }
  );
  const { content } = await res.json();
  return content as string;
}

// Prompt template in PromptForge (e.g.):
//   You are a {{persona}} assistant. Always respond in {{language}}.
// 1. Fetch your versioned Claude system prompt with dynamic variables
const content = await fetchPrompt("helpful_assistant", "English");

// 2. Use the versioned system prompt with Claude
//    Swap personas or languages by changing the PromptForge template,
//    no application redeployment required
const message = await anthropic.messages.create({
  model: "claude-opus-4-6",
  max_tokens: 1024,
  system: content,
  messages: [{ role: "user", content: "Summarise the quarterly report." }],
});

console.log(message.content[0].text);

How It Works

Anthropic + PromptForge in Three Steps

Add a prompt management layer to your Anthropic integration without refactoring your application.

Step 1

Store your Claude system prompt as a template

System prompts are the single highest-leverage part of any Claude conversation. Write yours in PromptForge with {{variable}} placeholders for persona, language, or instructions.

Step 2

Iterate safely with full version history

Anthropic recommends careful, incremental prompt iteration. PromptForge gives you a full diff of every change, lets you A/B between versions, and lets you roll back to any previous version in one click.

Step 3

Inject the versioned prompt at runtime

Fetch the prompt via REST API and pass `content` directly to `anthropic.messages.create({ system: content })`. Update the system prompt in the dashboard; running applications pick it up on the next request.

Features

Powerful Prompt Management Features for AI Developers

From simple prompt storage to production-ready APIs with version control, dynamic variables, rollback, and a public gallery.

Dynamic Variables

Use {{variable}} syntax to create reusable prompt templates. Pass different values via API for endless customization across any LLM.

Instant Prompt API

RESTful API endpoint ready in seconds. Fetch any prompt with version pinning and variable interpolation. No redeployment needed.

Rollback with Diff Checker

View a line-by-line diff of every change and roll back to any previous version in one click. Never lose a working prompt again.

Publish to Gallery

Share your best prompts with the community in the public gallery. Get discovered by other developers and grow your personal library.

Start for free, upgrade anytime

No credit card required to get started. Paid plans include a 14-day free trial.

HobbyNo credit card

1k API requests/month
1 prompt
Unlimited versions
Dynamic variables
Version pinning
API key management

Start Managing My Prompts

Starter

14-day free trial

$9/month

No charge until your trial ends

10k API requests/month
5 prompt
Unlimited versions
Dynamic variables
Version pinning
API key management

Start 14-day Free Trial

Anthropic + PromptForge: Common Questions

Specific answers for developers integrating PromptForge with Anthropic.

How do I manage Claude's system prompt separately from the human turn?

Create two separate prompts in PromptForge, one for the system prompt and one for any repeating user-turn scaffold. Fetch each independently at runtime: pass the system prompt to the `system` parameter and the user scaffold to the first `messages` entry. This way each has its own version history and can be updated independently.

Does PromptForge work with Claude's extended thinking mode?

Yes. Extended thinking is enabled by adding `thinking: { type: 'enabled', budget_tokens: N }` to the `anthropic.messages.create` call, it is an API parameter, not part of the prompt text. Manage and version the system prompt in PromptForge as normal, and enable extended thinking in your application code alongside it.

Can I use PromptForge with the Anthropic Messages Batches API?

Yes. The Batches API accepts the same message structure as the standard Messages API. Fetch your versioned system prompt from PromptForge once, then reuse the `content` string across all requests in your batch payload. Because PromptForge returns plain text, there is no additional transformation needed before inserting it into each batch item.

How do I version multi-turn conversation prompts for Claude?

For multi-turn conversations, the system prompt sets the persistent persona and rules. Store that in PromptForge. Dynamic user messages are generated by application logic and don't need versioning. If you use a recurring assistant-turn template (for example, a structured output format reminder), store that as a second PromptForge prompt and inject it where needed in the messages array.