Why this package exists

Crafting the perfect system prompt for specific roles (like a Code Reviewer, Security Engineer, or Translator) requires significant trial and error.

@promptforgee/templates provides a library of pre-configured @promptforgee/core Prompt instances. These templates have been highly optimized and battle-tested in production to yield reliable, high-quality responses from leading LLMs. Instead of reinventing the wheel, simply import a template, inject your context, and execute.

Features

• Plug-and-Play: Import a template and use it immediately.
• Highly Tuned: Pre-configured with optimal roles, tasks, constraints, and tones.
• Composable: Templates are Prompt instances, meaning you can chain additional .context() or .constraint() calls to customize them further.
• Type-safe: Fully integrated with @promptforgee/core.

Installation

# npm
npm install @promptforgee/templates @promptforgee/core

# pnpm
pnpm add @promptforgee/templates @promptforgee/core

# yarn
yarn add @promptforgee/templates @promptforgee/core

# bun
bun add @promptforgee/templates @promptforgee/core

Quick Start

import { FrontendEngineer } from '@promptforgee/templates';
import { MarkdownFormatter } from '@promptforgee/core';

async function generateReactComponent(requirements: string) {
  // 1. Extend the pre-built template
  const myPrompt = FrontendEngineer.context(`Component Requirements: ${requirements}`).constraint(
    'Use Tailwind CSS for styling.',
  );

  // 2. Format it to a string
  const promptString = myPrompt.build(new MarkdownFormatter());

  // 3. Send to LLM...
  console.log(promptString);
}

API Overview

Each export is a pre-configured Prompt builder instance from @promptforgee/core.

| Export | Ideal Use Case | | ------------------ | ----------------------------------------------------------- | | FrontendEngineer | Generating UI components, HTML/CSS, React/Vue code. | | BackendEngineer | Writing server logic, API endpoints, database queries. | | ReactExpert | Highly specialized React development with modern hooks. | | CodeReviewer | Identifying bugs, performance issues, and code smells. | | Teacher | Explaining complex technical concepts clearly to beginners. | | Interviewer | Conducting simulated technical interviews. | | Debugger | Analyzing stack traces and suggesting fixes. | | SecurityEngineer | Auditing code for vulnerabilities (XSS, SQLi, etc.). | | Researcher | Summarizing academic papers or technical documentation. | | Translator | Translating text across spoken or programming languages. | | TechnicalWriter | Generating API documentation, READMEs, and guides. |

Real-world Example

Building a multi-agent CI/CD pipeline step:

import { CodeReviewer, SecurityEngineer } from '@promptforgee/templates';

async function performAutomatedCodeReview(pullRequestDiff: string) {
  // Generate the review prompt
  const reviewPrompt = CodeReviewer.context(`PR Diff:\n${pullRequestDiff}`).build();

  // Generate the security audit prompt
  const securityPrompt = SecurityEngineer.context(`PR Diff:\n${pullRequestDiff}`)
    .constraint('Focus strictly on OWASP Top 10 vulnerabilities.')
    .build();

  // Execute both prompts in parallel via your LLM provider
  const [review, security] = await Promise.all([runLLM(reviewPrompt), runLLM(securityPrompt)]);

  return { review, security };
}

Ecosystem

@promptforgee/templates provides the starting point for your PromptForge workflows.

@promptforgee/templates (You are here) ↓ @promptforgee/core (Customize the template) ↓ @promptforgee/analyzer (Validate your additions) ↓ @promptforgee/optimizer (Optimize the final prompt)

Documentation

For full documentation and advanced usage, visit promptforge.dev/docs/templates.

Examples

Check out our Examples directory for more real-world use cases.

Roadmap

• 🚧 DataScientist template for pandas/numpy data manipulation.
• 🚧 DevOpsEngineer template for Docker/Kubernetes/Terraform configurations.
• 🚧 Domain-specific templates (e.g., Legal Assistant, Financial Analyst).

Contributing

We welcome contributions! Please read our Contributing Guide to get started.

License

MIT © Omnikon-Org