Examples

Real-world examples of composable AI prompts

Basic Text Analysis

A simple prompt with parameters

---
id: text-analyzer
name: Text Analyzer
version: 1.0.0
description: Analyzes text for sentiment and key themes
parameters:
  - name: text
    type: string
    required: true
  - name: analysis_type
    type: enum
    values: [sentiment, themes, summary]
    default: sentiment
---

# Text Analysis

Analyze the following text for {analysis_type}:

{text}

Provide detailed insights and actionable recommendations.

Data Processing with Inheritance

Building on a base prompt template

---
id: advanced-data-processor
name: Advanced Data Processor
version: 2.0.0
description: Advanced data processing with visualization
inherits: "@prompd/base-data-analyzer@1.5.0"
parameters:
  - name: dataset
    type: string
    required: true
  - name: visualize
    type: boolean
    default: false
---

# Advanced Data Processing

Process the dataset: {dataset}

{% if visualize %}
Create visualizations for the key metrics.
{% endif %}

Apply statistical analysis and generate insights.

Document Analysis with Binary Assets

Extracting content from Excel and PDF files

---
id: financial-report-analyzer
name: Financial Report Analyzer
version: 1.0.0
description: Analyzes financial reports from Excel and PDF
assets:
  - path: ./quarterly-data.xlsx
    type: excel
    extract: all
  - path: ./annual-report.pdf
    type: pdf
    extract: text
parameters:
  - name: fiscal_year
    type: string
    required: true
---

# Financial Report Analysis - FY {fiscal_year}

Analyze the quarterly data from the Excel file and cross-reference
with the annual report PDF.

Provide:
1. Revenue trends
2. Key performance indicators
3. Risk assessment
4. Strategic recommendations

Code Review with Model Configuration

Advanced example using object parameters for flexible model configuration

---
id: code-reviewer
name: Code Reviewer
version: 1.0.0
description: Comprehensive code review assistant
parameters:
  - name: code
    type: string
    required: true
  - name: language
    type: string
    required: true
  - name: focus_areas
    type: array
    default: [security, performance, maintainability]
  - name: model_config
    type: object
    properties:
      provider:
        type: string
        enum: [openai, anthropic, ollama]
      model:
        type: string
      temperature:
        type: number
    required: [provider, model]
    default:
      provider: openai
      model: gpt-4o
      temperature: 0.7
---

# Code Review - {language}

Review the following {language} code:

```{language}
{code}
```

Focus on: {% for area in focus_areas %}{{ area }}{% if not loop.last %}, {% endif %}{% endfor %}

Provide:
- Security vulnerabilities
- Performance optimizations
- Code quality improvements
- Best practice recommendations

CLI Usage Examples

Create and Run a Prompt

# Initialize a new project folder
prompd init ./my-project

# Run with parameters
prompd run my-project/example.prmd --param text="This product is amazing!"

Package and Publish

# Create a package from current directory
prompd package create . my-analyzer.pdpkg

# Validate package
prompd package validate my-analyzer.pdpkg

# Publish to registry
prompd login
prompd publish my-analyzer.pdpkg

Install and Use Published Packages

# Search registry
prompd search "sentiment analysis"

# Install package
prompd install @username/sentiment-analyzer@1.0.0

# Run installed package
prompd run @username/sentiment-analyzer --param text="Great service!"

Explore More Examples

Browse hundreds of community-created prompts and packages

Visit Registry →