Skip to main content

Why PromptPack?

The Engineering Challenge​

Prompts have become critical business logic. They orchestrate AI systems, integrate with tools, handle multi-turn conversations, and make decisions that impact users directly. But most prompt development lacks the discipline we apply to traditional software engineering.

Consider what we have for traditional code:

  • Version control (Git)
  • Package managers (npm, pip, Maven)
  • Testing frameworks (Jest, pytest, JUnit)
  • CI/CD pipelines
  • Dependency management
  • Reproducible builds

Now consider typical prompt development:

  • Prompts scattered across codebases
  • Copy-pasted between projects
  • No standard format
  • Framework lock-in
  • Difficult to test systematically
  • Hard to version independently

PromptPack brings software engineering discipline to conversational AI.

Framework Independence​

AI frameworks come and go. Your prompt logic shouldn't.

PromptPack is deliberately NOT a framework—it's a specification. Just as:

  • OpenAPI specifies REST APIs independent of implementation
  • Docker images are portable across orchestration platforms
  • SQL is a standard across different databases

PromptPack provides a standard format that works across:

  • Any LLM provider (OpenAI, Anthropic, Google, local models)
  • Any runtime implementation (PromptKit, custom runtimes)
  • Any orchestration framework (when they add PromptPack support)

Multi-Prompt Architecture​

Generic prompts trying to do everything perform worse than specialized prompts optimized for specific scenarios.

A production customer service AI needs:

  • Support prompt: Optimized for empathy, ticket creation, escalation
  • Sales prompt: Focused on product knowledge, opportunity detection
  • Technical prompt: Detailed troubleshooting, diagnostic tools
  • Billing prompt: Payment handling, invoice generation, PII protection

Each prompt:

  • Uses appropriate temperature settings for its task
  • Has specialized tools and validators
  • Can evolve independently with its own version
  • Shares common fragments and configuration

See real examples in our Specification Examples.

Production Ready​

PromptPack isn't just for documentation—it's designed for production use:

Testing: Built-in test metadata tracks which models have been tested and their success rates. PromptArena provides systematic multi-provider testing.

Safety: Validators and guardrails travel with your prompts. Define content filters, length limits, and custom validation rules once.

Observability: Structured format enables monitoring, logging, and analytics across your prompt infrastructure.

Governance: Version every prompt independently. Track changes. Roll back when needed. Audit who changed what.

Current Status and Roadmap​

PromptPack is an emerging specification:

✅ Available Now:

  • Complete v1.1 specification with multimodal support
  • JSON Schema for validation
  • Comprehensive documentation and real-world examples
  • PromptArena testing tool for multi-provider evaluation
  • RFC process for community-driven evolution

🚧 Under Development:

  • PromptKit reference runtime implementation
  • Language-specific SDK libraries (Python, JavaScript)
  • Validation and linting tools

🔮 Future Vision:

  • Growing ecosystem of compatible tools
  • Community-contributed PromptPack library
  • Framework integrations (LangChain, LlamaIndex, etc.)
  • PromptPack Hub for sharing and discovering packs

We're building toward an ecosystem. Join us in shaping the future of prompt engineering.

Getting Started​

Ready to try PromptPack?

  1. Read the Spec: Understand the core concepts and structure
  2. See Examples: Review real-world PromptPacks
  3. Validate Your Ideas: Use the JSON Schema to validate your PromptPacks
  4. Test at Scale: Try PromptArena for multi-provider testing
  5. Contribute: Join the RFC process to shape the specification

Get Started →