PromptL, a templating language designed for LLM prompting

Hacker News

Share card

PromptL, a templating language designed for LLM prompting

Hey HN! We just launched PromptL: a templating language built to simplify writing complex prompts for LLMs like GPT-4 and Claude. Why PromptL? Creating dynamic prompts for LLMs can get tricky, even with standardized APIs that use lists of messages and settings. While these formats are consistent, building complex interactions with custom logic or branching paths can quickly become repetitive and hard to manage as prompts grow. PromptL steps in to make this simple. It allows you to define and manage LLM conversations in a readable, single-file format, with support for control flow and chaining, while maintaining compatibility with any LLM API. Key Features - Role-Based Structure: Define prompts with roles (user, system, assistant) for organized conversations. - Control Flow: Add logic with if/else and loops for dynamic, responsive prompts. - Chaining Support: Seamlessly link prompts to build multi-step workflows. - Reusable Templates: Modularize prompts for easy reuse across projects. PromptL compiles into a format compatible with any LLM API, making integration straightforward. We created PromptL to make prompt engineering accessible to everyone, not just technical users. It offers a readable, high-level syntax for defining prompts, so you can build complex conversations without wrestling with JSON or extra code. With PromptL, even non-technical users can create advanced prompt flows, while developers benefit from reusable templates and a simple integration process. We’d love to hear your thoughts!