PromptStore

A lightweight Python package for managing and versioning LLM prompt templates.

Features

• Simple JSON-based storage
• Template versioning
• Tag-based organization
• Jinja2 template support
• Variable highlighting in Markdown exports
• Package integration utilities
• Easy to extend and customize

Quick Start

from promptstore import PromptStore

# Create a new store
store = PromptStore("./prompts")

# Add a prompt template
prompt = store.add(
    content="Write a {{language}} function that {{task}}",
    description="Code generation prompt",
    tags=["coding", "generation"]
)

# Use the prompt
filled = prompt.fill({
    "language": "Python",
    "task": "sorts a list in reverse order"
})

This automatically generates a markdown representation like this:

```markdown

uuid: 77a6b015-efea-436e-8636-6ae06c438ad8 name: example namespace: project1 description: Code generation prompt version: 1 tags: - coding - generation variables: - language - task subset: subproject1 created_at: '2025-11-12T11:17:24.128334+00:00' updated_at: '2025-11-12T11:17:24.128334+00:00'

---

Write a {{language}} function that {{task}}

This makes it easy to identify which parts of the prompt are variables when viewing the markdown files.

Why PromptStore?

• Lightweight: Minimal dependencies, just Jinja2
• Simple: JSON-based storage, easy to inspect and modify
• Flexible: Support for both local development and package distribution
• Versioned: Track changes to your prompts
• Organized: Tag and search your prompts
• Package-Ready: Easy to bundle prompts with your package