> ## Documentation Index
> Fetch the complete documentation index at: https://prompt-lockbox.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# Introduction

> Welcome to Prompt Lockbox! From chaotic text files to well kept prompt templates.

<Note>
  **Project Under Development!**

  This project is currently in active development. You may experience occasional bugs or inconsistencies. Please report any issues on [Report Bugs](/how_to_contribute) section—your input is valuable.
</Note>

### What is Prompt Lockbox ?

As language models become central to our applications, the prompts that control them are often treated as an afterthought—scattered across code, text files, and documents. This chaos leads to **inconsistency**, **lost work**, and a **lack of security** for your most critical AI assets.

<Card>
  Prompt Lockbox is a **command-line tool** and **Python SDK** designed to fix this. It provides a professional framework for bringing **structure**, **security**, and **version-controlled** management to your entire prompt engineering workflow, transforming your prompts from simple text into robust, reliable, and manageable code.
</Card>

<Tip>
  **Quick Tip!**

  As all project configuration and state are stored in centralized files, you can use the CLI or SDK interchangeably. Actions performed with one are immediately reflected when using the other.
</Tip>

### What problems does it solve ?

Here’s what it fixes for you -

<AccordionGroup>
  <Accordion title="Scattered Prompts Problem">
    * **Problem**: Your team's best prompts are lost in a chaotic mess of text files, Slack messages, and Google Docs. No one knows where the "official" version is, leading to confusion and duplicated effort
    * **Solution**: It creates a centralized and structured library, making it effortless for your entire team to find and use the correct prompt.
  </Accordion>

  <Accordion title="Which Version Was That? Problem">
    * **Problem**: You tweak a great prompt and accidentally make it worse, with no way to go back to the version that worked.
    * **Solution**: It provides Git-like versioning for prompts. Safely create new versions and manage modifications, just like code.
  </Accordion>

  <Accordion title="Did Someone Change This? Problem">
    * **Problem**: A critical prompt is changed without approval, breaking your application and causing unpredictable outputs..
    * **Solution**: It lets you lock and secure key prompts. The system instantly detects any unauthorized edits, ensuring production reliability.
  </Accordion>

  <Accordion title="Reinventing the Wheel Problem">
    * **Problem**: Your team wastes time building prompts that already exist because they're impossible to find.
    * **Solution**: It makes your entire library instantly searchable. Find what you need in seconds with powerful, context-aware search.
  </Accordion>

  <Accordion title="Good Prompt, Bad Documentation Problem">
    * **Problem**: You write a brilliant prompt but have no time to document it, making it unusable for others (or your future self).
    * **Solution**: It uses AI to automate documentation. One command generates a clear description and search tags, turning your prompts into reusable assets.
  </Accordion>
</AccordionGroup>

### Prompt Format

Let's have a quick look on how it creates an awesome prompt.

<Accordion title="Code to create prompt" icon="code">
  ```python theme={null}
  from prompt_lockbox import Project
  project = Project()

  new_prompt = project.create_prompt(
    name="Email-agent",
    version="1.0.0",
    description="Create a persuasive email template for small tech business owners to promote a chatbot product.",
    namespace=["email-tools", "ai-agent"],
    tags=[
        "email",
        "persuasive",
        "small business",
        "tech industry",
        "chatbot"
    ],
    author="ananya868 <ananya8154@gmail.com>",
    intended_model="gpt-9",
    model_parameters={
        "temperature": 0.7
    },
    linked_prompts=[
        "email-monitor-agent",
        "client-scrapper-agent"
    ],
    notes="Do not use in production before testing!"
  )
  ```
</Accordion>

The above code creates a prompt template with metadata. Go ahead and plug in your actual prompt. **Here’s the final version**:

<img src="https://mintcdn.com/prompt-lockbox/w9NVcZ5JBDkt-pSS/images/prompt_template.png?fit=max&auto=format&n=w9NVcZ5JBDkt-pSS&q=85&s=9dc595e32185709ccfdfa9d3cebb182c" alt="Prompt Template" height="300" className="rounded-lg" data-path="images/prompt_template.png" />

### Core Features in a nutshell

A brief look at what Prompt Lockbox can do for you.

<Note>
  This is just an overview of the features. Please checkout [Use Cases](/usecase/first) for detailed usage !
</Note>

<CardGroup col={3}>
  <Card title="Standardized Structure">
    Define every prompt in a clean, lintable **YML format**. This enforces consistency, separates logic from metadata, and makes your entire prompt library easy to understand and maintain.
  </Card>

  <Card title="Python SDK Integration">
    A lightweight and intuitive Python SDK allows you to seamlessly **load**, **render**, and **execute** your version-controlled prompts directly within your applications.
  </Card>

  <Card title=" Versioning & History">
    Treat your prompts like code. Create new versions with **semantic versioning** (v1.0.0 -> v1.1.0) to build a complete audit trail and safely experiment without breaking what works.
  </Card>

  <Card title="Integrity & Security">
    Lock your production-ready prompts to protect them from accidental or unauthorized changes. The framework will instantly **detect any tampering**, ensuring application's reliability.
  </Card>

  <Card title="Powerful Search">
    Stop reinventing the wheel. Instantly find any prompt in your library with **fuzzy, keyword, or advanced semantic search** that understands the intent behind your query.
  </Card>

  <Card title="AI Features">
    Let AI do the tedious work. Automatically **generate documentation**, get expert **critiques on your prompts**, and receive **suggestions for improvement**, all from a single command.
  </Card>
</CardGroup>

### Quickly Access Documentation

<CardGroup cols={2}>
  <Card title="Quickstart" href="/quickstart">
    Get started with the framework.
  </Card>

  <Card title="Basics" href="/guide/first">
    Learn the essentials of this toolkit.
  </Card>

  <Card title="Use Cases" href="/usecase/first">
    Explore advanced features and use cases.
  </Card>

  <Card title="API Reference" href="/api-reference/Introduction">
    Complete technical docs for the CLI and Python API.
  </Card>

  <Card title="Contribution" href="www.google.com">
    Contribute to this toolkit.
  </Card>
</CardGroup>

<Info>
  To contribute to this open source toolkit. Please follow guidelines described at [Contribution](/how_to_contribute) !
</Info>

## Next Steps

Here's how to install and get started with the toolkit -

<Card title="Quickstart" href="/quickstart" icon="play">
  Setup the toolkit and explore the basics.
</Card>