Build responsible, controlled and transparent applications on top of LLM models!

• 💙 Working on the Book language v1
• 📚 Support of .docx, .doc and .pdf documents
• ✨ Support of OpenAI o1 model

If you have a simple, single prompt for ChatGPT, GPT-4, Anthropic Claude, Google Gemini, Llama 3, or whatever, it doesn't matter how you integrate it. Whether it's calling a REST API directly, using the SDK, hardcoding the prompt into the source code, or importing a text file, the process remains the same.

But often you will struggle with the limitations of LLMs, such as hallucinations, off-topic responses, poor quality output, language and prompt drift, word repetition repetition repetition repetition or misuse, lack of context, or just plain w𝒆𝐢rd resp0nses. When this happens, you generally have three options:

1. Fine-tune the model to your specifications or even train your own.
2. Prompt-engineer the prompt to the best shape you can achieve.
3. Orchestrate multiple prompts in a pipeline to get the best result.

In all of these situations, but especially in 3., the ✨ Promptbook can make your life waaaaaaaaaay easier.

• Separates concerns between prompt-engineer and programmer, between code files and prompt files, and between prompts and their execution logic. For this purpose, it introduces a new language called the 💙 Book.
• Book allows you to focus on the business logic without having to write code or deal with the technicalities of LLMs.
• Forget about low-level details like choosing the right model, tokens, context size, temperature, top-k, top-p, or kernel sampling. Just write your intent and persona who should be responsible for the task and let the library do the rest.
• We have built-in orchestration of pipeline execution and many tools to make the process easier, more reliable, and more efficient, such as caching, compilation+preparation, just-in-time fine-tuning, expectation-aware generation, agent adversary expectations, and more.
• Sometimes even the best prompts with the best framework like Promptbook :) can't avoid the problems. In this case, the library has built-in anomaly detection and logging to help you find and fix the problems.
• Versioning is build in. You can test multiple A/B versions of pipelines and see which one works best.
• Promptbook is designed to use RAG (Retrieval-Augmented Generation) and other advanced techniques to bring the context of your business to generic LLM. You can use knowledge to improve the quality of the output.

Promptbook project is ecosystem of multiple projects and tools, following is a list of most important pieces of the project:

Promptbook pipelines are written in markdown-like language called Book. It is designed to be understandable by non-programmers and non-technical people.

# 🌟 My first Book

-   INPUT PARAMETER {subject}
-   OUTPUT PARAMETER {article}

## Sample subject

> Promptbook

-> {subject}

## Write an article

-   PERSONA Jane, marketing specialist with prior experience in writing articles about technology and artificial intelligence
-   KNOWLEDGE https://ptbk.io
-   KNOWLEDGE ./promptbook.pdf
-   EXPECT MIN 1 Sentence
-   EXPECT MAX 1 Paragraph

> Write an article about the future of artificial intelligence in the next 10 years and how metalanguages will change the way AI is used in the world.
> Look specifically at the impact of {subject} on the AI industry.

-> {article}

This library is divided into several packages, all are published from single monorepo. You can install all of them at once:

npm i ptbk

Or you can install them separately:

⭐ Marked packages are worth to try first

• ⭐ ptbk - Bundle of all packages, when you want to install everything and you don't care about the size
• promptbook - Same as ptbk
• @promptbook/core - Core of the library, it contains the main logic for promptbooks
• @promptbook/node - Core of the library for Node.js environment
• @promptbook/browser - Core of the library for browser environment
• ⭐ @promptbook/utils - Utility functions used in the library but also useful for individual use in preprocessing and postprocessing LLM inputs and outputs
• @promptbook/markdown-utils - Utility functions used for processing markdown
• (Not finished) @promptbook/wizzard - Wizard for creating+running promptbooks in single line
• @promptbook/execute-javascript - Execution tools for javascript inside promptbooks
• @promptbook/openai - Execution tools for OpenAI API, wrapper around OpenAI SDK
• @promptbook/anthropic-claude - Execution tools for Anthropic Claude API, wrapper around Anthropic Claude SDK
• @promptbook/azure-openai - Execution tools for Azure OpenAI API
• @promptbook/langtail - Execution tools for Langtail API, wrapper around Langtail SDK
• @promptbook/fake-llm - Mocked execution tools for testing the library and saving the tokens
• @promptbook/remote-client - Remote client for remote execution of promptbooks
• @promptbook/remote-server - Remote server for remote execution of promptbooks
• @promptbook/pdf - Read knowledge from .pdf documents
• @promptbook/documents - Read knowledge from documents like .docx, .odt,…
• @promptbook/legacy-documents - Read knowledge from legacy documents like .doc, .rtf,…
• @promptbook/website-crawler - Crawl knowledge from the web
• @promptbook/types - Just typescript types used in the library
• @promptbook/cli - Command line interface utilities for promptbooks

The following glossary is used to clarify certain concepts:

• Prompt drift is a phenomenon where the AI model starts to generate outputs that are not aligned with the original prompt. This can happen due to the model's training data, the prompt's wording, or the model's architecture.
• Pipeline, workflow or chain is a sequence of tasks that are executed in a specific order. In the context of AI, a pipeline can refer to a sequence of AI models that are used to process data.
• Fine-tuning is a process where a pre-trained AI model is further trained on a specific dataset to improve its performance on a specific task.
• Zero-shot learning is a machine learning paradigm where a model is trained to perform a task without any labeled examples. Instead, the model is provided with a description of the task and is expected to generate the correct output.
• Few-shot learning is a machine learning paradigm where a model is trained to perform a task with only a few labeled examples. This is in contrast to traditional machine learning, where models are trained on large datasets.
• Meta-learning is a machine learning paradigm where a model is trained on a variety of tasks and is able to learn new tasks with minimal additional training. This is achieved by learning a set of meta-parameters that can be quickly adapted to new tasks.
• Retrieval-augmented generation is a machine learning paradigm where a model generates text by retrieving relevant information from a large database of text. This approach combines the benefits of generative models and retrieval models.
• Longtail refers to non-common or rare events, items, or entities that are not well-represented in the training data of machine learning models. Longtail items are often challenging for models to predict accurately.

Note: Thos section is not complete dictionary, more list of general AI / LLM terms that has connection with Promptbook

• Organization (legacy name collection) group jobs, workforce, knowledge, instruments, and actions into one package. Entities in one organization can share resources (= import resources from each other).
  • Jobs
    • Task
    • Subtask
  • Workforce
    • Persona
    • Team
    • Role
  • Knowledge
    • Public
    • Private
    • Protected
  • Instruments
  • Actions

• Book file
  • Section
    • Heading
    • Description
    • Command
    • Block
    • Return statement
  • Comment
  • Import
  • Scope

• 📚 Collection of pipelines
• 📯 Pipeline
• 🎺 Pipeline templates
• 🤼 Personas
• ⭕ Parameters
• 🚀 Pipeline execution
• 🧪 Expectations
• ✂️ Postprocessing
• 🔣 Words not tokens
• ☯ Separation of concerns

• 📚 Knowledge (Retrieval-augmented generation)
• 🌏 Remote server
• 🃏 Jokers (conditions)
• 🔳 Metaprompting
• 🌏 Linguistically typed languages
• 🌍 Auto-Translations
• 📽 Images, audio, video, spreadsheets
• 🔙 Expectation-aware generation
• ⏳ Just-in-time fine-tuning
• 🔴 Anomaly detection
• 👮 Agent adversary expectations
• view more

• Anonymous mode
• Application mode

• Simple usage
• Usage with client and remote server

• When you are writing app that generates complex things via LLM - like websites, articles, presentations, code, stories, songs,...
• When you want to separate code from text prompts
• When you want to describe complex prompt pipelines and don't want to do it in the code
• When you want to orchestrate multiple prompts together
• When you want to reuse parts of prompts in multiple places
• When you want to version your prompts and test multiple versions
• When you want to log the execution of prompts and backtrace the issues

See more

• When you have already implemented single simple prompt and it works fine for your job
• When OpenAI Assistant (GPTs) is enough for you
• When you need streaming (this may be implemented in the future, see discussion).
• When you need to use something other than JavaScript or TypeScript (other languages are on the way, see the discussion)
• When your main focus is on something other than text - like images, audio, video, spreadsheets (other media types may be added in the future, see discussion)
• When you need to use recursion (see the discussion)

See more

• 🤸‍♂️ Iterations not working yet
• ⤵️ Imports not working yet

• ➿ No recursion
• 🏳 There are no types, just strings

If you have a question start a discussion, open an issue or write me an email.

• ❔ Why not just use the OpenAI SDK / Anthropic Claude SDK / ...?
• ❔ How is it different from the OpenAI`s GPTs?
• ❔ How is it different from the Langchain?
• ❔ How is it different from the DSPy?
• ❔ How is it different from anything?
• ❔ Is Promptbook using RAG (Retrieval-Augmented Generation)?
• ❔ Is Promptbook using function calling?

See CHANGELOG.md

Promptbook by Pavol Hejný is licensed under CC BY 4.0

See TODO.md

I am open to pull requests, feedback, and suggestions. Or if you like this utility, you can ☕ buy me a coffee or donate via cryptocurrencies.

You can also ⭐ star the promptbook package, follow me on GitHub or various other social networks.