Skip to main content

A framework for creating AI applications and agents using natural language programming

Project description

Text changing depending on mode. Light: 'So light!' Dark: 'So dark!'

Create AI agents with natural language programs

Playbooks AI is a powerful framework for building AI agents with Natural Language Programming. It introduces a new "english-like", semantically interpreted programming language with reliable, auditable execution.

Playbooks AI is still in early development. We're working hard and would love your feedback and contributions.

Playbooks AI goes well beyond LLM tool calling. You can fluidly combine:

  • Business processes written as natural language playbooks
  • Python code for external system integrations, algorithmic logic, and complex data processing
  • Multiple local and remote AI agents interacting in novel ways

Unlike standard LLM prompts that offer no execution guarantees, Playbooks provides full visibility into every step of execution, ensuring your AI system follows all rules, executes steps in the correct order, and completes all required actions. Track and verify the entire execution path with detailed state tracking, call stacks, and execution logs.

🚀 Key Features

  • Natural Language Programming - Write agent logic in plain English with markdown playbooks that look like a step-by-step recipe
  • Python Integration - Seamlessly call natural language and Python playbooks on the same call stack for a radically new programming paradigm
  • Multi-Agent Architecture - Build systems with multiple specialized agents, interact and leverage external AI agents
  • Event-Driven Programming - Use triggers to create reactive, context-aware agents
  • Variables, Artifacts and Memory - Native support for managing agent state using variables, artifacts and memory
  • Execution Observability - Full audit trail of every step of execution and explainability for every decision made by the AI agent

🏁 Quick Start

Installation

pip install playbooks

Create Your First Playbook

Create a file named hello.pb:

# Personalized greeting
This program greets the user by name

## Greet
### Triggers
- At the beginning of the program
### Steps
- Ask the user for their name
- Say hello to the user by name and welcome them to Playbooks AI
- End program

Run Your Playbook

python -m playbooks.applications.agent_chat hello.pb --verbose

VSCode Support (Optional)

Install the Playbooks Language Support extension for Visual Studio Code:

  1. Open VSCode
  2. Go to Extensions (Ctrl+Shift+X / Cmd+Shift+X)
  3. Search for "Playbooks Language Support"
  4. Click Install

The extension provides debugging capabilities for playbooks programs, making it easier to develop and troubleshoot your AI agents. Once the plugin is installed, you can open a playbooks .pb file and start debugging!

📚 Documentation

Visit our documentation for comprehensive guides, tutorials, and reference materials.

License

This project is licensed under the MIT License - see the LICENSE file for details.

Contributors

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

playbooks-0.3.2.tar.gz (63.3 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

playbooks-0.3.2-py3-none-any.whl (77.6 kB view details)

Uploaded Python 3

File details

Details for the file playbooks-0.3.2.tar.gz.

File metadata

  • Download URL: playbooks-0.3.2.tar.gz
  • Upload date:
  • Size: 63.3 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: poetry/1.8.5 CPython/3.13.1 Darwin/24.3.0

File hashes

Hashes for playbooks-0.3.2.tar.gz
Algorithm Hash digest
SHA256 316f367ef75eb6fa5bfe20803d2c0a340a73d61e54d025a36c873bfa503c1b51
MD5 09f2a668328a921c33c37750830bd5dc
BLAKE2b-256 ed22a5ccad6c7005a8e932f8e366980e662bfb9cd1d828d7f5d5536138cc2e92

See more details on using hashes here.

File details

Details for the file playbooks-0.3.2-py3-none-any.whl.

File metadata

  • Download URL: playbooks-0.3.2-py3-none-any.whl
  • Upload date:
  • Size: 77.6 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: poetry/1.8.5 CPython/3.13.1 Darwin/24.3.0

File hashes

Hashes for playbooks-0.3.2-py3-none-any.whl
Algorithm Hash digest
SHA256 d789149090e10abe15b14935346c833f5746d2c11ea7a052d389732024c5a57c
MD5 deb61723c6dd6b43d8a8808e6a1c0720
BLAKE2b-256 3f2a73cbf37fccc98d57334258f93c872e914dd999ec28642e1d83e4180f6c81

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page