Skip to content

breba-apps/breba-docs

Repository files navigation

Breba Docs · PyPI version

AI documentation validator

workflow

Features

Scans your documentation file and executes commands in the documentation to make sure that it is possible to follow the documentation.

Getting Started

Prerequisites

Docker engine needs to be installed and running. Use docker installation instructions for your system.

Get an OpenAI API Key and set environment variable like this:

export OPENAI_API_KEY=[your_open_ai_api_key]

Install and Run

To install and run breba-docs, run the following commands:

pip install breba-docs
breba-docs new sample_proj
cd sample_proj
breba-docs run

Then you will need to provide location of a documentation file. For example: sample_project/sample.md

The software will then analyze the documentation and run the commands found in the documentation inside a docker container with python installed.

The AI will then provide feedback regarding how it was able to follow the instructions.

Features

  • Create a New Project:
    Run breba-docs new to set up a new project with:

    • data directory
    • prompts directory
    • config.yaml file containing project and model configuration
  • Run the Project:
    Run breba-docs run to load and display your project's configuration.

    When providing a path to file, that file will be copied to the data directory.

    When using a github url, the repo will be cloned inside the data directory.

    The software will then analyze the documentation and run the commands found in the documentation

    Commands will be run inside a docker container with python installed.

    Commands to fix the documentation will run against the documentation in the data directory.

Contributing

For contributing to the project, please refer to Contribution Guide.