Load Gitbook documentation into LlamaIndex for RAG
Load and convert content from a GitBook space into LlamaIndex documents.
Why it matters
Extract and convert documentation content from Gitbook spaces into structured documents that can be indexed and queried by LlamaIndex for retrieval-augmented generation workflows.
Outcomes
What it gets done
Connect to Gitbook API using authentication token
Fetch all content from a specified Gitbook space by ID
Convert Gitbook pages into LlamaIndex document format
Attach metadata fields like path, title, description, and parent to documents
Install
Add it to your toolbox
Run in your project directory:
curl -fsSL https://spark.entire.vc/get/li-reader-readers-gitbook | bash Overview
LlamaIndex Readers Integration: Gitbook
A LlamaIndex reader that loads content from a GitBook space via its API and converts it into documents. Use to index a GitBook-hosted documentation space for search or question-answering.
What it does
The Simple Gitbook Reader loads data from a GitBook space, collecting and converting its content into documents usable by LlamaIndex. It gives a straightforward path from a documentation site hosted on GitBook to an indexable set of documents, without a custom scraper.
SimpleGitbookReader is initialized with a GitBook API token, and load_data takes the target space_id plus an optional metadata_names list controlling which fields get attached as document metadata - available options are path, title, description, and parent, letting the resulting documents carry structural context about where each piece of content sits within the GitBook space.
When to use - and when NOT to
Use it when you need to index the contents of a GitBook-hosted documentation space into LlamaIndex - for example to build a search or Q&A system over product docs. Use metadata_names when downstream retrieval or filtering needs to know a document's path, title, description, or parent within the space's structure, not just its text. Do not use it for content outside GitBook; this reader is scoped specifically to a GitBook space via its API.
Capabilities
load_data fetches all content from a given GitBook space ID and converts it into LlamaIndex documents, optionally attaching path, title, description, or parent metadata per document.
How to install
pip install llama-index-readers-gitbook
Requires a GitBook API token.
Who it's for
Developers who maintain documentation on GitBook and want its content indexed into LlamaIndex for search or question-answering, with optional structural metadata preserved.
Source README
LlamaIndex Readers Integration: Gitbook
Overview
Simple Gitbook Reader allows loading data from a gitbook space. It collects & converts contents from gitbook space into documents used by LlamaIndex.
Installation
You can install Gitbook Reader via pip:
pip install llama-index-readers-gitbook
Usage
from llama_index.readers.gitbook import SimpleGitbookReader
### Initialize SimpleGitbookReader
reader = SimpleGitbookReader(
api_token="<Gitbook API Token>", # Gitbook API Token
)
### load data from Gitbook
documents = reader.load_data(
space_id="<Gitbook Space Id>", # Id of the gitbook space
metadata_names=None, # Names of the fields to add to metadata attribute (available: 'path', 'title', 'description', 'parent')
)
FAQ
Common questions
Discussion
Questions & comments · 0
Sign In Sign in to leave a comment.