# Copilot prompt:

# Write a Python script that connects to a Confluence space (cloud or server) using username/API token authentication.

# The script should:

# 1. Fetch the content of one or more Confluence pages (title and body).

# 2. Clean and preprocess the page data (strip HTML, handle newlines, remove extra formatting).

# 3. Convert the text into documents that can be used as context for an LLM using LangChain.

# 4. Use LangChain's Document and TextSplitter classes to chunk the page data into manageable pieces.

# 5. Optionally, demonstrate inserting embeddings into a vector database (like FAISS or Chroma).

# 6. Keep the code modular (functions for fetch_confluence_page, clean_text, prepare_langchain_docs).

#

# Dependencies: atlassian-python-api (or requests if not available), langchain, beautifulsoup4