To continuously download dbt docs from an automated CI/CD pipeline (like GitHub Actions or GitLab CI), use the dbt Cloud Administrative API. Bash Automation Script:
Navigate to your project directory and run the compile command to build the latest manifest and catalog: dbt docs generate Use code with caution. Step 2: Locate the Files how to download dbt docs
Copy these three files to your desired local folder or archive them into a zip file for distribution: zip dbt_docs.zip index.html manifest.json catalog.json Use code with caution. Method 2: Download dbt Docs from dbt Cloud To continuously download dbt docs from an automated
import json import os with open("index.html", "r") as f: html = f.read() with open("manifest.json", "r") as f: manifest = json.load(f) with open("catalog.json", "r") as f: catalog = json.load(f) # Search and replace the payload markers html = html.replace('window.gc_manifest = null', f'window.gc_manifest = json.dumps(manifest)') html = html.replace('window.gc_catalog = null', f'window.gc_catalog = json.dumps(catalog)') with open("local_dbt_docs.html", "w") as f: f.write(html) Use code with caution. Run the script: python bundle_docs.py Use code with caution. Method 2: Download dbt Docs from dbt Cloud
By default, opening the downloaded index.html locally will result in a blank page or a loading spinner due to browser CORS (Cross-Origin Resource Sharing) restrictions. You can merge the three files into one single, portable file to bypass this. Option A: Using the community Python script
If your team uses dbt Cloud, the documentation is hosted automatically, but you can still download the raw assets for offline use.
The structural metadata from your data warehouse, including column types and table sizes.