Streamlit application index maintenance

The public directory at /streamlit-apps/ is generated by scripts/update_streamlit_index.py and refreshed automatically each Monday at 06:17 UTC. It can also be started manually from the repository’s Actions tab using Update Streamlit application index.

What is automated

The updater scans public, non-fork repositories in the mboudour GitHub account. A repository is considered a Streamlit project only when it contains implementation or dependency evidence (for example, a Python import of Streamlit or a Streamlit entry in a dependency file). It then lists only deployments that respond successfully.

When a live app is found for the first time, the updater adds it to _data/streamlit_apps.json with:

• section: unclassified
• a neutral placeholder title derived from the repository name
• a neutral placeholder description

This prevents automatic discovery from making an unsupported subject classification.

How to classify a new application

1. Open _data/streamlit_apps.json in GitHub.
2. Find the new repository under apps.
3. Change its section from unclassified to an existing section identifier, for example:
   • research-methods-ai
   • statistics-networks-scientometrics
4. Replace the placeholder title and description with the approved public wording.
5. Commit the edit. GitHub Pages will publish the revised placement automatically.

Adding a new section

Add an object to the top-level sections list, with an id and title, then use that id in the relevant application records. The generated directory follows the order in this list.

Important boundary

A repository without a verifiable live Streamlit deployment is not added to the public directory. This includes a project that has Streamlit code but has not yet been deployed, and a project whose deployment is temporarily unavailable.