Indy News

The API used by the Indy News assistant.

It uses a selection of trusted news sources to retrieve their publications for a time window and a potential topic of interest.

Search for relevant independent media outlets

• /media

Search for relevant youtube videos

• /youtube

Search for relevant X tweets

• /x

Prerequisites: Please first log in to X and log your cookies an place them in .env under X_COOKIE. It should look like this:

X_COOKIES='  lang=en; guest_id=123; night_mode=2; _twitter_sess=ABC--DEF; d_prefs=GJHHJVS; guest_id_ads=v1%3A1223423434; guest_id_marketing=v1%3A23423423423; personalization_id="v1_dasdiuhjKIvhhb=="; kdt=KJBNDFSbfsdksLKSDFJSDF; auth_token=sfsefsdfs234n2bn34j23kj4n2k; ct0=jsdfkdnksdndajsk21323kjn42kj; att=1kjfnaskNknlnlsfsaldkf; lang=en; twid=u%3D1342342342'

Check the ratings DBs we used

It also contains some endpoints to see where the ratings come from:

• /allsides
• /mediabiasfactcheck

Databases used (see data/ folder)

• all.csv: our source selection of media outlets
• allsides.com.json: snapshot of the AllSides db
• mediabiasfactcheck.com.json: snapshot of the MediaBiasFactCheck db

Dev instructions

1. Install and activate .venv and pip install -r requirements.txt

2. Run

• FastAPI: .venv/bin/uvicorn --host "0.0.0.0" -p 8088
• StreamLit .venv/bin/streamlit run Home.py