Loading repository data…
jaydayal-ai / repository
Command-line tool that downloads news article images from the newsdata.io API and assembles a daily collage PNG plus an HTML gallery.
A small command-line tool that pulls recent headlines from the newsdata.io news API (https://newsdata.io), downloads the article images, and stitches them into a single daily collage PNG plus a self-contained HTML gallery.
The gallery lists each headline with its source and any keywords the API returns, and — when your plan provides them — a sentiment badge, AI tags, and a short AI summary. Grab a free API key at https://newsdata.io to run it.
/latest endpoint, following
the nextPage cursor to gather enough articles.index.html: a dark gallery with thumbnails, sources, keyword chips,
and a sentiment breakdown chart.sentiment, ai_tag, and ai_summary from each article when they are
present and falls back to clearly labeled placeholders otherwise, so it runs
unchanged on a free key.Requires Python 3.9 or newer.
git clone <this repo>
cd news-image-collager
pip install -r requirements.txt
Set your newsdata.io API key (free keys work):
export NEWSDATA_API_KEY="your_key_here"
python main.py # latest English headlines
python main.py -q "climate" -n 16 # 16 images about climate
python main.py -c technology --country us --cols 5
Options:
-q, --query keyword to search for
-c, --category category, e.g. technology, sports, business
-l, --language language code (default: en)
--country 2-letter country code, e.g. us, gb
-n, --num number of images in the collage (default: 12)
--cols collage columns (default: 4)
-o, --outdir output directory (default: output)
Output lands in output/:
output/
collage-2026-06-11.png
index.html
Open output/index.html in a browser to see the gallery.
$ export NEWSDATA_API_KEY=pub_xxx
$ python main.py -q "space" -n 9 --cols 3
Fetching headlines from newsdata.io ...
Got 18 articles, 12 with images.
Wrote collage -> output/collage-2026-06-11.png
Wrote gallery -> output/index.html
Open output/index.html in a browser.
The sentiment, ai_tag, ai_org, ai_region, and ai_summary values are only
returned on paid newsdata.io plans. The tool reads them automatically when they
are present; on a free key those areas show a labeled placeholder and the
sentiment chart shows a sample distribution so the layout stays intact.
If you have a paid plan and want larger result pages, set:
export NEWSDATA_PAID=1
This lets --num request more than 10 articles per page (size greater than 10
is a paid-only parameter). If a paid-only request is rejected, the client retries
once without those parameters so you still get results.
NEWSDATA_API_KEY and is never written to disk.MIT
Selected from shared topics, language and repository description—not editorial ratings.
payaltandon-xyz /
Command-line tool that downloads news article images from the newsdata.io API and builds a daily collage PNG plus a browsable HTML gallery.