arXiv Preprint Tracker — New Papers by Keyword/Category
Pricing
Pay per usage
arXiv Preprint Tracker — New Papers by Keyword/Category
Track new research preprints on arXiv by keyword and/or category (cs.AI, econ.GN, stat.ML, and more). Get title, authors, abstract, category, and PDF link the moment a paper is submitted.
Pricing
Pay per usage
Rating
0.0
(0)
Developer
Timothy Kelvin
Maintained by CommunityActor stats
0
Bookmarked
2
Total users
1
Monthly active users
5 days ago
Last modified
Categories
Share
Track new research preprints on arXiv by keyword and/or category. Get the title, authors, abstract, category, and PDF link the moment a paper is submitted.
Built for researchers, VCs tracking a technical trend, and anyone who wants to know what's new in a field without checking arXiv by hand every morning.
Input
{"keyword": "prompt injection","category": "cs.AI","daysBack": 7,"maxResults": 25}
| Field | Type | Description |
|---|---|---|
keyword | string (optional) | Free-text search across title and abstract. |
category | string (optional) | arXiv category code, e.g. "cs.AI", "cs.CL", "stat.ML", "econ.GN", "q-bio.GN". Full list at arxiv.org/category_taxonomy. |
daysBack | number | How many days back from today to search, by submission date. Default 7, max 90. |
maxResults | number | Max papers to return, most recently submitted first. Default 25, max 100. |
Leave both keyword and category blank to get the most recently
submitted papers across all of arXiv.
Output
One record per paper:
{"arxivId": "2608.12172v1","title": "Rethinking Agent Security as a Networking Problem","summary": "AI agents are rapidly becoming more capable and widely deployed...","authors": ["Van Tran", "Taveesh Sharma", "Tajveer Singh Dhesi", "Nick Feamster"],"primaryCategory": "cs.MA","categories": ["cs.MA"],"published": "2026-08-12T15:29:51Z","updated": "2026-08-12T15:29:51Z","comment": null,"journalRef": null,"doi": null,"absUrl": "http://arxiv.org/abs/2608.12172v1","pdfUrl": "https://arxiv.org/pdf/2608.12172v1"}
A search with no matching papers returns no items but is still billed once for the search.
How it works
Direct calls to the official arXiv API
(export.arxiv.org) — no proxy, no key, no scraping. Free and open,
arXiv explicitly does not require API key registration.
Pricing note
Billed per search, not per paper returned — one charge whether the search returns 0 papers or 100.
Related products
- PubMed Research Alert — the biomedical/life-science equivalent, via NCBI E-utilities instead of arXiv