SNEWPapers launches remote MCP turning AI tools into US newspaper archive researchers
Jolly-Newspaper-6769 · reddit · 2026-11-17
SNEWPapers' founder released a remote MCP server that turns AI tools into archive researchers for US historical newspapers (1736–1963) from the Library of Congress Chronicling America collection. Each issue is segmented into individual items (news, obituaries, notices, letters), labeled, and linked back to original page scans. As of November 17, 2026 the index holds 6,988,773 extracted components across 69,762 issues and 2,901 titles — about 2.5% of English-language issues, growing daily, with hundreds of terabytes of scans still to process (20-page-or-fewer English issues first). The connector is approved in Claude's directory and can be added as a custom connector elsewhere. The founder claims best-in-class segmentation and text extraction accuracy, with search returning single readable components linked across pages, and offers free access to the first 100 Reddit users.
More from coding & agent
- REA-Video skill lets Claude reverse-engineer video editing styles; rival claims 20x speed at SSIM 0.96 — toolstelegraph · 2026-11-17
- Developer retries Amp Code after months and finally gets the hype: "completely different" — g3off_who · 2026-11-17
- Dev shares 5 AI automations for service businesses that actually make owners money — Crazy-Movie7633 · 2026-11-17
- Team designing open-source mock MCP server, asks community for testing pain points first — OkSkirt1039 · 2026-11-17
- Gemini 4 Hydrogen adds xhigh thinking level, with official guidance on all four tiers — patloeber · 2026-11-17
- Dev publishes live agent key with ¥100/day cap and challenges the internet to break it — kokesi12 · 2026-11-17