osFoundry / Platform
News reader
Pull every feed into one quiet reader. Summaries generated on demand.
What it is
News reader is the feed reader built into osFoundry, the bring-your-own-key desktop AI workspace. It pulls every source you follow into one quiet, unified reading surface, so instead of hopping between publisher sites, newsletters, and browser tabs, you open a single reader and see everything in one stream. It lives in the same workspace as your chat, agents, apps, code, and knowledge, which means the articles you read are already sitting next to the tools you use to act on them.
Because it runs inside a local-first desktop app, your feeds and reading history can stay on your device rather than on someone else's server. When you want an article distilled, summaries are generated on demand — you ask, and only then does the reader call a model. That summarization runs on your own API key with no reseller markup, so you pay the model provider directly at their rate; there's nothing forcing every headline through an LLM in the background. You choose which model handles the summary, whether that's a local model on your machine, a cloud provider, or something self-hosted.
The design goal is calm. Rather than an algorithmic timeline optimized for engagement, News reader gives you the plain list of what your sources published, in one place, with summaries available the moment you want to skim rather than read in full. There are no per-seat fees to add it to your workflow, and because it's part of osFoundry, a summary you generate can flow straight into a chat, an agent task, or your notes without leaving the workspace.
Key capabilities
- Aggregates every feed you follow into one unified, distraction-free reading view inside the desktop workspace
- Generates article summaries on demand — nothing is sent to a model until you ask for it
- Summaries run on your own API key with no reseller markup; you pay the model provider directly
- Local-first: feeds and reading history can stay on your device rather than a third-party server
- Choose the model that summarizes — local, cloud, or self-hosted — from the same BYOK workspace as your chat, agents, and notes
Use cases
- Analyst tracking an industry: follows a dozen publisher and newsletter feeds in one reader, then asks for on-demand summaries only on the articles worth a closer look
- Privacy-conscious researcher: keeps feeds and reading history local on the desktop app and points summarization at a local model so nothing leaves the machine
- Solo founder or small team: adds a calm news stream to the same workspace as their agents and notes without paying per-seat fees or a markup on summary calls
Frequently asked questions
Does every article get summarized automatically?
No. Summaries are generated on demand — the reader only calls a model when you ask for one. By default you get the plain feed of what your sources published, and you decide which articles are worth distilling.
What does it cost to summarize articles?
You bring your own API key, so summarization is billed directly by whatever model provider you use, at their rate. osFoundry adds no reseller markup on those calls, and there are no per-seat fees for the reader itself.
Where are my feeds and reading data stored?
osFoundry is a local-first desktop app, so your feeds and reading history can stay on your device rather than on a third-party server. You also choose which model handles summaries, including a local one, if you want reading to never leave your machine.