Get unread articles

Unread articles, newest first. Slots are shared across feeds so one high-volume source cannot fill the page; balance_feeds=false restores strict recency. since_hours narrows the window; feed_id scopes to one source; lean=true (limit up to 1000) fits the whole set in one call.

post/get_unread_articles

Request

  • Base URL: https://newsmind.app/api/v1
  • URL: https://newsmind.app/api/v1/get_unread_articles
  • Auth: OAuth 2 (authorization code, authorization URL https://newsmind.app/oauth/authorize, token URL https://newsmind.app/oauth/token, scopes: mcp)

Request body

limitinteger

Max articles to return (1–1000). Default 50; raise toward 1000 when clustering the whole unread set.

since_hoursnumber

Only articles published in the last N hours.

feed_idinteger

Limit to a specific feed.

exclude_feed_idsinteger[]

Exclude articles from these feeds. Useful when one or two high-volume feeds (HN, WIRED) drown out the rest.

leanboolean

Return a slim payload — id, feed_id, feed_title, title, url, published_at only — dropping summary, image_url/width/height/mime, and added_at. ~5x smaller per row, lets a single call carry hundreds of articles for AI-side topic clustering.

boost_engaged_feedsboolean

Re-rank the recent unread pool by per-feed engagement (how often the user reads/stars that feed) instead of pure recency. Feeds the user values rise; skipped feeds sink; recency breaks ties. Use for 'what should I read first' / 'the good stuff'. Default false (strict newest-first).

balance_feedsboolean

Share the page out across feeds that published recently, so one high-volume source cannot fill it. Default true. Set false for strict newest-first, e.g. when the user explicitly wants the raw timeline or is chasing a fast-moving story.

Response

Success

balancedobject

Present only when balancing changed the result.

Changes

No recorded changes to this operation across all 1 revision of this API.