How do I feed RSS feeds to an AI agent?

Feed RSS feeds to an AI agent by calling fetch_url on the feed URL — AgentFetch fetches the feed, strips the XML boilerplate, and returns the item titles, links, and summaries as clean markdown the model can read straight through. Example: fetch_url("https://news.ycombinator.com/rss") gives the model the current front-page items without handing it 25KB of raw XML. The agent can then pass the story links it picked to fetch_multiple and pull up to 20 of them in one parallel round-trip. This pattern is the backbone of automated news, monitoring, and competitive-intel agents — feed the agent a list of 10 RSS sources every morning and it produces a 200-word digest for $0.02-$0.05 in token cost. For sources without RSS (most modern sites), call fetch_url(listing_url, format="json") against the listing page and let the model pull the item list out of the returned structure. For high-frequency polling, lean on the six-hour cache rather than re-fetching: use_cache=True (the default) returns a cached copy inside that window, which saves both credits and publisher bandwidth. Pass use_cache=False only when freshness genuinely matters, such as breaking news or live prices.