What just got announced, and who announced it.
54 Press Release Wires tools behind one URL.
Frontier models: 100% correct with Pipeworx, 63% with web search
Six named wires, each its own pack, searchable in one call instead of six separate feeds. Business Wire covers its all-news channel (about a week of the whole wire), GlobeNewswire its public-company channel, PR Newswire its full site-wide feed, PRWeb Cision's small-business wire, EIN Presswire its public news sitemap, Newswire.com its public newsroom — headline, issuer and publish time back as data, not a page to parse. None is a full archive or the whole wire: each reads a live rolling window, and each pack says how big that window is. US public companies already put material news in SEC 8-Ks, which Pipeworx also covers; what these wires add is everything that isn't a filing — private-company announcements, routine news, and the minutes before the filing catches up.
Included Packs
Public-company releases — the wire's own feed description calls it “the last 20 releases”. Headline, issuer, tickers, publish time and excerpt, from globenewswire.com directly
Site-wide “All News Releases” feed — every category, the ~20 most recent releases. Headline, issuer, industry tags, publish time and excerpt, from prnewswire.com directly
Public news sitemap, not RSS (EIN Presswire's RSS now requires a free account). Headline, keyword tags, geo location and publish time, covering roughly the trailing 48 hours across every category
The all-news channel — roughly the last seven days of the whole wire (~3,500 releases), searchable in one call. Headline, dateline, publish time, excerpt and release id, from feed.businesswire.com directly
Cision's small-business wire — site-wide “All News Releases” feed, the ~20 most recent. Headline, issuer, industry and subject tags, publish time and excerpt, from prweb.com directly
Public newsroom feed, the 50 most recent releases. An Issuer Direct wire that shares many releases with ACCESS Newswire. Headline, publish time and sub-headline excerpt, from newswire.com directly
No historical archive on any of the six — every call re-reads the source's own live window. No ACCESS Newswire pack: its feeds and article pages sit behind a Cloudflare challenge, though Newswire.com (same parent company) carries a large share of the same releases.
Setup
Choose one of these options:
Open claude.ai/customize/connectors, click Add custom connector, name it Pipeworx Press Releases and paste this URL. Leave the OAuth client ID and secret blank — the connector registers itself.
https://gateway.pipeworx.io/mcp?vertical=press-releasesThen click Connect and sign in with GitHub. That step is free and takes you from 50 calls a day to 200 — same connection either way, and you can skip it and sign in later. In a new chat, open the tools menu and enable the connector.
Custom connectors need a paid Claude plan — that is an Anthropic gate, not a Pipeworx one.
Add this to your claude_desktop_config.json:
{
"mcpServers": {
"pipeworx-press-releases": {
"command": "npx",
"args": [
"-y",
"mcp-remote",
"https://gateway.pipeworx.io/mcp?vertical=press-releases",
"--header",
"Authorization:Bearer YOUR_API_KEY"
]
}
}
}Paste this into Claude to use for one conversation:
Use this MCP server for this conversation:
npx -y mcp-remote "https://gateway.pipeworx.io/mcp?vertical=press-releases" --header "Authorization:Bearer YOUR_API_KEY"
Then tell me the latest press releases from GlobeNewswire's public-company feed and PR Newswire's site-wide feed, and search EIN Presswire for anything published about healthcare today.All six packs are keyless — no account, no clickthrough terms, nothing to configure.
What each source is good for
Latest releases
globenewswire_latest_releasesNewest public-company releases, with a since_hours filterprnewswire_latest_releasesNewest site-wide releases across every category, with a since_hours filtereinpresswire_latest_releasesNewest releases across the trailing ~48 hours, with a since_hours filterSearch by company or keyword
globenewswire_search_releasesMatches headline, issuer and excerpt in the current windowprnewswire_search_releasesMatches headline, issuer, excerpt and industry tags in the current windoweinpresswire_search_releasesMatches headline, keyword tags and geo location in the current windowOne release
globenewswire_get_releaseFetches one release by URL from the current windowprnewswire_get_releaseFetches one release by URL from the current windoweinpresswire_get_releaseFetches one release by URL from the current windowCommon questions
Which wires does this cover?
Six, each its own pack rather than bundled behind a publisher argument: Business Wire's all-news channel, GlobeNewswire's public-company feed, PR Newswire's site-wide feed, PRWeb's site-wide feed (Cision's small-business wire), EIN Presswire's news sitemap, and Newswire.com's public newsroom feed. ACCESS Newswire has no pack: its feeds and article pages sit behind a Cloudflare challenge, but Newswire.com (same parent company, Issuer Direct) carries a large share of the same releases. More wires get added as their own packs, not folded into these six.
Is this the full archive of each wire?
No — each pack re-reads a live, rolling window on every call, not a historical archive. Business Wire's all-news channel is the widest: roughly the last seven days of the whole wire, about 3,500 releases, searchable in one call. GlobeNewswire's, PR Newswire's and PRWeb's feeds hold roughly their last 20 releases (public-company-only for GlobeNewswire; every category for the other two), Newswire.com's its last 50, all refreshed continuously. EIN Presswire's sitemap covers roughly the trailing 48 hours across every category. A search only ever searches what is currently in that window.
Why does EIN Presswire read a sitemap instead of an RSS feed?
Its RSS feeds turned out to require a free account — every category link on einpresswire.com/all-rss redirects to a login page. Logging in to scrape is out of scope under our standing sourcing rule, so this pack instead reads the public Google-News-format sitemap EIN Presswire lists in its own robots.txt for crawling: no account, denser (roughly 700 releases across ~48 hours), but it carries a headline, timestamp, keyword tags and geo location — no body excerpt and no separate issuer field the way an RSS item would.
Do I need API keys for this vertical?
No. All six packs are keyless — no account, no clickthrough terms, no rate-limited vendor key to manage.
Can I fetch the full text of one release?
Each pack's get_release tool re-fetches the live feed and returns the matched item's headline, issuer (where the source exposes one) and the excerpt the feed itself carries — not a separate full-article fetch. If the release has scrolled out of the source's current window, get_release says so rather than returning nothing silently.
How is this different from an SEC 8-K filing?
US public companies already have to file material news as an 8-K, and Pipeworx covers EDGAR directly — that overlap is real. What these wires add is everything that never becomes a filing: private-company announcements, product news, partnerships, awards and routine updates that don't meet 8-K materiality, plus speed — a release usually goes out before, sometimes well before, the filing that covers the same event.