Financial news alerts

1. 01List each outlet's business section

   Each news API lists a section in one request. Reuters takes a section path such as business, AP a section such as business, and Bloomberg a rail such as markets. Results have the title, URL, and publish time; Bloomberg and Reuters add the section and authors.

   GET /v1/search?engine=bloomberg_news&section=latest · results.0

   {
     "position": 1,
     "title": "US, Iran Said to Be Exploring Phased Deal to Open Hormuz",
     "url": "https://www.bloomberg.com/news/articles/2026-09-24/us-iran-said-to-be-exploring-phased-deal-to-open-hormuz",
     "section": "politics",
     "kicker": "Middle East",
     "authors": [
       {
         "name": "Fiona MacDonald",
         "url": null
       }
     ],
     "premium": false
   }

2. 02Match the watchlist

   Compare each title with your company names. Headlines use short names, such as Apollo for Apollo Global Management, so list the names the press uses.

   GET /v1/search?engine=reuters_news&section=world · results.0

   {
     "position": 1,
     "id": "ZYK6TDJ2VBKEHICD7UGXKGPR3Q",
     "title": "Trump warns in UN speech he could 'annihilate' Iran without peace deal",
     "url": "https://www.reuters.com/world/americas/trump-set-whirlwind-un-meetings-with-wars-iran-ukraine-agenda-2026-09-22/",
     "published_at": "2026-09-22T05:04:00.006Z",
     "section": "World",
     "kicker": "Americas",
     "authors": [
       {
         "name": "Gram Slattery",
         "url": "https://www.reuters.com/authors/gram-slattery/"
       }
     ],
     "content_type": "article",
     "premium": false
   }

3. 03Read each match

   The article endpoint returns the paragraphs, authors, publish time, and tags. On a metered Bloomberg or Reuters story only the public paragraphs come back and body_truncated is true; the headline and lead are usually enough for an alert.

   GET /v1/article?engine=bloomberg_news&url=https%3A%2F%2Fwww.bloomberg.com%2Fnews%2Farticles%2F2026-09-22%2Fapollo-caps-private-credit-fund-again-after-14-7-look-to-exit · article

   {
     "id": "TLGPOGKK3NYH00",
     "title": "Apollo Caps Private Credit Fund Again After 14.7% Look to Exit",
     "description": "Apollo Global Management Inc. is limiting redemptions from a private credit fund for the third straight quarter, as its investors join the rush to pull cash from the $1.8 trillion direct lending market.",
     "url": "https://www.bloomberg.com/news/articles/2026-09-22/apollo-caps-private-credit-fund-again-after-14-7-look-to-exit",
     "language": "en",
     "authors": [
       {
         "name": "Olivia Fishlow",
         "url": "https://www.bloomberg.com/authors/AXMJhR-IRl0/olivia-fishlow"
       }
     ],
     "published_at": "2026-09-22T21:00:42.020Z",
     "updated_at": "2026-09-22T21:00:42.020Z",
     "section": "markets",
     "subsection": null,
     "tags": [
       "Private Credit",
       "Software",
       "Private Market"
     ],
     "image": {
       "url": "https://assets.bwbx.io/images/users/iqjWHBFdfxIU/iXBnBEqKkiAc/v1/-1x-1.webp",
       "caption": "Apollo Global Management signage in New York.",
       "credit": "Jeenah Moon/Bloomberg"
     },
     "highlights": [],
     "paragraphs": [
       "Apollo Global Management Inc. is limiting redemptions from a private credit fund for the third straight quarter, as its investors join the rush to pull cash from the $1.8 trillion direct lending market.",
       "Apollo Debt Solutions BDC, which has about $26 billion in assets, told shareholders that it would again cap withdrawals at 5% of outstanding shares on Tuesday after 14.7% sought to pull their cash, ..."
     ],
     "word_count": 86,
     "reading_time_minutes": 1,
     "access": "metered",
     "body_truncated": true,
     "related": [
       {
         "title": "Murky Default Metrics",
         "url": "https://www.bloomberg.com/news/articles/2026-09-17/private-credit-defaults-are-1-6-or-19-depending-who-you-ask"
       }
     ]
   }

# Every 15 minutes: new Reuters, AP, and Bloomberg business headlines that
# name a company on the watchlist, with the article text for each match.
import json, os, pathlib, requests

API = "https://api.clair.im"
HEADERS = {"Authorization": f"Bearer {os.environ['CLAIR_API_KEY']}"}
WATCHLIST = ["Nvidia", "Apollo", "Stripe", "OpenAI"]
LISTINGS = [("reuters_news", {"section": "business"}),
            ("ap_news", {"section": "business"}),
            ("bloomberg_news", {"section": "markets"})]
STATE = pathlib.Path("seen.json")
seen = set(json.loads(STATE.read_text())) if STATE.exists() else set()

def get(path, **params):
    r = requests.get(f"{API}{path}", headers=HEADERS, timeout=60, params=params)
    r.raise_for_status()
    return r.json()

for engine, params in LISTINGS:
    for item in get("/v1/search", engine=engine, **params)["results"]:
        names = [n for n in WATCHLIST if n.lower() in item["title"].lower()]
        if not names or item["url"] in seen:
            continue
        seen.add(item["url"])
        article = get("/v1/article", engine=engine, url=item["url"])["article"]
        lead = " ".join(article["paragraphs"][:2])
        cut = " (truncated)" if article["body_truncated"] else ""
        print(f"{', '.join(names)} | {engine}: {article['title']}{cut}\n  {lead[:300]}")

STATE.write_text(json.dumps(sorted(seen)))

Set CLAIR_API_KEY to a key subscribed to the Reuters News API and Associated Press API and Bloomberg News API. Each call is one request against that API's monthly quota.

• Each news API is its own subscription with its own quota. Start with the outlet your readers rely on most.
• Listings are one page of the section's latest stories. Poll often enough that a busy section does not scroll a story off between runs.
• Reuters blocks site search, so match on section listings. AP and Bloomberg also accept q for keyword search.
• This is the public website, not a licensed wire feed: no embargoed or subscriber-only stories, and not for republishing.