Apify news monitoring

News Monitoring: Brand and Topic Watch From Google News

Turn Google News into a monitoring feed. Poll one query per watchlist term on a schedule, dedupe what you have already seen, and deliver only the new articles as alerts, a digest, or rows in a database.

When to use this skill

• The user wants news monitoring or media monitoring for a brand, product, executive, or competitor.
• They want brand mentions from news sources as a feed, digest, or alert.
• They want to track a market, industry, regulation, or crisis as it develops.
• They ask for "google news tracking", "google news monitoring", or "news alerts by API".

Not for: one-off bulk exports of a single search (use the google-news-api skill), full article text (the feed carries title, snippet, link, source, and date), or social media mentions (news outlets only).

What each poll returns (one dataset item per page)

• news_results: articles with position, title, link, source, snippet, date (relative, such as "2 hours ago")
• search_metadata: total_results, news_count, pages_processed, pagination_limit_reached
• search_information: query_displayed, news_results_state
• search_parameters, search_timestamp, page_number

link is the dedupe key; date plus search_timestamp is the freshness signal.

Prerequisites

• Apify account (sign up at https://apify.com?fpr=9n7kx3&fp_sid=skillrepo).
• Authentication via apify login, or an APIFY_TOKEN environment variable (Apify Console, Settings, Integrations).

The Actor

• Store page: https://apify.com/johnvc/GoogleNewsAPI?fpr=9n7kx3&fp_sid=skillrepo
• Actor ID: johnvc/GoogleNewsAPI
• Pricing: pay per page of results processed, plus a small per-run setup fee (see references/gotchas.md).

Run it with the Apify CLI

One watchlist poll, bounded to one page with duplicate coverage collapsed:

apify actors call "johnvc/GoogleNewsAPI" -i '{"q":"Acme Robotics","filter":"1","max_pages":1}' \
  --json \
  --user-agent apify-awesome-skills/apify-news-monitoring \
  2>/dev/null

Read a run's items later, for example from a scheduled run:

apify datasets get-items <DATASET_ID> --format json --user-agent apify-awesome-skills/apify-news-monitoring 2>/dev/null

Every call carries the three flags this repo expects: --json (or --format json), --user-agent apify-awesome-skills/apify-news-monitoring, and 2>/dev/null.

Run it from Claude or another AI agent (MCP)

The Actor is MCP-ready. Add the hosted server URL:

https://mcp.apify.com/?tools=actors,docs,johnvc/GoogleNewsAPI

Then ask, for example: "Check Google News for new coverage of our company since yesterday and summarize anything negative." MCP setup docs: https://docs.apify.com/platform/integrations/mcp

Workflow

1. Define the watchlist. One query per term: brand names, product names, executives, competitors, or topics. Quote exact phrases inside q for precise brand matching. Each term is its own Actor run.
2. Bound each poll. Keep max_pages at 1 or 2 per term; new coverage almost always surfaces on page one. Set filter: "1" so one story does not arrive once per outlet.
3. Localize per market. Use gl, hl (base codes), and google_domain for non-US watch terms; add location for regional brands.
4. Estimate cost per poll and per month, then confirm with the user. See references/gotchas.md.
5. Poll, then post-process: flatten news_results, dedupe against your seen-links store on link, and convert relative date strings ("2 hours ago") to timestamps using search_timestamp as the reference.
6. Deliver only what is new: an alert per article, a daily digest grouped by term, or rows appended to a database. For recurring polls, wrap the same input in an Apify Schedule or a cron job.

Inputs

• q (string, required): one watchlist term per run
• filter (enum string "0" or "1"): collapse duplicate coverage, recommended "1" for monitoring
• gl, hl, google_domain, location: per-market localization
• safe (enum: active or off), nfpr (enum string): content and autocorrect controls
• max_pages (integer, default 1): the per-poll cost bound

Cost

Each poll costs a per-run setup fee plus a per-page fee: a one-page poll is about three cents. A daily one-page poll on one term is roughly a dollar a month; a five-term watchlist polled daily is roughly four to five dollars a month. Estimate with the live prices in references/gotchas.md before scheduling.

Honest limits

• This is a polled feed, not a push API: freshness is your poll interval.
• date is a relative string; the workflow converts it client-side. There is no server-side date-range filter, so dedupe by link does the real work.
• No article body or author; fetch full text downstream from link when an alert needs context.
• One query per run: a watchlist of N terms is N runs per poll, each with its own setup fee.
• Some polls legitimately return nothing new; quiet brands have quiet news cycles.

Troubleshooting

• Same story arrives repeatedly: dedupe on link across polls; within a poll, set filter: "1".
• Too much noise for a generic brand name: quote the exact phrase in q and add a qualifier, such as "Acme Robotics" funding.
• Nothing new for days: normal for narrow terms; widen the term or slow the poll instead of raising max_pages.
• Missed a story you saw on Google News: inventory and ranking vary by gl and google_domain; align them with the market where the story ran.

See references/gotchas.md for cost guardrails and error recovery, and references/actor-index.md for the Actor routing table.

Related monitoring Actors

• Google Forums Search API, community chatter about the same brand: https://apify.com/johnvc/google-forums-search-api?fpr=9n7kx3&fp_sid=skillrepo
• Google AI Overview API, how AI answers describe the brand: https://apify.com/johnvc/Google-AI-Overview-API?fpr=9n7kx3&fp_sid=skillrepo
• Google Finance API, market data for public companies on the watchlist: https://apify.com/johnvc/google-finance-api?fpr=9n7kx3&fp_sid=skillrepo