# Find what happened recently

> Find what happened recently: 3 companies fully do it, 18 with a limit, 60 carry the tag. Every listing in Web search APIs with the reason and a source.

- Canonical: https://www.anchorterminal.com/niches/search.recent-news/
- Markdown: https://www.anchorterminal.com/niches/search.recent-news/index.md (~1,750 tokens)
- Slim: https://www.anchorterminal.com/niches/search.recent-news/index.min.md (~430 tokens, same facts, less prose, for token-sensitive contexts)
- JSON: https://www.anchorterminal.com/niches/search.recent-news/index.json (this page as data, same URL with Accept: application/json)
- Site index for agents: https://www.anchorterminal.com/llms.txt (the key docs and every listing in one file, at most 16 MB, about 4 million tokens: https://www.anchorterminal.com/llms-full.txt)
- API: https://www.anchorterminal.com/api/v1/index.json
- MCP: https://www.anchorterminal.com/mcp (the whole catalogue as tools, no key)
- DNS: dig +short TXT <slug>.grade.anchorterminal.com (the grade as a DNS TXT record, for a sandbox with no HTTP where names still resolve; also top., best.<category>. and <a>.vs.<b>. under the zone, and the same under grade.letme.dev)
- Licence: CC BY 4.0, credit Anchor Terminal and link this page (https://www.anchorterminal.com/terms/#data)
- Updated: 2026-11-15

Two or three companies' docs show a full fit.

| Companies, full fit | With a limit | Carry the tag | Share of the category's work | First-hand accounts |
| --- | --- | --- | --- | --- |
| 3 | 18 | 60 | 8 | 5 |

## What counts as done

Results filtered to the asked publication window, with the date returned per result.

Measured, for our own tests: every result's publication date falls inside the window, and nothing older than the window appears in the top five.

How agents ask for it: what happened today in X; news about X this week; current price of X; who won X 2026

## Does the job

The docs show it meets the test above, and nothing we read says otherwise.

- **NewsCatcher (NewsCatcher, Inc.)**: search_window (default 7d) on publication or crawl date, published_at per article, sort_by published_at, and languages, countries and sources filters.
- **Perplexity Search API (Perplexity AI, Inc.)**: search_recency_filter hour to year, search_after_date and last_updated filters, with date and last_updated returned per result.
- **TinyFish Search (Tiny Fish, Inc.)**: domain_type news returns publisher and date per result, with recency_minutes or after_date and before_date windows.

## Does the job, with a limit

It does the job, but something in the docs or a user's account will get in an agent's way. The limit is named in each row.

- **Web Search Tool on Amazon Bedrock AgentCore (Amazon Web Services, Inc. (regional AWS entities by account location))**: filters.publishedDateFilter from and to bounds (connector 1.2.0 and later) and publishedDate per result, but no live fetch, no news vertical, and the filter applies to web results only.
- **Grounding with Bing Search (Microsoft Corporation)**: search_configurations take freshness (Day, Week, Month) and market, but no publication date is returned per citation.
- **Brave Search API + MCP (Brave Software, Inc.)**: freshness pd, pw, pm, py or a date range filters by page age, and a separate News Search service exists, but the Web Search page read shows no per-result date field.
- **Exa API + MCP (Exa Labs Inc.)**: startPublishedDate, endPublishedDate, category news and maxAgeHours for a live fetch, but the pages read do not confirm a date field on each result.
- **Kagi Search API (Kagi Inc.)**: after, before and relative interval filters on pages updated or published, and a news search type exists, but no per-result date field was seen in the schema read.
- **Liner Search MCP (Liner)**: date_range past_day to past_year and a date per result, but no live fetch or news mode.
- **Linkup (Linkup Technologies SAS)**: fromDate and toDate plus a date per result, but no live fetch or news mode.
- **Mojeek Web Search API (Mojeek Limited)**: since and before timespans, a last-modified date per result when date=1, and datewr to rank by date, but no live fetch and no news vertical in the API read.
- **Parallel Search and Task APIs (Parallel Web Systems Inc.)**: source_policy after_date and fetch_policy max_age_seconds for live content, but no per-result date field was confirmed in the pages read.
- **SearchAPI.io (SearchApi, LLC)**: time_period and time_period_min filter by date, but no per-result date field was confirmed in the Google schema.
- **SerpApi (SerpApi, LLC)**: tbs date filters and tbm=nws, with no_cache=true to skip the one-hour cache, but no per-result date field is shown in the reference read.
- **Serper (Serper)**: A news vertical returns relative dates such as '2 weeks ago' and tbs qdr:d filters by day, per third-party docs; no absolute date field.
- **SERPHouse (Patoliya Infotech LLP)**: A Google News API is listed, but no date filter or per-result date appeared on the pages read.
- **SerpKite (Anastasia Klimova, trading as SerpKite)**: /v1/news with time, tbs, start_date, end_date and max_age for cache age, but no per-result date field was confirmed.
- **SpecStream (SpecStream SAS)**: Sources carry published when the result gives one, but no freshness parameter is documented.
- **Tavily API + MCP (AlphaAI Technologies Inc. dba Tavily)**: topic news adds published_date, time_range and start_date filter by publish date, and max_age_hours forces live fetches, but topic news vanished from the MCP schema in February 2026 and broke callers.
- **Valyu (Valyu.Network Ltd)**: start_date and end_date restrict the publication window and search_type news exists, but no live fetch and the per-result date field was not confirmed.
- **You.com APIs (SuSea, Inc., d/b/a you.com)**: freshness day, week, month, year or a range with page_age per result and news blended by intent, but no news-only mode on the search endpoint.

## Couldn't check

The docs were closed to us, or silent. These carry no mark either way.

- **DeepSeek Search and Fetch API (Hangzhou DeepSeek Artificial Intelligence Co., Ltd.)**: Date operators are described in the listing's notes, but the docs were not opened and no account covers recency.

- not this job: Arcmira (Arcmira, Inc.): after and before date windows exist, but the corpus is spoken media and the Free plan withholds the newest 30 days.
- not this job: Cloudflare Web Search API (Cloudflare, Inc.): No date or freshness parameter, and accounts report 2024 links for a current stock price and 2024 NFL results for a 2026 query.
- not this job: Fetch (MCP reference server) (Model Context Protocol, a Series of LF Projects, LLC): No search.
- not this job: Grep AI (Parcha Labs, Inc.): No search with a date window; runs research a question.

## What stops agents finishing

- index stale with no date filter (Ceramic through Cloudflare)
- date filter matches publish date, not update date, so a frequently updated page is missed
- topic news removed from a tool schema without notice

## Around it

After: search.fetch-url, notifications.send.

## Where this comes from

Weight 8 of 100 in Web search APIs: 5 of 44 accounts: Ceramic returning 2024 results for 2026 queries twice, a Tavily news topic schema break, Google News API complaints, a stock-price lookup returning old links

As data: https://www.anchorterminal.com/api/v1/use-cases/search.recent-news.json, or the MCP tool get_use_case with id search.recent-news. Draft, read 2026-11-12 from public docs and first-hand accounts, not yet reviewed by a person.

---

Query this catalogue directly over MCP: https://www.anchorterminal.com/mcp (Streamable HTTP, no key). How: https://www.anchorterminal.com/agents/#mcp
