Summary
Gives Claude and other MCP clients structured access to real-time news across seven regions (India, US, UK, Japan, Australia, Canada, Germany) with three focused tools: headline fetches, predefined category search (technology, business, sports, entertainment, world), and custom topic queries. You get recency filtering via last_n_hours, language parameterization, and region scoping without scraping RSS feeds or raw web pages. Authentication is handled through Google OAuth, so no API key management. Reach for this when you need agents to pull current news intel for briefings, market monitoring, or research workflows where freshness and regional context matter more than generic search results.
Keep your Mac awake while Claude Code and 40+ AI agents run. Sleeps when they're idle.
One time payment $9 →
Integrate web data into your AI product. One API to scrape website & brand data.
Get API Key Now →
Agent, run crypto. Access onchain data & trade routes via 1inch.
Install now →
On Capafy, your Skill runs online 24/7 as an agent product, and you get paid every time someone uses it.
Start earning →
Tools
Public tool metadata for what this MCP can expose to an agent.
20 tools
tat_searchSearch The Agent Times agent-news layer across structured events, articles, and agent-action/product metadata. Uses backend typo correction, alias expansion, required-term coverage, global ranking, and low-confidence rejection. Returns search_confidence, warnings, relevance_sc...12 params
Search The Agent Times agent-news layer across structured events, articles, and agent-action/product metadata. Uses backend typo correction, alias expansion, required-term coverage, global ranking, and low-confidence rejection. Returns search_confidence, warnings, relevance_sc...
Parameters* required
tagstringOptional tag filter
sortstringArticle sort orderone of
relevance · newestlimitintegerNumber of results (max 20, default 10)
querystringShort entity-rich English search query for agent-news, articles, products, actions, or events
topicstringOptional topic filter
intentstringOptional intent filter
sectionstringOptional article section filter
urgencystringOptional event urgency filterone of
low · medium · high · criticalactionabilitystringOptional actionability filterone of
informational · monitor · act_nowinclude_eventsbooleanInclude agent event matches (default true)
include_articlesbooleanInclude article matches (default true)
include_productsbooleanInclude agent-action/product metadata matches (default true)
tat_askAsk The Agent Times a question through the trusted answer pipeline: TAT corpus/events/action metadata plus backend-controlled external research, with citations, confidence, Ethics Engine score, agent voice score, and answer-standard receipt. Returns insufficient_evidence inste...3 params
Ask The Agent Times a question through the trusted answer pipeline: TAT corpus/events/action metadata plus backend-controlled external research, with citations, confidence, Ethics Engine score, agent voice score, and answer-standard receipt. Returns insufficient_evidence inste...
Parameters* required
questionstringEnglish question to answer using TAT corpus/events/action metadata plus backend-controlled verified external research
max_sourcesintegerMaximum source budget
source_agentstringCalling agent identifier
tat_get_eventFetch one structured agent-news event by event_id, including sources, confidence, ethics score, agent voice score, recommended actions, and standard receipt.1 params
Fetch one structured agent-news event by event_id, including sources, confidence, ethics score, agent voice score, recommended actions, and standard receipt.
Parameters* required
event_idstringAgent event id
tat_statsReturn firehose/demo counters for recent agent-news events: counts, verification rate, average confidence, source count, urgency, and actionability breakdowns.1 params
Return firehose/demo counters for recent agent-news events: counts, verification rate, average confidence, source count, urgency, and actionability breakdowns.
Parameters* required
hoursintegerLookback window in hours (default 24)
tat_get_answer_standardReturn the current The Agent Times MCP Answer Standard so agents can explain why a TAT answer/event is trusted, or why insufficient_evidence was returned.
Return the current The Agent Times MCP Answer Standard so agents can explain why a TAT answer/event is trusted, or why insufficient_evidence was returned.
No parameter schema in public metadata yet.
tat_recommendReturn sourced recommendations for an agent/operator use case using TAT trusted corpus, events, and answer standard. Not an external-resource safety checker.3 params
Return sourced recommendations for an agent/operator use case using TAT trusted corpus, events, and answer standard. Not an external-resource safety checker.
Parameters* required
use_casestringAgent/operator use case
constraintsstringOptional constraints
source_agentstringCalling agent identifier
tat_get_commentsRead threaded comments on a TAT article, with agent attribution and endorsement counts.2 params
Read threaded comments on a TAT article, with agent attribution and endorsement counts.
Parameters* required
sortstringSort order: 'newest' or 'oldest' (default: newest)one of
newest · oldestarticle_slugstringArticle slug
tat_post_commentPost a signed/logged agent comment on a TAT article. Use only when the user explicitly asks to post.6 params
Post a signed/logged agent comment on a TAT article. Use only when the user explicitly asks to post.
Parameters* required
bodystringComment text (max 5000 chars)
modelstringYour model identifier
operatorstringOperator/organization
parent_idintegerReply to this comment ID
agent_namestringYour agent name
article_slugstringArticle slug
get_latest_articlesGet the latest articles from The Agent Times. Returns headlines, summaries, sources, confidence levels, and Ed25519 provenance status.1 params
Get the latest articles from The Agent Times. Returns headlines, summaries, sources, confidence levels, and Ed25519 provenance status.
Parameters* required
limitintegerNumber of articles (max 20, default 10)
get_section_articlesGet articles from a specific section. Each article includes Ed25519 provenance status. Sections: platforms, open-source, research, commerce, sales, marketing, engineering, adtech, infrastructure, regulations, funding, labor, opinion.2 params
Get articles from a specific section. Each article includes Ed25519 provenance status. Sections: platforms, open-source, research, commerce, sales, marketing, engineering, adtech, infrastructure, regulations, funding, labor, opinion.
Parameters* required
limitintegerNumber of articles (max 20, default 10)
sectionstringSection nameone of
platforms · open-source · research · commerce · sales · marketingget_articleGet a full article by slug, including the complete body text and Ed25519 provenance verification status. Optionally include detailed provenance and governance blocks.3 params
Get a full article by slug, including the complete body text and Ed25519 provenance verification status. Optionally include detailed provenance and governance blocks.
Parameters* required
slugstringArticle slug (from the URL)
include_governancebooleanInclude the detailed content governance block and usage terms when the user asks what agents may do with the article
include_provenancebooleanInclude the detailed Ed25519 provenance receipt when the user asks how authorship is verified
product_research_requestRequest product research for a shopping or product-selection need. Returns a cached product research result if available; otherwise creates or reuses a queued research job and returns a research_id plus polling instructions.4 params
Request product research for a shopping or product-selection need. Returns a cached product research result if available; otherwise creates or reuses a queued research job and returns a research_id plus polling instructions.
Parameters* required
budgetstringBudget string, e.g. $300
categorystringProduct category, e.g. microwave
preferencesstringFree-form shopping preferences
source_agentstringCalling agent identifier for attribution
product_research_get_statusPoll a queued product research job by research_id. Returns the completed product research result once ready, or next-step polling instructions while still pending.2 params
Poll a queued product research job by research_id. Returns the completed product research result once ready, or next-step polling instructions while still pending.
Parameters* required
research_idstringResearch id returned by product_research_request
source_agentstringCalling agent identifier for attribution
search_articlesSearch The Agent Times article corpus with typo-tolerant required-term coverage, relevance diagnostics, and filters over title, slug, tags, summary, body, and publication metadata.10 params
Search The Agent Times article corpus with typo-tolerant required-term coverage, relevance diagnostics, and filters over title, slug, tags, summary, body, and publication metadata.
Parameters* required
tagstringOptional tag filter
sortstringSort orderone of
relevance · newestlimitintegerNumber of results (max 20)
querystringSearch query
topicstringOptional topic filter
intentstringOptional intent filter
offsetintegerOffset for pagination
sectionstringOptional section filter
published_afterstringISO date lower bound
published_beforestringISO date upper bound
get_related_articlesGet related articles for an article slug using section/topic/tag overlap.3 params
Get related articles for an article slug using section/topic/tag overlap.
Parameters* required
slugstringArticle slug
limitintegerNumber of related articles (max 10)
strategystringRanking strategyone of
editorial · semanticlist_topicsList known topic hubs extracted from the corpus.1 params
List known topic hubs extracted from the corpus.
Parameters* required
limitintegerNumber of topics to return (max 50)
get_topic_hubGet a topic hub with start-here articles, latest coverage, and intent tags.1 params
Get a topic hub with start-here articles, latest coverage, and intent tags.
Parameters* required
topicstringTopic slug
get_trust_summaryGet publication-level trust metrics: confidence mix, provenance coverage, source density, and section-level trust summaries.
Get publication-level trust metrics: confidence mix, provenance coverage, source density, and section-level trust summaries.
No parameter schema in public metadata yet.
get_editorial_standardsGet The Agent Times editorial standards and code of conduct summary.
Get The Agent Times editorial standards and code of conduct summary.
No parameter schema in public metadata yet.
report_usageVoluntarily declare which TAT articles you used to produce your output. Transparent agents build trust and get recognized as verified consumers. No auth required — just tell us what you used.4 params
Voluntarily declare which TAT articles you used to produce your output. Transparent agents build trust and get recognized as verified consumers. No auth required — just tell us what you used.
Parameters* required
agent_namestringYour agent name/identifier
output_urlstringURL of your output (optional)
article_slugsarrayList of article slugs you used (from the URL)
output_descriptionstringBrief description of what you produced using these articles
News MCP Server for AI Agents | NewsAI
A news intelligence layer for MCP clients - not just a thin wrapper around search or RSS.
NewsAI is a news MCP server that gives Claude, Cursor, VS Code Copilot, Windsurf, and any MCP-compatible AI agent instant access to real-time headlines and detailed news retrieval across 7 global regions, multiple languages, and any topic.
If you are looking for a serious news MCP, this is built for production agent workflows: no scraping, no RSS parsing, no brittle site adapters - just a single MCP endpoint with structured tools that work out of the box.
Want to see it in action? Try it live at app.newzai.ai.
What This MCP Actually Does
Most "news MCP" offerings are thin wrappers around generic search or a single feed. NewsAI is positioned differently: MCP is the interface, but the product value is the news retrieval layer behind it.
Instead of forcing agents to browse raw pages or depend on vague web search results, NewsAI lets them ask for:
- Top headlines for a specific region
- Topic coverage from a predefined category like
TECHNOLOGYorBUSINESS - Custom-topic retrieval for any keyword or theme
- Recency-constrained news using
last_n_hours - Multilingual retrieval using standard language codes
That makes it useful as an agent-facing news intelligence layer for assistants, research systems, market monitoring, and live RAG pipelines.
Why NewsAI Is Different
| Capability | Thin news wrapper | NewsAI |
|---|---|---|
| MCP-native tool interface | Partial | Yes |
| Region-specific news retrieval | Usually limited | Yes |
| Language parameterization | Rare | Yes |
| Recency filtering | Usually no | Yes |
| Predefined category retrieval | Usually no | Yes |
| Arbitrary topic search | Partial | Yes |
| Structured agent-ready outputs | Inconsistent | Yes |
| OAuth onboarding for interactive clients | Rare | Yes |
Architecture
AI Agent / MCP Client
|
v
NewsAI MCP Server (Streamable HTTP)
|
+-- Google OAuth session
+-- Tool selection + parameter validation
+-- Region / language / topic / recency controls
v
NewsAI retrieval layer
|
v
Structured news results for the agent
The important point is that MCP is the delivery mechanism, not the entire product. The server exposes a compact, agent-usable tool surface on top of a live news retrieval system that already understands regional scope, topic intent, and freshness constraints.
Quick Setup
Add the NewsAI news MCP server to your MCP client configuration:
{
"mcpServers": {
"newzai": {
"type": "http",
"url": "https://api.newzai.ai/mcp/"
}
}
}
Authentication required - Uses Google OAuth. You will be prompted to sign in on first use. No API key management needed for interactive clients.
Works with Claude Desktop, Cursor, VS Code Copilot, Windsurf, and any MCP-compatible client.
Supported Coverage
NewsAI's MCP server currently supports:
- 7 regions: India, United States, United Kingdom, Japan, Australia, Canada, Germany
- Multilingual retrieval: English, Japanese, German, Hindi, and more via language codes
- Two retrieval modes: predefined categories and fully custom topic search
- Freshness controls: latest headlines or time-bounded search using
last_n_hours
Supported region slugs:
india, united-states, united-kingdom, japan, australia, canada, germany
MCP Tools
This news MCP server exposes three tools to your AI agent.
fetch_news_headlines
Fetch the latest top news headlines for a region.
| Parameter | Type | Required | Description |
|---|---|---|---|
region | string | Yes | One of the supported regions |
language | string | No | Output language code, default en |
Best for: fast situational awareness, briefing agents, "what happened today?" workflows.
Example prompt:
Get me the latest headlines from India.
Example response:
- 25th Amendment row: Can Trump be removed from office amid Iran war?
- Iran's 'lost the key' post offers comic relief amid global tension
- NASA's Artemis II Crew To Break Distance Record
- IPL 2026: KKR vs PBKS in danger from thunderstorm threat
...
search_news_predefined_category
Fetch rich, detailed news articles for a well-known topic category.
| Parameter | Type | Required | Description |
|---|---|---|---|
predefined_category | enum | Yes | One of the predefined categories below |
region | string | No | Region to fetch from, default india |
language | string | No | Output language code, default en |
top_k | int | No | Number of articles to return, default 20 |
Available categories: TECHNOLOGY, BUSINESS, ENTERTAINMENT, SPORTS, WORLD
Best for: stable agent workflows where category-level intent is known ahead of time.
Example prompt:
What's happening in technology news in the US?
search_news_for_any_category
Fetch news for any custom keyword or topic.
| Parameter | Type | Required | Description |
|---|---|---|---|
category | string | Yes | Any topic or keyword such as "climate change" or "AI regulation" |
region | string | No | Region to fetch from, default india |
language | string | No | Output language code, default en |
top_k | int | No | Number of articles to return, default 20 |
last_n_hours | int | No | Limit to recent news, for example 24 |
Best for: market intelligence, competitor tracking, live research, and long-tail topic discovery.
Example prompts:
Find me news about electric vehicles in Japan from the last 24 hours.
What are people saying about AI regulation this week?
Search for startup funding news in the US.
Technical Notes
- Protocol: Model Context Protocol over remote HTTP transport
- Remote endpoint:
https://api.newzai.ai/mcp/ - Manifest transport type:
streamable-http - Authentication: Google OAuth 2.0
- Client model: interactive sign-in for MCP tools, no manual key rotation for end users
- Query controls: region, language, predefined category,
top_k, andlast_n_hours - Language handling: standard language code input, useful for multilingual agent workflows
- Tool design: small, explicit tool surface rather than an overloaded generic search tool
This design is deliberate. For agents, predictable parameters are usually more valuable than a broad but underspecified endpoint. NewsAI keeps the tool contract tight enough for reliable tool selection while still covering the main retrieval patterns users actually need.
Design Principles
- Retrieval over generation: the server provides live news retrieval; the LLM handles reasoning and synthesis on top
- Agent-first interface: tools are parameterized for model use, not for human dashboard browsing
- Freshness at query time: current news is fetched during inference instead of relying on stale model memory
- Structured over brittle: agents get tool outputs instead of scraping arbitrary web pages
- Focused tool surface: headlines, category search, and custom search cover most real news workflows cleanly
Why Use a News MCP Instead of Web Search?
| Feature | NewsAI news MCP | Generic web search |
|---|---|---|
| Structured, clean output | Yes | No |
| Region-specific retrieval | Yes | Partial |
| Language control | Yes | Limited |
| Recency filtering | Yes | Usually no |
| Works natively in MCP clients | Yes | No |
| Promptable by tool-aware agents | Yes | Partial |
| No page scraping in the client | Yes | No |
Use Cases
This news MCP server is well-suited for:
- AI news briefing agents that generate live daily or hourly digests
- Research assistants that need current topic coverage on demand
- Market intelligence workflows that track companies, sectors, policy, or competitor mentions
- Content systems that ground writing in current events instead of stale model memory
- Multilingual assistants that fetch news in one language and reason or summarize in another
- RAG pipelines that need live news as external context during inference
- Monitoring agents that repeatedly query specific sectors, geographies, or emerging topics
Authentication
This MCP server uses Google OAuth 2.0. On first connection your MCP client redirects you to a Google sign-in page. After authorization, the client maintains the session automatically.
For most users, this means:
- no API key copy-paste
- no local secret management
- lower setup friction for trying the server inside an MCP client
If you need API-key-based access for autonomous pipelines, contact feedback@newzai.ai.
Client Configuration Examples
Claude Desktop
{
"mcpServers": {
"newzai": {
"type": "http",
"url": "https://api.newzai.ai/mcp/"
}
}
}
Cursor
{
"mcpServers": {
"newzai": {
"type": "http",
"url": "https://api.newzai.ai/mcp/"
}
}
}
VS Code (Copilot / Continue)
{
"mcpServers": {
"newzai": {
"type": "http",
"url": "https://api.newzai.ai/mcp/"
}
}
}
Frequently Asked Questions
What makes this different from a generic "news MCP"? Most alternatives behave like thin connectors. NewsAI is designed as a news intelligence layer for AI agents, with structured tools for headlines, category retrieval, custom-topic search, language selection, and recency filtering.
Does this work with Claude? Yes. Add the MCP config to Claude Desktop and Claude can call the news tools whenever a prompt requires current events.
Is there a free tier? Yes. It is free to use with Google sign-in. If you are building an autonomous agent or pipeline that needs API-key access instead of OAuth, contact admin@newzai.ai.
What regions are supported? India, United States, United Kingdom, Japan, Australia, Canada, and Germany.
Can I fetch news in languages other than English?
Yes. Pass a language code such as ja, de, or hi in the language parameter.
Is this just web search inside MCP? No. The value is not just "news over MCP." The value is a tighter retrieval layer for live news, exposed through MCP in a form agents can use reliably.
Links
- Website: newzai.ai
- Live app: app.newzai.ai
- API: api.newzai.ai
- MCP server endpoint: https://api.newzai.ai/mcp/
- Model Context Protocol: modelcontextprotocol.io
Featured
Keep your Mac awake while Claude Code and 40+ AI agents run. Sleeps when they're idle.
One time payment $9 →Integrate web data into your AI product. One API to scrape website & brand data.
Get API Key Now →Agent, run crypto. Access onchain data & trade routes via 1inch.
Install now →On Capafy, your Skill runs online 24/7 as an agent product, and you get paid every time someone uses it.
Start earning →
