MCP NewsAPI.ai
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| NEWSAPI_AI_API_KEY | Yes | Your NewsAPI.ai API key (Dashboard). Secret — never commit it. | |
| NEWSAPI_AI_MAX_TOKENS_PER_SESSION | No | Max NewsAPI.ai tokens this server may spend per session (until the server process restarts). Default 200. | 200 |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| search_articlesA | Search news articles (NewsAPI.ai / Event Registry). Cost: 1 token per call (measured live) when limited to the last 31 days (the default); querying the historical archive (allow_archive=true) costs extra tokens per year touched. At least one of keyword / concept_uris / category_uris / source_uris / location_uris is required. Prefer concept_uris (from suggest_concepts, which is free) over free-text keyword for any company, person, place or organisation: keyword is for free terms/topics and is noisy with common words (e.g. "Telefónica" also matches the Spanish adjective "telefónica"), so resolve entities with suggest_concepts first and pass concept_uris. Results are compact: title, truncated body (body_len, default 300 chars), url, source, date, sentiment. Defaults to 10 results, newest first, duplicates skipped. Paginate with |
| get_articleA | Get one article by its URI (the |
| search_eventsA | Search news EVENTS: clusters of articles about the same real-world happening, with a title, summary, date, location, key concepts and article count. Best for "what is happening with X" — one event replaces dozens of near-duplicate articles. Cost: 5 tokens (measured live) per call within the last 31 days (default); allow_archive=true costs extra per year touched. At least one of keyword / concept_uris / category_uris / source_uris / location_uris is required. Defaults to 10 events, newest first; use sort_by="size" for the biggest stories. Use get_event for details on one. For companies, people, places or organisations, resolve them with suggest_concepts first and pass concept_uris; keyword is for free terms/topics and is noisy with common words (e.g. "Telefónica" also matches the Spanish adjective "telefónica"). |
| get_eventA | Get one event by URI (the |
| suggest_conceptsA | Resolve a name to concept URIs (people, organisations, places, Wikipedia topics). Pass the returned |
| suggest_categoriesA | Resolve a topic name to category URIs (e.g. "news/Business", "dmoz/Sports/Soccer"). Pass the |
| suggest_sourcesA | Resolve a publication name or domain to news source URIs (e.g. "bbc.co.uk"). Pass the |
| suggest_locationsA | Resolve a city or country name to a location URI. Pass the |
| get_trending_conceptsA | Get the concepts (people, organisations, places) that are trending right now in the news — computed by comparing mentions in the last 2 days against the last 2 weeks. Use |
| get_usageA | Get the token budget of this NewsAPI.ai account: tokens available, used and remaining for the current period. Call it before a batch of searches (measured costs: search_articles and get_article 1 token, search_events and get_event info 5, get_trending_concepts 10). Free: does not consume tokens itself. Suggest tools are free too. Also reports this server’s session budget (NEWSAPI_AI_MAX_TOKENS_PER_SESSION, default 200): calls that would exceed it fail locally without reaching the API. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 10 tools
Each tool has a distinct role: search vs. get for articles/events, resolver tools for different filter types, usage, and trending. search_articles and search_events overlap conceptually but are clearly differentiated by article-level vs. event-cluster results. The descriptions make selection boundaries explicit.
All names use snake_case and predictable verb_noun phrasing: get_*, search_*, suggest_*, and get_usage/get_trending_concepts. The resolver family is especially consistent. No mixed conventions or vague names.
Ten tools is well-scoped for a news search API. The set covers article/event retrieval, entity/filter resolution, trending concepts, and account usage without obvious redundancy. Each tool earns its place.
The surface covers core news workflows: searching articles/events, retrieving full details, resolving all major query filters, checking trending concepts, and monitoring token usage. Minor gaps exist, such as no direct historical trend/timeline tool or source/category detail lookup beyond suggestions, but agents can work around these.