Skip to main content
Glama
Sign In
News Site Article Archive MCP Server

News Site Article Archive MCP Server

by mambalabsdev
OverviewSchemaSourceRelated ServersDiscussionsAdmin
JavaScript
Remote

News Site Article Archive MCP Server

npm version license

An MCP server for the Mamba Labs News Site Article Archive actor on Apify. Full article text and dates from named news sites across a past date range.

What's Inside

Related MCP server: article-scraper-mcp

What it does

Give news sites and a date range. The actor finds article addresses in the site sitemaps and the Internet Archive capture index, reads each page over plain HTTP, and returns one row per article with full text, published date, title, author, and section. A page gone from the site can be read from an Internet Archive copy. Paywalled and refused pages return a free row with no text.

All of the work runs on Apify. This package is a thin client that starts the actor run, waits for it, and hands back the dataset rows.

Quick start

You need Node.js 18 or newer and an Apify account with an API token.

Add this to your Claude Desktop config (claude_desktop_config.json), Cursor, or any MCP client:

{
  "mcpServers": {
    "mamba-news-site-article-archive": {
      "command": "npx",
      "args": [
        "-y",
        "@mambalabsdev/mcp-news-site-article-archive"
      ],
      "env": {
        "APIFY_TOKEN": "your-apify-token"
      }
    }
  }
}

Restart the client. The collect_news_site_articles tool appears.

Prerequisites

Example prompts

  • "Get every article this news site published in January 2016."

  • "Collect 10 years of reports from these three news sites."

  • "Pull the full text of this site's politics stories from last year."

Tool

collect_news_site_articles. One tool. Example arguments:

{
  "sites": [
    "techcrunch.com"
  ],
  "date_from": "2016-01-01",
  "date_to": "2016-01-31",
  "max_articles_per_site": 5
}

Inputs

Every buyer facing input of the actor. Omitted inputs take the actor's own defaults.

Input

Type

Required

Description

sites

array

no

News sites by domain, for example techcrunch.com, up to 20.

date_from

string

no

The earliest published date, as YYYY-MM-DD. Default: 30 days before Date to.

date_to

string

no

The latest published date, as YYYY-MM-DD. Default: today.

max_articles_per_site

number

no

Articles to return from each site, up to 5,000. Default 100.

url_contains

array

no

Optional. Keep only addresses with any of these texts, for example /politics/.

min_words

number

no

Pages with fewer words are not counted as articles and are free. Default 80.

use_archive_index

boolean

no

yes (default): when the sitemaps run out, also find addresses in the Internet Archive capture index, which reaches years that sitemaps no longer list. no: sitemaps only.

use_archive_copy

boolean

no

yes (default): a page the site answers with 404, 410, 500, 502, or 504 is read from the Internet Archive, at the archived row price. Refused and paywalled pages are never read from an archive. no: gone pages return a free row.

obey_robots

boolean

no

yes (default): pages robots.txt disallows are skipped and counted. no: read them anyway.

max_charge_usd

string

no

Optional. The run stops before a charge that would pass this amount. Summary rows are always returned.

Pricing

Pay per event on Apify, billed to your Apify account. Prices in USD.

Event

Free plan

Bronze

Silver

Gold

article-row, one article read from the publisher's page

$0.003

$0.00285

$0.0027

$0.00255

archived-article-row, one article read from an Internet Archive copy, only for a page gone from the site

$0.003

$0.00285

$0.0027

$0.00255

apify-actor-start, per run start

$0.021

$0.021

$0.021

$0.021

Refused, paywalled, gone, and out of range pages, summary rows, and refused input are free. Mamba Labs pays the platform costs of the run, Apify proxy traffic included. Mamba Labs pays the platform costs of the run, Apify proxy traffic included.

Full actor documentation

apify.com/mambalabs/news-site-article-archive

License

MIT. See LICENSE.

Available Tools

1 tool
collect_news_site_articlesCollect articles from news sites by date rangeA
Read-onlyIdempotent

Find the articles a named news site published in a past date range, from its sitemaps and the Internet Archive capture index, and return full text, published date, title, author, and section. Paywalled and refused pages return a free row with no text. No browser, so sites built by JavaScript return little. Pay per event on Apify: 0.003 USD per article.

NameRequiredDescriptionDefault
sitesNoNews sites by domain, for example techcrunch.com, up to 20.
date_toNoThe latest published date, as YYYY-MM-DD. Default: today.
date_fromNoThe earliest published date, as YYYY-MM-DD. Default: 30 days before Date to.
min_wordsNoPages with fewer words are not counted as articles and are free. Default 80.
obey_robotsNoyes (default): pages robots.txt disallows are skipped and counted. no: read them anyway.
url_containsNoOptional. Keep only addresses with any of these texts, for example /politics/.
max_charge_usdNoOptional. The run stops before a charge that would pass this amount. Summary rows are always returned.
use_archive_copyNoyes (default): a page the site answers with 404, 410, 500, 502, or 504 is read from the Internet Archive, at the archived row price. Refused and paywalled pages are never read from an archive. no: gone pages return a free row.
use_archive_indexNoyes (default): when the sitemaps run out, also find addresses in the Internet Archive capture index, which reaches years that sitemaps no longer list. no: sitemaps only.
max_articles_per_siteNoArticles to return from each site, up to 5,000. Default 100.

TDQS

A4.6/5.0
Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Beyond the readOnly/idempotent annotations, the description discloses the failure modes that matter: paywalled and refused pages yield a free row with no text, JS-built sites yield little, and pricing is 0.003 USD per article on Apify. It also explains the cost-control mechanism (summary rows always returned, runs stop before a charge). This is behavior an agent could not infer from annotations.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Four sentences, front-loaded with purpose, then limitations, then cost. No filler, no restatement of the title, and each sentence carries distinct information an agent needs.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a 10-parameter, no-required-parameter, no-output-schema tool, the description covers purpose, return fields, failure modes, and cost, which is most of what an agent needs. It could say more about default volume (max_articles_per_site default 100) and how the archive flags interact, but nothing essential is missing.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, so the baseline is 3, but the description adds economics and semantics the schema does not: the per-article price that makes max_charge_usd meaningful, and the fact that below-min_words pages are free. It does not explain the interaction between the two archive flags or the per-site cap, leaving some parameter interpretation to the schema.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description opens with a specific verb and resource ("Find the articles a named news site published in a past date range") and names both retrieval channels (sitemaps, Internet Archive capture index). It also enumerates the returned fields (full text, published date, title, author, section), so an agent knows exactly what it gets without opening the schema.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Usage context is clear: date-range article harvesting from a named domain, with explicit exclusions about where it fails ("No browser, so sites built by JavaScript return little"). There are no sibling tools to route against, and no explicit statement of when not to use it beyond the JS caveat, so it stops short of a 5.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

  1. 1 tool updatev1.0.0
    • First observedcollect_news_site_articles
Deploy Server
One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

TDQS

A4.4/5.0

Scored across 1 tool

Disambiguation5/5

There is only one tool, so there is no possibility of confusing it with another. Its purpose — retrieving archived articles from a named news site for a date range — is stated explicitly and cannot be mistaken for anything else.

Naming Consistency4/5

The single name collect_news_site_articles follows a clear verb_noun snake_case convention consistent with common MCP patterns. With only one tool there is no cross-tool pattern to evaluate, so it cannot demonstrate full consistency across a set.

Tool Count3/5

The narrow scope (one retrieval operation over sitemaps plus archive index) means a single tool is defensible, but a one-tool surface is thin and offers no way to discover supported sites or adjust output. It sits at the borderline of minimal but adequate.

Completeness3/5

The core retrieve-articles operation is covered end to end, including full text and metadata, but there is no companion tool for listing supported sites, paginating large date ranges, or handling failed/JavaScript-heavy sites. Agents hitting those limits have no fallback within the server.

Maintenance

ActivityMaintained
ResponsivenessNo recent community issues

Resources

Related MCP Connectors

Related MCP Servers

  • Newsnow MCP Server

    BACH-AI-Tools
    A
    license
    C
    quality
    D
    maintenance
    Enables access to news from multiple sources including Google News and DuckDuckGo, supporting queries by location, category, site, date range, and providing top news headlines with detailed information.
    7
    MIT
  • newspaper-mcp

    newspaper-mcpofficial

    AndyTheFactory
    A
    license
    Not graded
    quality
    C
    maintenance
    Enables LLM agents to discover news sources, list articles, parse and extract article text, search news, and perform NLP analysis such as keyword extraction and summarization.
    MIT

  • mining-news-mcp

    joker2still
    F
    license
    Not graded
    quality
    B
    maintenance
    Enables searching and fetching news articles with configurable date ranges, returning structured, sourced evidence for daily mineral-rights briefings.
    -