# Arquivo Pt [Health: Active]

**Category:** 🔎 Search & Data Extraction  
**Repository:** https://github.com/pipeworx-io/mcp-arquivo-pt  
**GitHub Stars:** 0  
**npm Downloads (last month):** 1203  
**Views:** 5  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/arquivo-pt

## Description
Arquivo.pt MCP — full-text search over the Portuguese web archive.

## Tools
Capabilities this server exposes over MCP:

- **arquivo_search_pages** — Full-text search of archived web pages in Arquivo.pt, the Portuguese web archive (captures since 1996), when the URL is unknown. Finds pages whose TEXT contains the given terms or a quoted phrase and returns, per hit, the page title, original URL, capture date, a text snippet with the matching terms, the archived (replay) URL, and links to the extracted text and a screenshot. Coverage: centred on the Portuguese web (.pt sites and Portuguese-language pages) but international English-language sites that Portuguese pages link to are captured too — probes for "climate change", "quantum computing" and "Federal Reserve interest rates" return roughly 50M, 1M and 2M estimated hits, with ipcc.ch, microsoft.com and federalreserve.gov among the top results. The full-text index lags the crawl by years (probed 2026-10: most hits end in 2020, a few reach 2023) while arquivo_url_history carries current captures, so use that for anything recent at a known URL. Supports a date window (YYYY or YYYYMMDD), restricting to one site, restricting to a file type (pdf, html, doc), exclusion with a leading minus (Albert -Einstein), and paging by offset. Do not pass a URL as the query; use arquivo_url_history for that.
- **arquivo_url_history** — List every preserved capture of a specific URL in Arquivo.pt, the Portuguese web archive, newest first — the version history of a page or domain from 1996 to the current crawl. Each capture carries the capture timestamp (UTC), the archived replay URL, the crawl HTTP status, MIME type, content length and content digest, plus links to the extracted text, the raw original file and a screenshot. Accepts a bare domain (publico.pt), a host (www.publico.pt) or a full URL with path and query. Use this to see how a Portuguese page changed over time or to find the capture timestamp to pass to arquivo_page_text; use arquivo_search_pages when you only know words from the page, not its address.
- **arquivo_page_text** — Read the extracted plain text of one archived page capture held by Arquivo.pt, the Portuguese web archive — the page content with HTML stripped, as the archive indexed it. Takes the original URL and the capture timestamp (YYYYMMDDhhmmss) that arquivo_search_pages or arquivo_url_history returned, and returns the text (optionally truncated to max_chars) with the archived replay URL. Use it to quote or summarise what a Portuguese page said on a given date without fetching the archived HTML.

## Claude Desktop Quick Installation
Install path detected from listing signals. Uses `npx` (confidence: high):

```json
"mcpServers": {
  "arquivo-pt": {
    "command": "npx",
    "args": ["-y","@pipeworx/mcp-arquivo-pt"]
  }
}
```

## Documentation

## What Arquivo Pt does

Arquivo Pt connects MCP clients to Arquivo.pt, the Portuguese web archive operated by FCCN. Its three tools support discovery and inspection of archived material:

- `arquivo_search_pages` finds archived pages by words or quoted phrases when the address is unknown.
- `arquivo_url_history` lists preserved captures for a domain, host, or complete URL.
- `arquivo_page_text` returns the plain text extracted from one selected capture.

Search results include titles, original URLs, capture dates, matching snippets, replay links, extracted-text links, and screenshot links. URL history adds HTTP status, MIME type, content length, and a content digest for each capture.

## How it works

Use full-text search to locate relevant pages, then pass a result's original URL and capture timestamp to `arquivo_page_text`. When the address is already known, start with `arquivo_url_history` instead; it returns captures newest first and can identify the timestamp needed for text retrieval.

Search accepts date boundaries in `YYYY` or `YYYYMMDD` form, site and file-type filters, pagination offsets, and a per-site result limit. A leading minus excludes a term, and quoted text searches for a phrase. URL history accepts a bare domain, host, or full URL with a path and query.

The hosted MCP endpoint is `https://gateway.pipeworx.io/arquivo-pt/mcp`. The full Pipeworx gateway is also available at `https://gateway.pipeworx.io/mcp`. Both expose the Arquivo.pt tools through the same gateway; the scoped endpoint also includes shared Pipeworx meta-tools at the protocol level.

## Setup and configuration

Add the hosted endpoint to an MCP client with a server URL pointing to the Arquivo Pt gateway. Claude Desktop, Cursor, and Windsurf are shown as supported client options.

For a local stdio connection, run `npx -y @pipeworx/mcp-arquivo-pt`. The package forwards requests to the hosted gateway and requires network access; it does not provide offline operation. Anonymous calls are available with a daily limit. A free Pipeworx account and `PIPEWORX_API_KEY` raise the daily allowance. Set `PIPEWORX_META_TOOLS=1` if the local bridge should list the gateway-wide meta-tools as well as the Arquivo.pt tools.

## Tools and capabilities

- Search indexed page text with terms or exact phrases.
- Filter searches by date, site, file type, and result position.
- List historical captures for domains, hosts, and URLs.
- Read archive-extracted plain text with an optional character limit.
- Retrieve replay, text, raw-file, and screenshot links where returned.

## Limitations and notes

Coverage focuses on `.pt` sites and Portuguese-language pages, while some international pages linked from Portuguese sites are also captured. The full-text index may trail the crawl by several years, so recent research for a known address should use `arquivo_url_history` rather than text search.

Arquivo.pt distributes requests across index nodes that can disagree or return malformed or delayed results. The integration retries once, removes rows lacking a usable URL or timestamp, and reports upstream faults rather than treating them as empty search results. Search results are deduplicated to two captures per site by default; increase `per_site` to inspect more captures. Do not submit a URL as a search query; use URL history for address-based lookups.

_Full upstream README: https://allmcps.com/mcp/arquivo-pt/readme_