mcpbeat

Reading Room MCP Server

by randpaulcovid
answering

Reading Room is answering right now. Last checked 13 min ago. It exposes 7 tools.

Search the U.S. Senate's subpoenaed COVID-19 records: 18,094 communications, each page-cited.

Uptime history 10 hours of history
10 hours agonow
100.0%
Uptime 24h
34 of 34 checks
7
Tools
read from the server
460 ms
Response time
average over 24h
open, no key
Access
streamable-http

Connect this server

Endpoint below is the one we actually reach during checks — not the one copied from a README. Last verified 13 min ago.

run in your terminal
claude mcp add reading-room --transport http https://randpaulcovid.org/mcp
~/Library/Application Support/Claude/claude_desktop_config.json
{
  "mcpServers": {
    "reading-room": {
      "url": "https://randpaulcovid.org/mcp"
    }
  }
}
~/.codex/config.toml
[mcp_servers.reading-room]
url = "https://randpaulcovid.org/mcp"
.cursor/mcp.json
{
  "mcpServers": {
    "reading-room": {
      "url": "https://randpaulcovid.org/mcp"
    }
  }
}
.vscode/mcp.json
{
  "mcpServers": {
    "reading-room": {
      "url": "https://randpaulcovid.org/mcp"
    }
  }
}

Available tools 7

Read directly from the server with tools/list, grouped by what they act on. If a tool disappears, we record the date.

concept
get_concept
Fetch a concept document from the OKF bundle — frontmatter and body — with every cross-link rewritten to an absolute URL, so the bundle can be traversed from what this returns. A person carries activity, correspondents, organization and a chronological statement timeline; a topic carries its principal voices; a source carries provenance and SHA-256. Accepts a slug (anthony-s-fauci), a bundle path (people/anthony-s-fauci.md), a person_id from the site or JSON API (anthony_s_fauci), or a display name (Anthony S. Fauci).
concepts
search_concepts
Search the analysis layer: one cross-linked document per person, organization, topic, released package, media outlet and substantial conversation, each carrying provenance back to the pages it derives from. Use this when the question is who someone is, what a topic covers, or what a package contains — rather than which records match a string, which is search_records. Matches names, aliases and descriptions; get_concept then returns one in full.
fetch
fetch
Retrieve the full text of one item by id. Accepts a doc_id from search (fauci_diary:entry:00026) and also an OKF concept id (anthony-s-fauci), so a person, organization or topic document can be read the same way. Every response carries the citation to the page of the Committee's PDF it came from.
pages
search_pages
Literal search over the OCR text of all 5,247 pages, before segmentation into records. This is the completeness layer: it reaches duplicate copies and pages no record covers, and matches across a whole page rather than within a chunk. Use it to settle whether a phrase appears in the corpus at all — a miss here is much stronger evidence of absence than a miss in search_records, though OCR error and page-spanning lines still make it imperfect.
record
get_record
Retrieve a single communication in full by doc_id, with its citation, participants and neighbouring messages where it sits in a conversation.
records
search_records
Find communications matching a question or phrase. Returns records with speaker, date, grade and a citation to the page of the Committee's PDF. Retrieval only: nothing is summarised, and a record's wording is the record's, not this server's.
search
search
Search 18,094 communications from the U.S. Senate HSGAC COVID-19 releases. Returns id, title, excerpt and URL for each match; pass an id to fetch for the full text. This is the standard search contract — search_records offers keyword mode for exact strings and control over quoted material, and search_pages searches the raw OCR text, which is what settles whether something is absent from the corpus.

Endpoints

URLTransportStateLatencyChecked
https://randpaulcovid.org/mcp streamable-http answering 409 ms 13 min ago

Alternatives to Reading Room

same job, measured the same way
Case Doha Record
by klmlfl

Search 30,000+ decided U.S. DOHA security-clearance decisions, cited to the public record.

7 tools answering
Japan Parliament Search
by easakura

Search Japan's National Diet records (1947-present): statements, Q&A, and debates with citations.

2 tools answering
Nexus MCP
by thenexusintel

U.S. gov records, fact-checks with citations, news search, and fraud/corruption/PRC trackers.

43 installs/wk local only
Copus
by copus-io

Search human-curated content recommendations from Copus - The Internet Treasure Map

29 installs/wk local only
Korean Rnd Regs MCP
by smilemin07

Search South Korea's national R&D regulations in natural language with cited provisions.

2 880 installs/wk local only
Semantic Scholar MCP Server
by smaniches

MCP server for the Semantic Scholar API: search 200M+ papers, citations, authors, recommendations.

606 installs/wk local only
Gdpr Decisions
by mister-franklin

EU GDPR decisions by thedpo.eu — search all EU DPA decisions and recommendations

73 installs/wk local only
LinkList
by linklist

Search source collections your team vetted. Every passage cites its URL and fetch date.

answering

Reading Room — questions

Answers built from our own checks of this server.

What can Reading Room do?
It exposes 7 tools, read directly from the server on our last check. Among them: fetch, get_concept, get_record, search, search_concepts, search_pages and 1 more. The full list with descriptions is on this page — we take it from the server itself via tools/list, not from a README. How MCP servers expose tools in the first place →
Is Reading Room working right now?
We send a real MCP handshake every 15 minutes. Over the last 24 hours 34 of 34 checks got a reply (100.0%), average response time 460 ms. The bar chart above shows every period we have measured.
How do I connect Reading Room?
Copy the ready config from this page — we generate it for Claude Code, Claude Desktop, Codex, Cursor and VS Code, each with the file path that client actually reads. It is a remote server, so there is nothing to install — the client connects to the address.
Does Reading Room need an API key?
No. Reading Room completed a full MCP handshake with us as an anonymous client and listed its tools without asking for anything. All 7 of them are readable on this page. This is what we observed, not what the docs claim.
How fast is Reading Room?
It answers our handshake in 460 ms on average, which is faster than 28% of all working MCP servers we measure. The comparison comes from our own checks across the whole registry, every 15 minutes.