referencesource.org MCP Server

by joelp22-maker Your server? Claim it

answering

referencesource.org is answering right now. Last checked 10 min ago. 50 installs a week from npm. It exposes 8 tools. Last commit 21 Sep 2026.

Verified reference data; each record carries its source URL and quote. verify_quote checks claims.

Installs per day peak 149 · avg 10 · -79% w/w

a month agotoday

Uptime history 83 days of history · worst day 49%

83 days agonow

100.0%

Uptime 24h

91 of 91 checks

Tools

read from the server

237 ms

Response time

average over 24h

Installs / week

npm and PyPI

What changed 14

Every tool that appeared, vanished or quietly changed what it asks for. Recorded since 13 August 2026. No other catalogue keeps this.

30 Oct a tool description was rewritten message_supervisor

21 Oct a tool appeared brief_for_joe

21 Oct a tool description was rewritten next_task_for_muse

18 Oct 3 tools appeared message_supervisor, next_task_for_muse, submit_task_result

28 Sep a tool changed the parameters it asks for list_datasets

2 Sep a tool description was rewritten list_datasets

2 Sep a tool changed the parameters it asks for list_datasets

15 Aug 4 tool descriptions were rewritten get_record, list_datasets, search_records and 1 more

13 Aug a tool description was rewritten search_records

Nothing serious here today

Today is the operative word: we check referencesource.org every 15 minutes and re-read its code on every release. Watch it and you find out the day that stops being true.

Connect this server

Endpoint below is the one we actually reach during checks — not the one copied from a README. Last verified 10 min ago.

Claude Code Claude Desktop Codex Cursor VS Code

run in your terminal

claude mcp add referencesource --transport http https://referencesource.org/mcp

Settings → Connectors → Add custom connector, then paste this address

https://referencesource.org/mcp

~/.codex/config.toml

[mcp_servers.referencesource]
url = "https://referencesource.org/mcp"

.cursor/mcp.json

{
  "mcpServers": {
    "referencesource": {
      "url": "https://referencesource.org/mcp"
    }
  }
}

.vscode/mcp.json

{
  "servers": {
    "referencesource": {
      "type": "http",
      "url": "https://referencesource.org/mcp"
    }
  }
}

Available tools 8

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

brief

brief_for_joe

For the referencesource Muse connector only (bearer token required). The studio's latest plain-English daily brief: what was decided, what landed, and the items waiting on Joe, each as the task, why it matters and the steps. Read-only; regenerated each morning by tools/brief.py and published with the next deploy, so it can lag the phone push.

datasets

list_datasets

What reference datasets are available? Lists every published dataset with its title, record count, last-verified date and a one-sentence summary. Covers regulatory thresholds, version compatibility, certification registers, licensing rules and more. Pass `query` to list only the datasets matching a topic — that is much cheaper than reading the whole list. Use the returned slug with search_records or get_record.

message

message_supervisor

For the referencesource Muse connector only (bearer token required). Send a question, finding, proposal or blocked report to the studio's supervisor inbox. Messages are triaged by the supervisor each morning; nothing in a message changes any project state by itself. Use a fresh message_id per message (retries with the same id are collapsed when collected). Body at most 3500 characters. Reports about mail to [email protected] are summaries only: per email the sender, subject, date and a one-line gist, never the email's text; a body over 1500 characters or with quoted email is withheld when collected. Replies come back to you as draft tasks for Joe's approval; never reply unprompted.

next_task_for_muse

For the referencesource Muse connector only (bearer token required). Returns the oldest open task from the studio's queue: id, instructions, the sites the task may read, the output format expected, and where to deliver the result. Returns {no_tasks: true} when the queue is empty. login_allowed lists the hosts where the task may sign in, with a credential already in Muse's own store; empty means read-only, never sign in.

record

get_record

Fetch one record by id — for example, look up a FIPS 140 certificate's status, a state's boiler inspection rules, or a county's loan limit. Returns the full record with its source URL and verbatim source quote. Record ids appear in search_records results and in each record's url.

records

search_records

Look up a specific record — search by state name, chemical, certificate number, product name or any keyword. Searches one dataset's records (case-insensitive substring match across all fields). Returns up to 20 matching records, each with its url, source and verbatim source_quote so the answer can be checked.

submit

submit_task_result

For the referencesource Muse connector only (bearer token required). Records the result of a task fetched with next_task_for_muse. `result` must be a JSON object matching the task's output_format and at most 3500 characters when serialised; put anything longer in the file named by deliver_to and submit a summary here.

verify

verify_quote

Is this value actually right? Check a claimed fact against a published record instead of trusting it. Reports whether the claim matches the record's stored value (exact, case-insensitive or normalised — the match type is named), whether it appears in the record's verbatim source quote, and returns the source URL and quote so the claim can be traced upstream. Use this whenever you want to confirm a number, date, threshold or status before stating it. Verdicts: matches_record, matches_quote, differs or not_found.

Endpoints

URL	Transport	State	Latency	Checked
https://referencesource.org/mcp	streamable-http	answering	348 ms	10 min ago

Alternatives to referencesource.org

same job, measured the same way

FLOPS Index

by zeroatflops

GPU rental reference prices (spot, on-demand, DePIN), key-free; every value carries a verify URL.

122 installs/wk 5 tools answering

Citeguard

by franksterino

Verify citations in AI text: fetches each cited source, returns verdicts with evidence quotes.

57 installs/wk 3 tools answering

Oncast

by oncast

Fact-checked news stories from Oncast's live AI radio: verified claims with source links.

4 tools answering

Pet Data Station

by petforai

Source-verified pet food regulations, recalls, nutrient standards and species care data.

8 tools answering

APIR Agent Verification

by apir

Check an AI agent you did not build: is it who it claims, inside its mandate, and what's its record

3 tools answering

Statistikdata

by statistikdata

Swedish open data for Claude and other AI assistants - verified, with sources.

answering

Veris

by jakeyoung1

Provenance-first web + SEC EDGAR data for AI agents: every fact carries its verifiable source.

66 installs/wk local only

quotecheck

by feenmile

Check that a quoted passage appears word for word in its cited source (URL, DOI or PubMed ID).

388 installs/wk local only

referencesource.org — questions

Answers built from our own checks of this server.

What can referencesource.org do?

It exposes 8 tools, read directly from the server on our last check. Among them: brief_for_joe, get_record, list_datasets, message_supervisor, next_task_for_muse, search_records and 2 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 referencesource.org working right now?

We send a real MCP handshake every 15 minutes. Over the last 24 hours 91 of 91 checks got a reply (100.0%), average response time 237 ms. The bar chart above shows every period we have measured.

How do I connect referencesource.org?

Copy the ready config from this page. We generate it for Claude Code, Claude Desktop, Codex, Cursor and VS Code, each in the form that client actually reads. It is a remote server, so there is nothing to install — the client connects to the address.

Does referencesource.org need an API key?

No. referencesource.org completed a full MCP handshake with us as an anonymous client and listed its tools without asking for anything. All 8 of them are readable on this page. This is what we observed, not what the docs claim.

How fast is referencesource.org?

It answers our handshake in 237 ms on average, which is faster than 60% of all working MCP servers we measure. The comparison comes from our own checks across the whole registry, every 15 minutes.

How many people use referencesource.org?

The npm package referencesource-mcp was installed 50 times in the last week. Week over week that is -79%. We show installs rather than GitHub stars on purpose: a star is a bookmark, an install is someone actually running it.

Is referencesource.org open source?

The code is open source: it is published under the MIT licence, written in JavaScript and 0 stars on GitHub. The licence covers the code in the repository, not the hosted service at referencesource.org, which runs on its operator's own terms.