Read this document with your agent

Paste this into Claude, ChatGPT or Cursor once the RegAlert connector is added (steps below). The read is free with your account.

Using the RegAlert connector, read the document with uid be87f829-d1aa-4c56-a575-917c05bc4732 (get_document) and give me: what it requires, who it applies to, key dates, and anything unusual.

Or fetch it directly: GET https://mcp.regalert.today/v1/documents/be87f829-d1aa-4c56-a575-917c05bc4732 with your OAuth token. Back to the document

RegAlert inside your agent

Search 68,877 regulatory documents from ~300 financial regulators from Claude, ChatGPT, Cursor or your own agent — over MCP. Every document summarised in English, with the regulator's original a link away.

Free with an account
Fair use

Server URL · any MCP client

https://mcp.regalert.today/mcp

Sign in with your RegAlert account when the client asks — OAuth, nothing to paste. No API key to manage; the account page shows every read.

Add it in a minute

Pick your client. Each one takes the server URL above and asks you to sign in once. The Claude and ChatGPT buttons open the web settings; on a phone, add the connector there and the mobile app picks it up.

ChatGPT

Opens settings — then Connectors → Advanced → Developer mode, and add an MCP server with the server URL.

Open ChatGPT settings

Claude.ai / Claude Desktop

Opens the Add custom connector dialog — paste the server URL and sign in.

Add in Claude

Cursor

One click opens Cursor and adds the server; sign in when asked.

Install in Cursor

 ▐▛███▜▌
▝▜█████▛▘
  ▘▘ ▝▝

Claude Code

claude mcp add --transport http --scope user regalert https://mcp.regalert.today/mcp

VS Code

One click adds the server to VS Code's MCP list; sign in when asked.

Install in VS Code

{ }

Plain HTTP

GET /v1/documents/{uid} and GET /v1/account on https://mcp.regalert.today, with the same OAuth bearer token.

Try these

Once the connector is added, ask in plain language. Search is free; a read spends five cents.

“What did the Central Bank of Nigeria publish in the last two weeks? Summarise anything about capital or liquidity.”
“Search RegAlert for MAS guidance on outsourcing, then read the most recent one and list the obligations for a bank.”
“Which regulators in the EU published something about crypto-asset custody this month? Give me the documents and their dates.”

Tools

Seven tools, all read-only for the world outside RegAlert, all free with an account.

Tool	Cost	What it returns
`search_documents`	free	Full-text search with regulator, jurisdiction, region and topic filters. Each hit: title, date, regulator and the start of its summary. The total is exact.
`recent_documents`	free	What was published, newest first, with the exact total: by regulator, jurisdiction, region, topic and date range — or the newest of everything.
`list_regulators`	free	The regulators covered and the codes the other tools take. Finds a regulator by name or country.
`list_topics`	free	The topic tags, with the number of documents of each.
`get_lineage`	free	What a document implements, amends or replaces, and what amended or replaced it — the confirmed links, both directions.
`get_document`	free	One document: metadata, summary, lineage and the English text (markdown), in parts for a long one. With a query, only the passages that match. The original-language text on request.
`get_account`	free	Your access: whether full-text reads are available right now.

Access

Free with an account. Search, lineage and full-text reads cost nothing. Sign in, no card.

Fair use. Reads are for your own work. If an account reads at bulk scale, full-text reads pause until the next day; search, metadata and lineage keep working.

Bulk access. For bulk or commercial use, write to ask@regalert.today.

What you get, and what you don't

Every document, the day it appears

Captured from the regulator's official site, with the original file preserved, the text extracted (translated to English where needed, the original kept), dated, tagged by topic and summarised. The authoritative text is always the regulator's own, linked from every record.

Nothing of yours

We never see your conversation — only the calls the agent makes to us (privacy). Scraping the site is not a licensed path; the MCP server is. Building compliance software? We license emerging-market rule text with nightly exports, daily changes and the links between documents. Questions, bulk access, or a regulator we don't cover yet: ask@regalert.today.

The connector is the same corpus as the site — browse it here first if you want to see what your agent will be reading.

Add it in a minute Browse the documents