mcpField
Q
Qiniso

Qiniso

Remote
qinisolabs.github.ioAI & Agents1Updated 18w ago

About Qiniso

Qiniso is a local fact-validation library and MCP server for agents that need to check structured identifiers and values. It uses checksums and bundled reference data rather than live-data lookups.

Key features

  1. 01Checksum and curated-data validation
  2. 02Checks across eight data domains
  3. 0356 typed verification tools
  4. 04Local operation without network calls

Use cases

  1. 01Validate IBANs, VAT numbers, and phone numbers
  2. 02Check crypto addresses and financial identifiers
  3. 03Parse addresses, dates, and currencies

MCP support

Qiniso exposes verification tools to MCP clients through a hosted endpoint or a local stdio server. The same checks are also available as library functions without MCP.

Configuration

Remote servers need no local install. Pick your client, then copy the config or install in one click.

Endpointhttps://qiniso.qinisolabs.workers.dev/mcp
Copy install command
Claude CodeTerminal
Cursor.cursor/mcp.json
VS Code.vscode/mcp.json
Claude Desktopclaude_desktop_config.json
Windsurf~/.codeium/windsurf/mcp_config.json

Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.

Packages
npm@qinisolabs/qiniso0.5.0

Tools

56
ParameterTypeRequiredDescription
ibanstringYesThe IBAN to validate; spaces are ignored.

Details

From the project README
Qiniso

Qiniso

The deterministic fact-verification layer for AI agents.

Verified, trustworthy data tools for AI agents. "Qiniso" means "truth" in Zulu.

Website · npm · GitHub · MCP Registry


Agents confidently emit IBANs, phone numbers, domains, VAT numbers and crypto addresses that are subtly — and silently — wrong. Qiniso checks the structured facts an agent produces against checksums and curated authoritative data, so a bad value is caught instead of trusted.

On arbitrary identifiers, a frontier LLM validates them wrong ~91% of the time, cold and silently. Qiniso: 0%.

Install

npm i @qinisolabs/qiniso

Use as a library

Every check is a typed function — validate locally in one pass, no MCP required:

import { validateIban, validateVat, validatePhone } from "@qinisolabs/qiniso";

validateIban("GB82 WEST 1234 5698 7654 32");
// { valid: true, country: "United Kingdom", checkDigits: "82", ... }

validateVat("DE136695976");          // { valid: true, country: "Germany", ... }
validatePhone("020 7946 0123", "GB"); // { valid: true, e164: "+442079460123", ... }

Use as an MCP server (in Claude and other agents)

No install needed — add the hosted endpoint as a custom connector:

https://qiniso.qinisolabs.workers.dev/mcp

Or run it locally over stdio. Add this to your client's MCP config (e.g. claude_desktop_config.json):

{
  "mcpServers": {
    "qiniso": {
      "command": "npx",
      "args": ["-y", "-p", "@qinisolabs/qiniso", "qiniso-mcp"]
    }
  }
}

What it verifies — 56 tools across 8 domains

DomainChecks
IdentifiersIBAN, payment card (Luhn + brand), ISBN-13, VIN, GTIN/UPC/EAN barcodes (+ GS1 country)
Web / networkTLD & domain (IANA root zone), IP, UUID, URL, email
FinanceISIN, CUSIP, SEDOL, LEI, US ABA routing
CryptoEthereum (EIP-55), Bitcoin (Base58Check / Bech32) addresses
National & tax IDsBrazil CPF/CNPJ, South Africa ID, Spain DNI/NIE, India Aadhaar, Italy, Poland, Netherlands, Belgium, Nordics, Portugal, Turkey, China, Germany Steuer-IdNr, France NIR, Switzerland AHV, Mexico CURP, Croatia OIB, Romania CNP, Bulgaria EGN, Estonia, Czech/Slovak rodné číslo, EU/UK VAT
AcademicISBN-10, ISSN, ORCID
LocalePhone (global), date parsing, currency, holidays (~200 countries), UK VAT-by-date
AddressesUK/US address parsing

What it is not

  • Not a live-data provider — it verifies facts you give it; it does not return the current time, weather, or live FX.
  • Not a credential sink — it never asks for secrets or keys.
  • Not a registration check — it validates a VAT number's checksum, not whether it is live-registered (VIES); it confirms a TLD is real, not that a domain is registered.

Privacy

This tool runs locally on your machine and is built not to collect, store, or transmit your data — no analytics, no telemetry, no account. All reference data is bundled — no network calls, and nothing leaves your device. Full policy: https://qinisolabs.github.io/privacy.html.

License

Apache-2.0

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.