FauxGen MCP is a stdio MCP server that hands agents instant fictional test data — identities, addresses, phones, test cards, IBANs and more — across 66 locales.
Full personas: generate_identity returns a complete fake person (name, address, phone, email, job, test cards) with optional country and gender.
Locale-aware building blocks: generate_name, generate_address, generate_phone (national + international) for any supported country code.
Auth fixture staples: generate_email, generate_username, generate_password (length 6–128, toggleable upper/digits/symbols).
Payment-like test data: generate_credit_card (Luhn-valid test Visa/Mastercard/Amex) and generate_iban (mod-97 valid). Clearly not real instruments.
Company data: generate_company gives a fictional name, domain and catchphrase.
Device/network junk food: generate_uuid, generate_ip (v4/v6/both), generate_mac, generate_user_agent, plus generate_crypto_wallet (Bitcoin/Ethereum-looking strings).
Discovery: list_countries lists all country codes with dial codes and page URLs.
Web tool deep-links: fauxgen_tool_url returns fauxgen.com URLs for visual tools — WhatsApp/Telegram/SMS chat mockups, tweets, Instagram posts, license plates, barcodes, QR codes, plus generators.
Batching: most list-style tools accept count (1–50; user agents 1–20).
Extras: ships a seed_test_user prompt and a fauxgen://overview resource; all tools run synchronously with no task support.
Safety: output is explicitly fictional test data — not for deception, real accounts or payments. The README notes 66 countries matching the website.
Generates Luhn-valid Visa test card numbers (alongside Mastercard and Amex) so payment forms, input masks and card validators can be exercised with realistic-looking, non-real card data.
Exposes a deep link into fauxgen.com's fake WhatsApp chat generator, allowing agents to produce realistic WhatsApp conversation mockups/screenshots for demos and UI fixtures.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@FauxGen MCP seed a German checkout user with a fake address and test card"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Agents write forms, seeds and Playwright fixtures all day. They shouldn't invent SSNs from thin air or paste the same John Doe forever.
You: seed a German checkout user
Agent: → generate_identity({ country: "de" })
→ valid-looking DE address, phone, IBAN-ready country, test Visa…Related MCP server: misata-mcp
Add to .cursor/mcp.json (project) or ~/.cursor/mcp.json (global):
{
"mcpServers": {
"fauxgen": {
"command": "npx",
"args": ["-y", "fauxgen-mcp"]
}
}
}~/Library/Application Support/Claude/claude_desktop_config.json (macOS):
{
"mcpServers": {
"fauxgen": {
"command": "npx",
"args": ["-y", "fauxgen-mcp"]
}
}
}npx -y fauxgen-mcpTool | What you get |
| Full persona: name, address, phone, email, job, test cards |
| Locale-aware building blocks |
| Auth-fixture staples |
| Luhn-valid test Visa / MC / Amex |
| mod-97 test IBAN |
| Company + domain + buzzphrase |
| Device & network junk food |
| Random-looking BTC / ETH strings |
| All supported country codes |
| Deep-link into the visual web tools (WhatsApp chats, plates, QR…) |
Also ships a seed_test_user prompt and a fauxgen://overview resource.
66 countries — same coverage as the website.
All output is fictional test data. Do not use it to deceive people, open real accounts, or commit fraud. Test cards and IBANs pass checksums so your validators work — they are not real payment instruments.
npm install
npm run build
npm start
npm run typecheckMIT © FauxGen — fauxgen.com
Return the fauxgen.com URL for a visual tool (chat mockups, plates, barcodes, etc.). Use when the user needs screenshots or the full UI.
Inputs: tool (required), country.
Generate a street address for a country.
Inputs: country.
Generate a fictional company name, domain and catchphrase.
Inputs: country.
Generate a Luhn-valid TEST card number (Visa/Mastercard/Amex). Not a real account — never use for payments.
Inputs: brand.
Generate random-looking Bitcoin / Ethereum addresses for UI tests.
Inputs: chain.
Generate one or more fictional email addresses.
Inputs: count.
Generate a mod-97 valid TEST IBAN. Not a real bank account.
Inputs: country.
Generate a full fictional test identity (name, address, phone, email, job, test cards). For QA and demos only.
Inputs: gender, country.
Generate random IPv4 / IPv6 addresses.
Inputs: count, version.
Generate random MAC addresses.
Inputs: count.
Generate a locale-aware first/last name.
Inputs: gender, country.
Generate a random password with configurable character sets.
Inputs: upper, digits, length, symbols.
Generate a phone number for a country (national + international).
Inputs: country.
Generate realistic browser user-agent strings.
Inputs: count.
Generate fictional usernames.
Inputs: count.
Generate one or more UUID v4 values.
Inputs: count.
List all FauxGen country codes with dial codes and country page URLs.
Inputs: no named parameters.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Scored across 17 tools
Most generators target clearly distinct data types, so an agent can usually tell them apart. However, generate_identity overlaps substantially with generate_name, generate_email, generate_address, generate_phone, and generate_credit_card, which could cause hesitation when a full identity is not needed.
The dominant pattern is consistent snake_case with a generate_ prefix, which is predictable and easy to scan. The outliers list_countries and fauxgen_tool_url are minor deviations from the verb_noun convention but not enough to confuse selection.
17 tools is slightly above the typical 3-15 range, but the domain naturally supports many distinct fake-data categories. Each generator covers a different data type, so the count is reasonable rather than bloated.
The server covers core identity, financial, network, and crypto test data well, with list_countries and a UI URL helper filling peripheral needs. Minor gaps remain for common QA data such as lorem text, dates, numbers, and standalone job/DOB/SSN generators, but agents can often work around them using generate_identity.
300+ deterministic data utilities for AI agents: validate, normalize, parse, match, redact.
Generate AI images, video, voiceovers and music from Claude, ChatGPT or Cursor through 50+ models (Veo 3.1, Kling 3, Seedance, Nano Banana, GPT Image, ElevenLabs). Also image editing, upscaling, background removal, face swap, transcription, voice cloning and UGC-style video ads. Sign in with OAuth — no API key to paste. Tools are annotated (read-only vs. credit-spending); failed generations are refunded.
Email inboxes, webhooks, mock APIs, 2FA codes, approvals and 170 tools for AI agents.
People and property search for Claude and ChatGPT. Phones, emails, addresses, relatives, properties.
