A public MCP server for this site

MCP (Model Context Protocol) is the standard way an AI assistant, like Claude or ChatGPT, calls tools that live outside it.

This server gives your assistant the retirement simulator, Warren Buffett’s shareholder letters, and every post on this site. It’s read-only and needs no sign-in. Connect your assistant

https://api.mays.co/mcp

MCP inspector

Send request calls api.mays.co from your browser

Computed when this site was built

92.71%

At a 4% withdrawal rate and 60% stocks, 92.71% of 100,000 simulated retirements lasted 30 years. Median failure year: 26.

Simulation inputs

Request

> POST https://api.mays.co/mcp> Content-Type: application/json> Accept: application/json, text/event-stream> MCP-Protocol-Version: 2026-07-28> Mcp-Method: tools/call> Mcp-Name: simulate_retirement
{ "jsonrpc": "2.0", "id": 1, "method": "tools/call", "params": { "name": "simulate_retirement", "arguments": { "withdrawal_rate_percent": 4, "years": 30, "stock_share_percent": 60 }, "_meta": { "io.modelcontextprotocol/protocolVersion": "2026-07-28", "io.modelcontextprotocol/clientInfo": { "name": "mays.co-inspector", "version": "1.0.0" }, "io.modelcontextprotocol/clientCapabilities": {} } }}

Response

{  "jsonrpc": "2.0",  "id": 1,  "result": {    "structuredContent": {      "inputs": {        "withdrawal_rate_percent": 4,        "years": 30,        "stock_share_percent": 60      },      "paths": 100000,      "success_rate_percent": 92.71,      "failed_paths": 7286,      "median_failure_year": 26,      "cumulative_failure_percent_by_year": [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0.02, 0.04, 0.09, 0.18, 0.34, 0.55, 0.83, 1.21, 1.63, 2.13, 2.71, 3.37, 4.12, 4.89, 5.7, 6.51, 7.29],      "model": {        "returns": "real (inflation-adjusted) annual",        "stockMean": 0.08,        "stockSd": 0.18,        "bondMean": 0.025,        "bondSd": 0.08,        "correlation": 0.1      },      "simulator_url": "https://mays.co/tools/retirement-simulator"    },    "content": [      {        "type": "text",        "text": "… (same JSON as structuredContent, 511 chars)"      }    ]  }}
Protocol details

Stateless Streamable HTTP on AWS Lambda. Protocol 2026-07-28 has no handshake and no session, so each request carries its version and client info in _meta.

Connect your assistant

Connect it to Claude, ChatGPT, Cursor, or VS Code. Your assistant picks a tool when a question needs one.

Claude Code

Run this in a terminal:

claude mcp add --transport http mays-co https://api.mays.co/mcp

Start claude and ask:

Claude (web and desktop)

  1. In Claude on the web or the desktop app, open Customize › Connectors › Add custom connector.
  2. Paste the URL.

    https://api.mays.co/mcp
  3. Choose No sign in.

The Free plan allows 1 custom connector. On Team and Enterprise plans, an Owner must add it.

Then ask one of the example questions below.

ChatGPT

  1. Turn on Developer mode in ChatGPT’s settings.
  2. Add an app and paste the URL.

    https://api.mays.co/mcp

Developer mode is a beta for paid plans, on the web only. If the menu moved, search settings for Developer mode.

Then ask one of the example questions below.

Cursor

Add to Cursor

Cursor opens and asks you to confirm.

Or add it by hand to .cursor/mcp.json:

{"mcpServers":{"mays-co":{"url":"https://api.mays.co/mcp"}}}

Then ask one of the example questions below.

VS Code

Install in VS Code

VS Code opens and asks you to install the server.

Or add it by hand to .vscode/mcp.json:

{"servers":{"mays-co":{"type":"http","url":"https://api.mays.co/mcp"}}}

Then ask one of the example questions below.

Tools and prompts

Tools

Your assistant picks a tool when a question needs one. The inspector above runs 3 of them.

  • simulate_retirement

    Runs the Monte Carlo retirement simulator for a withdrawal rate, a time span, and a stock share.

    Parameters

    • withdrawal_rate_percent number, 0–10
    • years integer, 1–50
    • stock_share_percent number, 0–100
  • search_buffett_letters

    Searches 25 Berkshire Hathaway shareholder letters, 1998–2023, and returns passages with their year.

    Parameters

    • query string, 3–500 characters
    • limit integer, 1–10, default 5

    Ask:

  • list_posts

    Lists the posts on this site with their title, date, category, and tags.

    Parameters

    • none

    Ask:

  • get_post

    Returns one post as Markdown, with links that work outside this site.

    Parameters

    • slug string

    Ask:

  • about_steven

    Returns a short profile of Steven Mays and his work.

    Parameters

    • none

    Ask:

Prompts

Your client lists these in its prompt or slash-command menu.

  • stress_test_retirement

    Guides your assistant through simulator runs that test how a retirement plan holds up.

    Arguments

    • age string
    • savings string
    • annual_spending string

    Ask:

  • ask_warren

    Has your assistant answer from Buffett’s letters and cite the year of each passage it uses.

    Arguments

    • question string

    Ask:

How it’s built

The inspector on this page sends the same requests the server’s automated tests use, so the page can’t drift from the server.

No tool on the server asks a model to write anything. The server computes and retrieves, and your assistant writes every answer.

  1. Your assistant Claude, ChatGPT, Cursor, or VS Code
  2. POST /mcp One request, no session
  3. MCP server AWS Lambda. No model runs here.
  4. Tool code The Rust/WASM simulator, letter search, posts
  5. result
  6. Your assistant’s model Writes the answer

I build MCP servers like this one for internal systems: ERP, product, and analytics data. Contact me