mcpField
S
Social Security COLA 2027 MCP

Social Security COLA 2027 MCP

Remote
github.com/theluckystrikeOther0Updated 1d ago
Repository

About Social Security COLA 2027 MCP

A Social Security MCP server for people estimating COLA changes, retirement ages and benefit limits. It combines annual figures with SSA retirement rules and calculates claiming-age adjustments and earnings-test withholding.

Key features

  1. 01COLA values and history
  2. 02Monthly benefit increase calculations
  3. 03Full retirement age by birth date
  4. 04Claiming-age benefit estimates
  5. 05Earnings limits and withholding calculations

Use cases

  1. 01Estimate a benefit after a COLA
  2. 02Check retirement age and age-62 reduction
  3. 03Compare benefits at different claiming ages
  4. 04Calculate earnings-test withholding

MCP support

Exposes Social Security lookup and calculation tools to MCP clients, using Federal Register notices and 20 CFR rules as sources.

Configuration

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

Endpointhttps://social-security-cola-2027-mcp.lipmichal.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.

Tools

Details

From the project README

social-security-cola-2027-mcp

Social Security COLA 2027, full retirement age and 2027 limits for AI assistants. The COLA paid from January 2027 is 3.6 percent. Give a birth date and get the exact full retirement age month, the cut at 62 and the gain at 70.

This is a Model Context Protocol (MCP) server. Every amount is copied from the SSA notices in the Federal Register and from 20 CFR part 404. It needs no API key.

Tools

ToolWhat it returns
get_social_security_colaThe COLA for a payable year (2027 is 3.6 percent, effective December 2026, paid January 2027) and the history back to 1996. Without a year it checks the Federal Register API for a newer notice.
calculate_cola_increaseWhat a monthly benefit becomes after the COLA, with SSA rounding to 10 cents and to the whole dollar.
get_full_retirement_ageFull retirement age for a birth date, the date and month it is reached, the earliest claim month at 62, the percent cut at 62, the spouse percent, delayed credits to 70 and the survivor full retirement age.
calculate_claiming_age_benefitThe monthly benefit for any claiming age from 62 to 70, from the full-retirement-age amount (PIA).
get_social_security_limitsFor 2023 to 2027: maximum taxable earnings (2027 is $192,300), earnings test limits (2027 is $25,560 and $67,920), earnings per work credit, SGA, trial work month, maximum federal SSI and PIA bend points.
calculate_earnings_testHow much the retirement earnings test withholds for a year and annual earnings.

Every result has structured JSON, a one-paragraph summary, as_of, source (the Federal Register notice or the eCFR section) and more (a page on mcpstamp.com).

What it knows that your model does not

  • The 2027 COLA (3.6 percent) and the 2027 amounts. SSA published them on 2026-10-26, after most models were trained.
  • The full retirement age rules as written. A January 1 birthday counts as the year before. You reach an age the day before your birthday. Before full retirement age you must be 62 for the whole month, so most people start the month after they turn 62.
  • The exact reduction and delayed-credit arithmetic with SSA rounding. The two worked examples in 20 CFR 404.410 and 404.313 are test fixtures.

Install

Hosted (Claude.ai custom connector, ChatGPT developer mode)

Paste this URL. No token, no sign-up. It works now.

https://social-security-cola-2027-mcp.lipmichal.workers.dev/mcp
Claude Code
claude mcp add --transport http ss-cola https://social-security-cola-2027-mcp.lipmichal.workers.dev/mcp
Claude Desktop

Claude Desktop adds remote servers under Settings, Connectors, Add custom connector. Paste the URL above.

Cursor

Add to ~/.cursor/mcp.json:

{
  "mcpServers": {
    "ss-cola": { "url": "https://social-security-cola-2027-mcp.lipmichal.workers.dev/mcp" }
  }
}
VS Code

Add to .vscode/mcp.json:

{
  "servers": {
    "ss-cola": { "type": "http", "url": "https://social-security-cola-2027-mcp.lipmichal.workers.dev/mcp" }
  }
}
Local stdio (after npm release)

The npm package is not published yet. After it is, run it locally with Node 20 or newer:

claude mcp add ss-cola -- npx -y social-security-cola-2027-mcp

Example prompts

  • "What is the Social Security COLA for 2027?"
  • "My benefit is $2,008 a month. What will it be in January 2027?"
  • "I was born on July 15, 1960. What is my full retirement age and how much do I lose if I start at 62?"
  • "My PIA is $2,400. What do I get if I claim at 64 and 6 months, or at 70?"
  • "What is the Social Security earnings limit for 2027, and how much is withheld if I earn $40,000?"
  • "What is the maximum taxable earnings for Social Security in 2027?"

Data sources

Full list with fetch dates in DATA_SOURCES.md.

Limitations

  • It does not read your earnings record. Use your SSA statement for your PIA.
  • The COLA increase from a net payment is an estimate within about a dollar, because SSA applies the COLA to the PIA before the Medicare premium is taken out.
  • Spousal and survivor amounts are shown as percents only. Family maximum and disability benefits are not modeled.
  • Yearly limits cover 2023 to 2027. COLA history covers 1996 to 2027.

More

More MCP servers that answer questions your assistant cannot answer alone are on mcpstamp.com.

Part of the Zovo MCP collection: https://mcp.zovo.one

Built by theluckystrike.

Development

npm install
npm test                              # builds, then runs the fixture tests and an offline stdio test
node scripts/stdio-transcript.mjs     # stdio round trip, writes .live/stdio-transcript.jsonl

License

MIT

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.