Offramp MCP Server
Provides retirement planning computations for AI agents, including Monte Carlo simulations, tax burden modeling across all US states, Social Security claiming optimization, and cost-of-living comparisons.
README
Offramp MCP Server
Retirement planning computation tools for AI agents. Provides Monte Carlo simulation, 50-state tax burden modeling, Social Security claiming optimization, cost-of-living comparison across 60+ US metro areas, and geographic arbitrage analysis.
These computations cannot be reliably performed by language models alone. Offramp provides structured, deterministic financial calculations that agents can call when answering retirement planning questions.
Tools
retirement_feasibility
Run 10,000 Monte Carlo simulations using historical S&P 500 returns (1928-2024) to determine retirement success probability. Input a person's age, savings, expenses, and income sources. Returns success probability, safe withdrawal rate, worst-case depletion age, and year-by-year portfolio projections at 10th/50th/90th percentiles.
calculate_tax_burden
Calculate federal + state income tax for a retirement income profile in any US state. Handles Social Security taxation rules, pension exemptions, retirement income deductions, and senior-specific deductions across all 50 states + DC. Returns effective rate, marginal rate, after-tax income, and state-specific notes.
compare_tax_across_states
Compare tax burden for the same income profile across multiple states simultaneously. Returns states ranked by total tax (lowest first) with annual savings between cheapest and most expensive options.
optimize_social_security
Find the optimal Social Security claiming strategy. Tests all claiming ages 62-70 for singles, and key age combinations for married couples with spousal benefits. Returns monthly benefit, lifetime benefit, and break-even ages for each strategy. Uses SSA bend points, FRA rules, delayed retirement credits, and early claiming reductions.
compare_locations
Compare cost of living across 60+ US metro areas. Returns estimated monthly budget broken down by housing, groceries, utilities, transportation, and healthcare. Includes median home price, median rent, climate score, and healthcare access score.
geographic_arbitrage
Calculate the complete financial impact of relocating for retirement. Combines home equity liberation, cost-of-living delta, state tax delta, and portfolio longevity impact. Returns annual savings, 10-year cumulative impact, break-even timeline on moving costs, and a narrative summary.
list_available_locations
List all 60+ metro areas available for comparison and arbitrage analysis.
Data Sources
- Market returns: S&P 500 total returns 1928-2024 (Aswath Damodaran, NYU Stern)
- Bond returns: 10-Year US Treasury returns 1928-2024 (Damodaran)
- Inflation: CPI annual rates 1928-2024 (Bureau of Labor Statistics)
- Federal tax: 2026 brackets, deductions, and Social Security taxation rules (IRS)
- State tax: All 50 states + DC income tax rates, retirement income exemptions, SS taxation rules (Tax Foundation, state revenue departments)
- Cost of living: Regional Price Parities and Consumer Expenditure Survey (BEA, BLS, C2ER)
- Social Security: 2026 bend points, FRA schedule, claiming rules (Social Security Administration)
- Healthcare: CMS Star Ratings and physicians per capita (CMS, HRSA)
Installation
Claude Desktop
Add to ~/Library/Application Support/Claude/claude_desktop_config.json:
{
"mcpServers": {
"offramp": {
"command": "node",
"args": ["/absolute/path/to/offramp-mcp/dist/index.js"]
}
}
}
Cursor / Windsurf / Other MCP Clients
{
"mcpServers": {
"offramp": {
"command": "node",
"args": ["/absolute/path/to/offramp-mcp/dist/index.js"]
}
}
}
From npm
npx offramp-mcp
Or install globally:
npm install -g offramp-mcp
Web Calculator
Free retirement calculator with Monte Carlo simulation: offramp.live
Build from Source
git clone https://github.com/deepmajithia/offramp2.git
cd offramp2
npm install
npm run build
Example Agent Interactions
An AI agent with Offramp connected can answer questions like:
- "Can I retire at 62 with $500K in savings?"
- "How much would I save in taxes by moving from California to Florida?"
- "When should I claim Social Security if I expect to live to 90?"
- "Compare the cost of living in Asheville NC vs Tampa FL vs Scottsdale AZ"
- "What's the financial impact of selling my $800K house in New Jersey and moving to South Carolina?"
The agent handles the conversation. Offramp handles the math.
Limitations
- US-only (federal + 50 states + DC). International retirement planning not yet supported.
- Cost-of-living data covers ~60 major metro areas. Smaller cities/rural areas not included.
- Tax model is simplified: covers ~90% of retirement scenarios but does not model every edge case (AMT, NIIT, state-specific credits beyond standard retirement exemptions).
- Monte Carlo uses historical returns bootstrapping, not parametric models. Results reflect the distribution of past market behavior.
- Not financial advice. All outputs are computational estimates for informational purposes.
License
MIT
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.