Skip to content
MarketHeist logo

MarketHeist MCP server

Which servers in the official MCP registry are MarketHeist's own, whether they answer, and what they offer. A program checked each of these, nobody copied them from a product page.

read 2026-10-20

← Back to MarketHeist

Whose server it is
  • Registry nameio.marketheist/backtestverified in the MCP registry

The registry name sits under marketheist.io. The registry only lets a publisher use that namespace after proving control of the domain through DNS or HTTP.

When we asked itchecked 2026-10-20
  • Stateanswerschecked 2026-10-20
  • Tools10
  • Server calls itselfmarketheist
  • Server version1.3.0
  • Protocol version2025-06-18

On 2026-10-20 the server answered initialize and tools/list without a sign-in. First answer in 135 ms.

Should an agent connect itNot proven either way yet
  • Whether its tools do the product's job has not been judged yet.

Once connected

  • For "backtest a moving average or RSI strategy on any ticker", the likely first call is run_backtest.

Said by rules from the measurements above. The likely first call is a model's pick from the tool list; read its description before relying on it.

Measured dailydaily checks: 7, the first on 2026-10-20
  • INITE MCP score95 of 100answers 100 · definitions 100 · context 72
  • Answered100% of days7 of 7 daily checks
  • Median first answer184 ms
  • Space in an agent's context≈ 8,476 tokensthe definitions of all its tools (10), which an agent carries on every turn
  • Tool definitionsA · 100 of 100
  • Typed results0 of 10 toolsdeclared outputSchema: results come typed, read once
  • Every tool passes all six checks.

What an agent can do with it

  • not sure · test a proven allocation against your own portfolio weights
  • can do · backtest a moving average or RSI strategy on any ticker
  • not sure · measure drawdown and growth of a portfolio over 30 years

Each of the product's use cases, put to Jev against this server's tool list: can an agent do it with these tools?

Worth a look

None of the five warning signs in these definitions.

A program checks the server once a day: initialize and tools/list, no model. Tokens are the definitions' length at about four characters a token. The grade averages six checks of every tool's name, description, schema and annotations; it says how well the tools are described, not whether they work. The score weighs what was measured and nothing else: answering (30%), doing the product's job (25%), definitions (20%), context (10%) and use (15%). A part that was not measured is left out, and a hidden instruction caps the score at 40. The state of vendors' MCP servers: method and numbers →

Tools it offers10 tools · read 2026-10-20
  • portfolio_returns

    Look up how a portfolio, strategy or fund actually returned: seasonality by calendar month (average return, share of positive months, number of years and a t-statistic so a thin 'September effect' is not presented as a law), calendar-year…

  • portfolio_risk

    Answer both 'how bad did it get' and 'is it really diversified' for a built-in model portfolio, a canonical strategy or your own portfolio. Drawdowns: the deepest peak-to-trough episodes with months to trough and recovery, Ulcer index…

  • compare_portfolios

    Compare 2 to 4 portfolios (built-in model portfolios, canonical strategies or your own allocations) side by side over their COMMON window, which is the only fair comparison: two pages' headline numbers usually cover different periods (a…

  • analyze_portfolio

    Analyze an asset-allocation ('lazy') portfolio and get long-run performance computed from real monthly price history (proxy-extended for decades of data), not estimated: CAGR, max drawdown, volatility, Sharpe, Sortino, Ulcer index…

  • find_guides

    Search MarketHeist's 90 published research pages and get the page's own verdict and numbers, with its URL to cite: 18 strategy guides (60/40, All Weather, Golden Butterfly, DCA vs lump sum, ...), 26 fund-history pages (how far back an ETF…

  • list_portfolios

    List every portfolio and strategy the other tools accept by id, with a one-line description, holdings, headline CAGR / max drawdown / Sharpe with the start month and the data month, and the marketheist.io page and guide URLs. Three kinds…

  • fund_history

    Tell how far back a fund can be tested on MarketHeist once its history is extended with older substitute funds: the proxy chain (e.g. VXUS <- VGTSX), the extended start month, the ETF's own first month and, for the 26 funds with a history…

  • decompose_factors

    Explain WHAT DRIVES the returns of a ticker, ETF or built-in portfolio by decomposing them into common factor exposures (market, size, value, momentum, quality, low-volatility, duration, credit) plus an idiosyncratic residual. Use this…

  • run_backtest

    Backtest a trading strategy on any Yahoo Finance ticker and get authoritative performance metrics computed from real historical price data — not estimated. Use this whenever the user asks how a strategy or indicator would have performed…

  • list_indicators

    List the built-in technical indicators available for backtesting (RSI, moving-average crossovers, ADX, Bollinger, CCI, Stochastic, and more) with their IDs, parameters with defaults, and each indicator's default long rule. Call this to…

How to connect

Add the server to your MCP client config. Claude Desktop, Cursor and most other clients read this shape:

{
  "mcpServers": {
    "backtest": {
      "type": "http",
      "url": "https://api.marketheist.io/api/mcp"
    }
  }
}

Add it

Claude Code, in a terminal

claude mcp add --transport http marketheist https://api.marketheist.io/api/mcp

Add to CursorAdd to VS Code

claude.ai, ChatGPT and other clients: add https://api.marketheist.io/api/mcp as a custom connector in the settings.

Each line is something a program found when it went looking: the registry entry, the server's own answer, a spec that parses. A missing line means we did not find it, not that it does not exist.