QuantRisk MCP server

• Registry namedev.quantrisk/mcp-serververified in the MCP registry

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

• Stateanswerschecked 2026-10-26
• Tools10
• Server calls itselfquantrisk
• Server version1.0.0
• Protocol version2025-06-18

On 2026-10-26 the server answered initialize and tools/list without a sign-in. First answer in 336 ms.

• Its tools do every use case we checked: estimate one-month value at risk for a stock portfolio; simulate quarterly outcome paths and the chance of loss; replay past market crises against current holdings to gauge drawdown.
• Checked on 1 of the 7 days it takes to judge reliability.

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.

• INITE MCP scoreafter a weeka score needs 7 daily checks: 1 so far
• What an agent can do with it3 of 3 use cases
• Use23 npm downloads a week2 GitHub stars
• Answeredmeasuring1 of the 7 daily checks it takes to print a share
• Median first answer451 ms
• Space in an agent's context≈ 4,584 tokensthe definitions of all its tools (10), which an agent carries on every turn
• Tool definitionsB · 83 of 100
• Typed results0 of 10 toolsdeclared outputSchema: results come typed, read once

• Not saying whether it only reads or can destroy: 10 of 10

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 (25%), doing the product's job (25%), definitions (20%), context (10%), use (10%) and how readily a client connects (10%). 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 →

• analyze_risk

  Calculate core risk metrics for a portfolio — Value at Risk (VaR), Conditional VaR (CVaR), volatility, beta, and max drawdown.

• monte_carlo_simulation

  Run Monte Carlo simulation on a portfolio to model the distribution of future returns, including percentile outcomes and probability of loss.

• stress_test

  Stress test a portfolio against historical crisis scenarios (GFC 2008, COVID 2020, etc.) or custom shocks (paid tier).

• optimize_portfolio

  Find the optimal portfolio allocation using mean-variance optimization. Supports max Sharpe, min variance, and target return objectives. Paid tier only.

• correlation_matrix

  Compute the pairwise correlation matrix for a set of assets. Identifies highly correlated pairs and diversification opportunities.

• performance_attribution

  Break down portfolio performance into factor exposures, sector allocation, and position contributions. Computes Sharpe, Sortino, Treynor, Calmar, and Information ratios.

• sector_exposure

  Break down portfolio exposure by GICS sector, market cap, and asset class. Returns concentration metrics including the Herfindahl-Hirschman Index.

• price_history

  Fetch historical OHLCV price data for one or more tickers. Free tier: 1 ticker, 252 days. Paid tier: up to 20 tickers, 1260 days.

• compare_portfolios

  Compare two or more portfolio allocations head-to-head across all key risk and return metrics. Paid tier only.

• calculate_greeks

  Calculate option Greeks (delta, gamma, theta, vega, rho) for individual options or an options portfolio. Uses Black-Scholes for European, binomial for American style. Paid tier only.