Model Context Protocol · deterministic · stateless

ScenarioSim MCP

A transparent, 100% deterministic what-if & scenario simulation engine for AI agents. Hand it assumptions — growth, churn, pricing, costs, a time horizon — and get back projected outcomes over time, sensitivity analysis, and break-even solving, each with a plain-language explanation.

Live endpoint:

Streamable HTTP
https://scenariosim-mcp.pages.dev/mcp

● live · free tier: 20 calls/day, no key required

Why agents need a simulation engine

deterministic

Identical inputs always yield byte-identical output. Every number runs through 40-digit decimal math — never floats, never drift over 100 periods.

transparent

Every result echoes back the exact assumptions used (defaults filled in), the methodology, per-period projections, and a natural-language explanation.

stateless

No database, no sessions, no hidden state. Pure functions over your assumptions. Trivial to host and reason about.

agent-friendly

One consistent JSON envelope for success and error. Errors are structured and actionable — never raw exceptions.

sensitivity + break-even

Not just a projection: sweep any input to see what moves the outcome, or solve for the value that hits your target.

9 templates + free-form

SaaS growth, pricing, churn, cost cuts, hiring, runway, unit economics, funnels — or bring your own metrics.

Six tools

ToolPurpose
run_scenarioMain tool: project a template or free-form model over time → projections, key_results, assumptions, explanation.
sensitivity_analysisVary one or more inputs and show the impact on a target metric, with elasticity + influence ranking.
break_evenSolve for the input value required to reach a target metric value (bisection).
compare_scenariosRun 2–3 scenarios side by side with deltas and an optional winner.
list_templatesDiscovery: every template with inputs, defaults, and available outputs.
health_checkVersion, status, and capabilities.

saas_growthpricing_changechurn_impact cost_reductionhiring_plancash_runway unit_economicsmarketing_funnelcompound_growth custom

Connect your client

Cursor
Claude Desktop
VS Code
curl

~/.cursor/mcp.json

json
{
  "mcpServers": {
    "scenariosim": {
      "url": "https://scenariosim-mcp.pages.dev/mcp"
    }
  }
}

claude_desktop_config.json — via mcp-remote bridge

json
{
  "mcpServers": {
    "scenariosim": {
      "command": "npx",
      "args": ["-y", "mcp-remote", "https://scenariosim-mcp.pages.dev/mcp"]
    }
  }
}

.vscode/mcp.json

json
{
  "servers": {
    "scenariosim": {
      "type": "http",
      "url": "https://scenariosim-mcp.pages.dev/mcp"
    }
  }
}
bash
curl -s https://scenariosim-mcp.pages.dev/mcp \
  -H 'content-type: application/json' \
  -H 'accept: application/json, text/event-stream' \
  -d '{"jsonrpc":"2.0","id":1,"method":"tools/call","params":{
        "name":"run_scenario","arguments":{
          "template":"saas_growth",
          "inputs":{"starting_customers":200,"new_customers_per_period":40,
                    "churn_rate":0.03,"arpu":60},
          "horizon":12}}}'

Example result

run_scenario → structured result (trimmed)
{
  "status": "success",
  "scenario": "saas_growth",
  "period_label": "month",
  "horizon": 12,
  "key_results": {
    "ending_customers": 449.7,
    "ending_mrr": 26982.1,
    "ending_arr": 323785.2,
    "total_churned_customers": 82.4,
    "cumulative_revenue": 232104.6
  },
  "projections": [
    { "period": 0, "customers": 200, "mrr": 12000 },
    { "period": 1, "customers": 234, "mrr": 14040, "new_customers": 40, "churned_customers": 6 }
  ],
  "assumptions_used": { "template": "saas_growth", "churn_rate": "0.03", "arpu": "60", "horizon": 12 },
  "methodology": { "model": "SaaS Growth", "deterministic": true, "precision": "decimal.js (40 significant digits)" },
  "explanation": "Starting from 200 customers and adding 40 per month (churn 3%), after 12 months you reach ..."
}

Headline numbers are rounded for readability; every key result also carries a full-precision value_exact string so no precision is lost.

Pricing

Start free to evaluate. When your agent needs real volume, upgrade in two clicks — pay by card, get an API key instantly, and send it as X-API-Key. Every plan has all six tools and all templates. Cancel anytime.

Free

$0

  • 20 calls / day
  • All 6 tools · all templates
  • Full transparency & explanations
  • No API key needed
  • Shared IP-based quota
Use free endpoint

For evaluation & low-volume testing.

Starter Popular

$12/mo

  • 5,000 calls / day
  • All 6 tools · all templates
  • Private API key + per-key quota
  • Usage endpoint
  • Manage/cancel via billing portal
Get Starter →

Best for a single production agent.

Pro

$39/mo

  • 50,000 calls / day
  • Everything in Starter
  • Highest priority
  • Manage/cancel via billing portal
  • Email support
Get Pro →

For fleets & high-volume workloads.

Secure checkout by Stripe · cards & wallets · promo codes supported · manage anytime at the billing portal. Prices in USD.

After purchase — add your key
{
  "mcpServers": {
    "scenariosim": {
      "url": "https://scenariosim-mcp.pages.dev/mcp",
      "headers": { "X-API-Key": "ssim_live_your_key_here" }
    }
  }
}

Prefer to self-host? The engine is open-source (MIT) and fully stateless — run your own copy on Cloudflare Pages (or any Node/Deno/Bun runtime) with unlimited calls and no keys.