Simulatte MCP Server
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@Simulatte MCP Server Test this sleep coaching concept on 50 stressed parents."
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Simulatte MCP Server
Run any of Simulatte's 26 SKUs from Claude, Cursor, Zed, or any MCP-compatible AI client. Get results back as structured JSON. No new UI to learn.
User: "Test this sleep coaching concept on 50 stressed parents."
Claude: [calls simulatte_estimate_cost] → 150 credits (~$1.80). Approve?
User: "yes"
Claude: [calls simulatte_run_study] → Results in 4 minutes.Installation
One-off (no install):
SIMULATTE_API_KEY=sim_live_your_key npx @simulatte-io/mcp-serverGlobal install:
npm install -g @simulatte-io/mcp-server
SIMULATTE_API_KEY=sim_live_your_key simulatte-mcpRelated MCP server: simengine-mcp
Get an API key
Create a key — starts with
sim_live_Set it as
SIMULATTE_API_KEYin your environment or MCP client config
Claude Desktop setup
Edit ~/Library/Application Support/Claude/claude_desktop_config.json:
{
"mcpServers": {
"simulatte": {
"command": "npx",
"args": ["-y", "@simulatte-io/mcp-server"],
"env": {
"SIMULATTE_API_KEY": "sim_live_your_key_here"
}
}
}
}Restart Claude Desktop. Simulatte tools appear automatically.
Cursor setup
Edit ~/.cursor/mcp.json (or your project's .cursor/mcp.json):
{
"mcpServers": {
"simulatte": {
"command": "npx",
"args": ["-y", "@simulatte-io/mcp-server"],
"env": {
"SIMULATTE_API_KEY": "sim_live_your_key_here"
}
}
}
}Zed setup
In ~/.config/zed/settings.json, add:
{
"context_servers": {
"simulatte": {
"command": {
"path": "npx",
"args": ["-y", "@simulatte-io/mcp-server"],
"env": {
"SIMULATTE_API_KEY": "sim_live_your_key_here"
}
}
}
}
}Available tools
Tool | What it does |
| Run any of 26 research SKUs — concept testing, pricing, messaging, B2B committee, and more |
| Retrieve full structured results for a completed study |
| Ask a natural-language question across your entire study history |
| List all synthetic persona pools in your workspace |
| Create a new persona pool with custom demographic/psychographic specs |
| Run a multi-turn simulated depth interview with a synthetic persona |
| Get a credit + USD cost estimate before running a study (local — no API call) |
SKUs supported by simulatte_run_study
concept-viability · claim-credibility · brand-identity-test · message-resonance · price-sensitivity · feature-priority · ad-copy · b2b-committee · conjoint · iris-pulse · card-sort · open-end · ab-backlog · polarization-stress-test · name-test · founder-positioning · ad-concept-resonance · depth-interview · custom-study · iat · counterfactual-positioning · personalization-sensitivity · regulated-claim-preflight · volume-forecast · brand-tracker · creative-audit
Example prompts
Concept test:
"I have a new sleep coaching app concept. Test it on 50 stressed parent personas and tell me if it'll land."
Pricing sensitivity:
"Run a price sensitivity study on my premium plan ($99/mo) against 100 millennial professionals. Use pool_abc123."
Cross-study synthesis:
"What are the most common objections across all of our B2B studies this quarter?"
Depth interview:
"Conduct a depth interview with a skeptical 35-year-old UK parent about our onboarding flow. 15 turns."
Authentication
Preferred: per-customer API key
Send your sim_live_* key in either of these headers — both are accepted:
x-api-key: sim_live_your_key_here
# or
Authorization: Bearer sim_live_your_key_hereDeprecated: shared platform key
The shared key forge-prod-2026 (used in internal integrations before v0.4) is
deprecated and will be removed at v0.5. If your integration sends
x-api-key: forge-prod-2026, migrate to a sim_live_* per-customer key now.
The worker logs a deprecation warning on every request that uses the shared key.
Response headers
Every successful /v1/forge/* and /v1/iris/* response includes these headers:
Header | Description |
| Credits charged for this request |
| Workspace credit balance after this charge |
| Present only when your key has consumed ≥ 80% of its spend cap. Value: |
Sample curl showing all headers
curl -X POST https://forge-worker-production.up.railway.app/v1/forge/concept-viability \
-H "x-api-key: sim_live_your_key" \
-H "x-workspace-id: your-workspace-uuid" \
-H "Content-Type: application/json" \
-d '{"concept":"A sleep app for stressed parents","population_id":"default","sample_size":20}' \
-iExpected response headers (2xx):
HTTP/2 201
x-simulatte-credits-used: 28
x-simulatte-credits-remaining: 472If your key is at 85% of its spend cap:
x-simulatte-spend-warning: 85%-of-cap-consumedError reference
Status |
| Meaning |
|
| Key not found, revoked, or expired |
| (message) | Bearer JWT invalid or expired |
|
| Key has a spend cap and this request would exceed it. Body: |
|
| 60 req/min or 1000 req/hr limit hit. Check |
|
| Worker DB not reachable — retry in 30s |
Sample error bodies
402 — spend cap exceeded:
{
"detail": {
"error": "key_spend_cap_exceeded",
"cap": 500,
"used": 498,
"would_charge": 10
}
}429 — rate limited:
{
"detail": {
"error": "rate_limited",
"limit": "60/min",
"retry_after_seconds": 42
}
}Retry-After: 42 header is also present.
Rate limits and quotas
Rate limits and credit caps are enforced per API key:
Default limit | Value |
Per-minute | 60 req/min |
Per-hour | 1,000 req/hr |
Spend caps are optional, set per-key by workspace admins. A key with no spend cap has unlimited spend (bounded only by workspace credit balance).
See app.simulatte.io/settings/billing for your current usage.
Troubleshooting
"Missing SIMULATTE_API_KEY" — Make sure the env block in your MCP config contains your key. The key must start with sim_live_.
Tool not appearing in Claude Desktop — Restart Claude Desktop after editing claude_desktop_config.json. Check the MCP logs at ~/Library/Logs/Claude/mcp*.log.
simulatte_get_results returns status: "running" — Studies take 2–8 minutes. Poll again in 30 seconds, or ask Claude to wait and retry.
401 Unauthorized — Your API key may be revoked or incorrect. Generate a new one at app.simulatte.io/settings/api-keys.
429 Too Many Requests — You've hit the rate limit for your tier. Upgrade at app.simulatte.io/settings/billing.
Development
git clone https://github.com/Iqbalahmed7/simulatte-mcp-server
cd simulatte-mcp-server
npm install
npm run build # compiles TypeScript → dist/
npm test # runs vitest test suite
npm run dev # runs directly via tsx (no build step)License
MIT — see LICENSE.
Available Tools
7 toolssimulatte_ask_insightsA
Ask a question across your entire Simulatte research history using semantic search. Returns a cited answer with themes and source study references. Great for synthesizing findings across multiple studies.
Inputs: query (required), top_k.
simulatte_create_poolA
Create a new synthetic persona pool with a custom demographic and psychographic specification. Returns a pool_id you can reuse across multiple studies.
Inputs: name (required), market (required), size (required), pool_spec (required).
simulatte_depth_interviewA
Run a simulated depth interview with a synthetic persona. The persona responds in character across multiple turns, surfacing motivations, objections, and language naturally. Returns interview_id and a credits estimate.
Inputs: goal (required), product_context (required), persona_id, persona_pool_id, +1 more.
simulatte_estimate_costA
Estimate the credit cost and USD price for a Simulatte study before running it. Calculated locally — no API call needed. Credits = base_cost_per_persona × sample_size.
Inputs: sku (required), sample_size, max_turns.
simulatte_get_resultsA
Retrieve results for a completed Simulatte study. Returns full structured results JSON including verdicts, key drivers, objections, persona breakdowns, and Forge Loop recommendations.
Inputs: study_id (required).
simulatte_list_poolsA
List all persona pools in your Simulatte workspace. Returns pool IDs, names, markets, and sizes. Use pool IDs in simulatte_run_study to target a specific audience segment.
Inputs: no named parameters.
simulatte_run_studyB
Run a Simulatte synthetic research study. Choose from 26 SKUs covering concept testing, pricing, messaging, B2B committee simulation, depth interviews, and more. Returns a study_id you can poll with simulatte_get_results.
Inputs: sku (required), inputs (required), persona_pool_id, sample_size.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 7 tools
Every tool has a clearly distinct purpose, covering different stages of the research workflow: setting up pools, running studies/interviews, getting results and insights, and estimating costs. No two tools have overlapping functionality.
All tools consistently use the 'simulatte_' prefix followed by a verb_noun pattern (e.g., create_pool, get_results), making names predictable and easy to understand.
With 7 tools, the set is well-scoped for the server's purpose of synthetic research. Each tool addresses a necessary step without being excessive or insufficient.
The tools cover the core lifecycle—create pools, run studies, get results, ask insights, estimate costs. A minor gap is the lack of deletion or cancellation tools, but the essential workflows are complete.
Maintenance
Related MCP Connectors
Run AlgoriQ quant research from any MCP client: data, features, strategies, validation, portfolios.
Run user and customer interviews with real people from your AI agent, by MCP or REST API. Vetted, fraud-screened participants from a 4M+ consumer panel or your own list. AI moderation that ladders into why. Voice, video or chat in 80+ languages. Findings come back cited to quotes.
Build and run AI workflows, agents, data tables and 1000+ API integrations from any MCP client.
Minds is a synthetic market research platform. This MCP server lets ChatGPT, Claude, and Cursor build Audiences from a brief, run durable Studies and Question Blocks, compare segments, and export results. OAuth 2.1 or Minds API key.
Related MCP Servers
AlicenseNot gradedqualityDmaintenanceEnables running LLM evaluations, experiments, and custom evaluators through a standardized MCP interface.16Apache 2.0- AlicenseNot gradedqualityDmaintenanceEnables running Monte Carlo simulations, safety constraint evaluation, and optimization from JSON domain configs. Provides AI agents with a tool to simulate uncertain futures and obtain structured results.36 PyPIApache 2.0

Minds MCP Serverofficial
AlicenseNot gradedqualityBmaintenanceMinds MCP enables AI assistants to run synthetic market research end to end using grounded synthetic customer panels for concept testing, message testing, and segment comparison, with durable server-side studies exporting to PDF, CSV, XLSX, JSON, and Markdown.4MIT- FlicenseNot gradedqualityBmaintenanceEnables AI agents and developers to run deterministic, multi-perspective synthetic consumer focus groups and analyze market sentiment through JSON-RPC/MCP tools.8-