sentientUI
Official MCP server for SentientUI — read experiment data and manage variants for your projects from any AI assistant.
What it can do
- Create Project: Create a NEW SentientUI project (onboarding). Returns the project id and its pk_ public key for the SDK. Requires an account login: this works when connected via OAuth (the hosted MCP
- List Projects: List all SentientUI projects for the authenticated account.
- Get Project Stats: Get health stats for a project: event volume, session count, agent calls, and status.
What data it sees
Do you need an account
No: the server works without sign-in
Official MCP server for SentientUI — read experiment data and manage variants for your projects from any AI assistant. Connect at https://api.sentient-ui.com/mcp or run npx @sentientui/mcp.
How to connect
How to connect sentientUI to Claude
The server address to paste into Claude settings:
https://sentientui--carlos-sanchez.run.tools
- Open Claude (claude.ai or the desktop app).
- Go to Settings → Connectors.
- Click Add custom connector.
- Paste the server address copied below into Remote MCP server URL and click Add.
- In a chat, click + → Connectors and switch the new connector on.
Custom connectors are available on Free, Pro, Max, Team and Enterprise plans (Free is limited to one). On Team and Enterprise an organization Owner adds the connector first under Organization settings → Connectors.
Authorization
After you click Add, a sign-in window for the service opens. Sign in with your own account and approve access. Claude never sees your password.
How to connect sentientUI to ChatGPT
The server address to paste into ChatGPT settings:
https://sentientui--carlos-sanchez.run.tools
- Open ChatGPT in a browser (chatgpt.com). A Plus, Pro, Business, Enterprise or Edu plan is required.
- Turn on developer mode once: Settings → Apps → Advanced settings → Developer mode.
- Open Settings → Connectors and click Create.
- Fill in the form: Name (anything), Description (one line about what the service does), MCP server URL (copy it below).
- Under Authentication choose OAuth if the service requires sign-in, otherwise None. Click Create.
- In a new chat open + → Apps/Connectors and enable the connector.
OpenAI has renamed this section before (Connectors → Apps/Plugins). If the label differs, search settings for "developer mode". On Business/Enterprise workspaces an admin must allow custom connectors first.
Authorization
On first use ChatGPT opens the service's sign-in window. Sign in and approve access.
How to connect sentientUI to Cursor
The server address to paste into Cursor settings:
https://sentientui--carlos-sanchez.run.tools
Fastest: click Open in Cursor below and confirm the prompt.
Manually:
- In Cursor open Settings → Cursor Settings → MCP and click Add new global MCP server.
- Paste the JSON copied below into
~/.cursor/mcp.json(per project:.cursor/mcp.jsonin the repo root). - Save the file. The server appears in the MCP list; authorize it there if asked.
Authorization
If the service needs sign-in, an authorize button appears next to the server in the MCP list.
Server tool list (16)
Raw names from tools/list. Only developers need these.
| create_project | Create a NEW SentientUI project (onboarding). Returns the project id and its pk_ public key for the SDK. Requires an account login: this works when connected via OAuth (the hosted MCP URL) but NOT with a project-scoped sk_ server key or an anonymous demo token. After it succeeds, call get_integration_guide and help the user install @sentientui/react with the returned key. |
| list_projects | List all SentientUI projects for the authenticated account. |
| get_project_stats | Get health stats for a project: event volume, session count, agent calls, and status. |
| list_components | List all adaptive components in a project with variant counts and impression totals. |
| get_variant_performance | Get CVR and momentum for all variants in a project over the last 7 days vs prior 7 days. |
| get_insights | Get the latest AI-generated insights: narrator observations and (Growth tier) advisor recommendations. |
| get_persona_breakdown | Get the distribution of visitor persona clusters with session counts and reliability scores. |
| get_goal_funnel | Get goal hit counts, unique-session conversion rates, and per-variant breakdown. |
| list_guardrail_events | List variants currently paused by the guardrail in the last 24 hours. |
| get_layout_stats | Get per-persona section layout rankings and bandit reward weights. |
| get_variant_brief | Get an insight-driven brief for creating a new CODE-NATIVE variant of a component. Returns current variant performance, audience, insights, a data-sufficiency assessment (with a best-practice fallback when there is no data yet), and step-by-step instructions for writing the variant in the customer's code. Use this instead of create_variant when the variant will live in the codebase. |
| get_test_brief | Get a ready-to-paste test for a SentientUI-wrapped component, populated with the component's real variants and goals. This project uses @sentientui/react/testing. Use this so your tests force a specific variant/layout deterministically and never break when the optimizer serves a different version. Returns a React Testing Library example plus the URL-param recipe for E2E (Playwright/Cypress). |
| create_variant | Create a NO-CODE managed text variant for a component (content stored in SentientUI, rendered by <AdaptiveText>). Use this ONLY for text-only variants the user wants without a code change. For variants that will live in the codebase (full components — copy, markup, styling), use get_variant_brief and write the variant in code instead; those auto-register on deploy and do not need create_variant. Requires a paid plan (server keys are Starter+; anonymous demo tokens are read-only). |
| pause_variant | Pause a variant, stopping traffic from being assigned to it. |
| refresh_insights | Trigger fresh AI insight generation for a project. Returns immediately; use get_insights in ~15 seconds to see results. |
| get_integration_guide | Get the SentientUI adaptive-ladder integration guide: setup (keyless and keyed) plus copy-pasteable examples for every rung (Style, Swap, Reorder). Use this to integrate SentientUI into a codebase. |