User Intuition
What it does
Qualitative research with real people: AI-moderated interviews, 4M+ vetted panel, 80+ languages.
- Connects 电脑或云端
- Account or key Cloud: Sign in to the service when you connect (OAuth) · Local: No key needed
- Category 搜索与 AI 记忆
安装
选择你使用的 AI 应用,然后复制设置。
通过 5/5 项检查 · 你的提问会经过开发者的服务器(mcp.userintuition.ai) · ✓ 命令已检查 2026年11月13日
第一步:你用哪个 AI?
选择你用的应用,步骤会显示在下方。
可连接 Claude · 约 2 分钟
不需要密钥,不用安装程序 · 在 claude.ai 或桌面应用设置一次,即可在聊天中使用 · 免费账户可添加 1 个 · Team/Enterprise 方案由所有者(Owner)添加
打开 Claude 的 Connectors 页面
CustomizeConnectors
点击“Add” → “Custom” → “Web”,填写这 2 项
把这里的内容复制到 Claude 中同名的输入框,然后点击 Continue
Name · 名称User IntuitionRemote MCP server URL · 地址https://mcp.userintuition.ai/mcp在 Claude 检测到的登录设置页点击 Continue · Authentication 按作者说明选择(Sign in now、Sign in when needed 或 No sign in),然后点击 Add
在聊天中开启
+ConnectorsUser Intuition
试一试
用 User Intuition 告诉我你能帮我做什么,然后做一件给我看✓ 如果 Claude 说正在使用 User Intuition,或回答中出现 User Intuition 的工具名称,就表示连接成功
菜单名称已于 2026年11月11日 对照 Anthropic 帮助页面核实 · 你的界面可能因版本略有不同
能否连接 ChatGPT 取决于方案
完整 MCP:网页版 ChatGPT Business、Enterprise 或 Edu · Pro:以读取/获取权限连接,修改数据的工具可能用不了 · Plus:官方文档尚未确认,如果在插件页看到添加自定义 MCP 的选项可以试试 · 仅限网页
在网页版 ChatGPT 打开添加 MCP 的页面
Plugins+Add custom MCP server
填写名称和地址
输入名称,在 Connection 下粘贴地址,勾选风险提示,然后点击 Create as a plugin
Name · 名称User IntuitionConnection · 地址https://mcp.userintuition.ai/mcp在聊天中选择 User Intuition
开启新聊天,输入前先从插件列表中选择 User Intuition
试一试
用 User Intuition 告诉我你能帮我做什么,然后做一件给我看✓ 如果 ChatGPT 说正在使用 User Intuition,或回答中出现 User Intuition 的工具名称,就表示连接成功
菜单名称已于 2026年11月11日 对照 OpenAI 帮助页面核实 · 你的界面可能因版本略有不同
可连接 Claude Code · 约 2 分钟
不需要密钥,不用安装程序
打开终端
Mac:按 Command+空格,输入 Terminal · Windows:打开 PowerShell
粘贴这条命令并按 Enter
claude mcp add --transport http --scope user research https://mcp.userintuition.ai/mcp在 Claude Code 中输入 /mcp
应能看到 User Intuition 在列表中并显示已连接
试一试
用 User Intuition 告诉我你能帮我做什么,然后做一件给我看✓ 如果 Claude Code 说正在使用 User Intuition,或回答中出现 User Intuition 的工具名称,就表示连接成功
菜单名称已于 2026年11月11日 对照 Anthropic 帮助页面核实 · 你的界面可能因版本略有不同
可连接 Cursor · 约 2 分钟
不需要密钥,不用安装程序
在 Cursor 中以 Agent 模式打开 AI 聊天
试一试
用 User Intuition 告诉我你能帮我做什么,然后做一件给我看✓ 如果 Cursor 说正在使用 User Intuition,或回答中出现 User Intuition 的工具名称,就表示连接成功
菜单名称已于 2026年11月11日 对照 Cursor 帮助页面核实 · 你的界面可能因版本略有不同
可连接 VS Code · 约 2 分钟
不需要密钥,不用安装程序
在 VS Code 中以 Agent 模式打开 AI 聊天
试一试
用 User Intuition 告诉我你能帮我做什么,然后做一件给我看✓ 如果 VS Code 说正在使用 User Intuition,或回答中出现 User Intuition 的工具名称,就表示连接成功
菜单名称已于 2026年11月11日 对照 VS Code 帮助页面核实 · 你的界面可能因版本略有不同
其他支持 MCP 的应用(如 Windsurf、Zed、Gemini CLI)添加方式各不相同,请在应用列表中查看
通过网页连接的应用使用的地址
https://mcp.userintuition.ai/mcpJSON 设置在下方“开发者专用”中
开发者专用:终端命令和 JSON 设置
通过网页连接
Claude Code
claude mcp add --transport http --scope user research https://mcp.userintuition.ai/mcpCursor (~/.cursor/mcp.json)
{
"mcpServers": {
"research": {
"url": "https://mcp.userintuition.ai/mcp"
}
}
}VS Code (mcp.json)
{
"servers": {
"research": {
"type": "http",
"url": "https://mcp.userintuition.ai/mcp"
}
}
}安装在电脑上
Claude Code
claude mcp add --scope user research -- npx -y @userintuition-ai/mcpClaude Desktop (claude_desktop_config.json)
{
"mcpServers": {
"research": {
"command": "npx",
"args": [
"-y",
"@userintuition-ai/mcp"
]
}
}
}Cursor (~/.cursor/mcp.json)
{
"mcpServers": {
"research": {
"command": "npx",
"args": [
"-y",
"@userintuition-ai/mcp"
]
}
}
}VS Code (mcp.json)
{
"servers": {
"research": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@userintuition-ai/mcp"
]
}
}
}这些设置是根据作者在注册中心的条目生成的。如果 README 的写法不同,请以 README 为准。
一览
哪些应用可以使用它
依据这个服务器的连接方式和各应用自己的文档,检查于 2026年11月10日。
通过互联网连接,适用于
ClaudeChatGPTGooseLM StudioMsty StudioCherry StudioCursorVS Code + GitHub CopilotVisual StudioWindsurf (Devin Desktop)ZedKiroClineContinueJetBrains AI AssistantClaude CodeCodex CLIGemini CLIOpenCodenodecodeWarpLibreChatOpen WebUIRaycast
在你的电脑上运行,适用于
ClaudeGooseLM StudioMsty StudioCherry StudioCursorVS Code + GitHub CopilotVisual StudioWindsurf (Devin Desktop)ZedKiroClineContinueJetBrains AI AssistantClaude CodeCodex CLIGemini CLIOpenCodenodecodeWarpLibreChatRaycast
它能做什么
服务器要求先登录,才会列出它的工具。
基础检查i
我们检查的是任何人都能看到的信息;我们没有阅读代码。不适用的检查项不计入总数。 · 检查于 2026年11月14日
- ✓ 通过
作者与仓库一致
注册名称已通过域名 userintuition.ai 验证(注册中心会核实域名所有权)
- ✓ 通过
仍在维护
代码最后更新于前天
- ✓ 通过
有许可证
使用 MIT
- ✓ 通过
端点有响应
在 1877 毫秒内响应(HTTP 401)
- ✓ 通过
使用 OAuth 登录
声明支持 OAuth 登录,你的 AI 应用可以带你去登录
- … 尚未检查
工具描述中没有隐藏指令
尚未检查;我们会在检查端点时读取工具列表
讨论
提问、分享技巧或报告无法使用。按“有帮助”数量排序。
用户评价
还没有评价
与“搜索与 AI 记忆”中的其他服务器比较i
- 星标数多于此分类中其他 6092 个服务器的 0%
- 发布到注册中心的时间比此分类中其他 6092 个服务器的 90% 更新
- 基础检查结果等于或优于此分类中其他 6092 个服务器的 100%
与注册中心里这个分类中所有活跃的服务器比较。只有当发布者拥有该仓库时,星标数才计入。
常见问题
User Intuition 安全吗?
它在 5 项适用的基础检查中通过了 5 项。我们没有阅读代码。在添加密钥或令牌之前,请先看看这个服务器要求哪些权限,并使用只具备所需权限的密钥。
它需要哪些密钥或代码?
注册中心没有列出必需的密钥,不过有些服务器会在首次使用时要求你登录。
它在我的电脑上运行还是在云端?
两种方式都可以:安装在你的电脑上,或者连接云端版本。
它还在维护吗?
GitHub 仓库最后更新于前天,最新版本(1.6.3)于4天前进入注册中心。
哪些应用可以使用它?
支持 MCP 的应用,例如 Claude Desktop、Claude Code、Cursor 和 VS Code。它通过 URL 连接,因此也能用于接受远程 MCP 的应用,例如网页版 ChatGPT(Business 及以上为完整 MCP;Pro 为只读)。
有什么可以替代它?
同一分类中有:Context7, Codebase Memory, OpenSEO
同类别的其他选择
同一分类中其他作者的服务器,按星标数排序。
如何看卡片:↓ 每周下载量i★ GitHub 星标i三大支柱:评价 · 检查 · 使用情况i更新 = 最新版本发布到注册中心的时间i
Codebase Memory
Codebase knowledge graph for AI agents — 162 languages, sub-ms queries, 99% fewer tokens.
OpenSEO
Keyword research, site audits, rank tracking and AI visibility for your AI agent.
作者的 README
打开 README
按原文从作者的仓库复制。PopNx MCP 没有审查或背书其内容。HTML、图片和脚本已被移除。链接会离开本站。
User Intuition: run real qualitative research from your AI agent
Qualitative research with real people: AI-moderated interviews, 4M+ vetted panel, 80+ languages. Run user interviews, customer interviews, concept tests and prototype tests from your AI agent.
Ask your agent to run a concept test with 20 frozen-dinner buyers, find out why shoppers trade down to a store brand, or learn why trial users of your app don't activate. It drafts the study, shows you the plan and the cost, and once you confirm, it fields AI-moderated interviews (typically 20–30 minutes) and returns findings with verbatim quotes. Interviews ladder 5–7 levels deep, from what people did to why it matters to them. Participants come from a vetted 4M+ consumer panel or from your own customers, by voice, chat or video in 80+ languages. No account yet? Ask for help with any research question: the user-research skill plans the study or synthesizes research you already have, then fields it with real people when you're ready.
These are Agent Skills for running qualitative research with User Intuition: create interview studies, design screeners, field panel recruits, and turn transcripts into stakeholder-ready findings. The plugin bundles the hosted User Intuition MCP server for the *tools* and the skills for the *play*: the workflow, guardrails, and order of operations that make research runs reproducible.
Works with: Claude Code (plugin), Claude Desktop and claude.ai (connector and skill ZIPs), ChatGPT (plugin), Codex, Cursor, and any agent that reads Agent Skills.
Guides: Set up User Intuition in Claude · Skills docs · How quality is controlled · Pricing
Example run
This is the shape of a run. The prompt is illustrative; the findings come from the public Walmart shopper study.
- Prompt. "Find out where Walmart shoppers get stuck on a trip, and whether it changes whether they come back."
- Saved plan. The agent drafts the study from the brief (audience, screener, discussion guide) and shows you the saved plan before anything is fielded.
- Estimate. For a panel study it gets a quote and stops. Nothing launches, and nothing is spent, until you approve the plan and the quote. The cost depends on audience and size; see pricing.
- Findings excerpt. From 43 interviewed shoppers, each finding linked back to what participants said:
- 34 of 43 hit at least one obstacle on their trip.
- 36 of 43 will shop there again.
- The shelf carried 20 of 51 friction moments.
- 19 of 43 named the shelf as the moment that shaped the trip.
Install
Every surface uses the same hosted MCP server (https://mcp.userintuition.ai/mcp, browser sign-in) and the same skill files. After installing, ask your agent: "Help me get started with User Intuition." The setup skill checks the connection and routes you to the right workflow.
| Agent | Skills | MCP connection |
|---|---|---|
| Claude Code | /plugin marketplace add user-intuition/skills, then /plugin install user-intuition-research@user-intuition | Bundled with the plugin. Run /mcp, select user-intuition, and sign in. |
| Claude Desktop / claude.ai | Download a skill ZIP from Releases and upload it in Settings → Capabilities → Skills. One ZIP per skill; the bundle ZIP holds all of them. | Settings → Connectors → add a custom connector with https://mcp.userintuition.ai/mcp. |
| Codex | npx skills add user-intuition/skills -a codex | codex mcp add user-intuition --url https://mcp.userintuition.ai/mcp, then codex mcp login user-intuition. |
| Cursor | Install the plugin defined in .cursor-plugin/plugin.json, or npx skills add user-intuition/skills -a cursor | Bundled with the plugin, or add https://mcp.userintuition.ai/mcp under Settings → MCP. Sign in when prompted. |
| ChatGPT | Install User Intuition from the ChatGPT plugin directory in one click. The skills ship inside the plugin. | Bundled with the plugin. Sign in when prompted. Without the plugin, add https://mcp.userintuition.ai/mcp as a custom MCP connector in Settings → Apps & Connectors (developer mode). |
| Any other agent | npx skills add user-intuition/skills (pick your agent when asked, or pass -a ) | Add https://mcp.userintuition.ai/mcp as a remote MCP server in your client. |
Install a single skill with npx skills add user-intuition/skills -s create-study-from-brief. To build the Desktop ZIPs yourself, run node scripts/build-desktop-zip.mjs (output in dist/).
Raw files
Each skill is a standalone Markdown file under plugins/user-intuition-research/skills/ /SKILL.md. You can also tell your agent:
Fetch https://raw.githubusercontent.com/user-intuition/skills/main/catalog.json, select the workflow needed for the task, and fetch its raw
SKILL.md. Install it in the skill directory supported by your agent client.
Connect your account
The Claude plugin includes the hosted MCP connection:
https://mcp.userintuition.ai/mcp
After installing the plugin, open /mcp, select user-intuition, and complete the browser-based OAuth flow. Claude stores and refreshes the resulting token securely.
If you prefer a local stdio connection instead of the bundled hosted connection, disable the plugin-provided MCP and use:
claude mcp add userintuition -- npx -y @userintuition-ai/mcp@latest
with USERINTUITION_API_KEY set (get a key from your dashboard).
The skills
| Skill | Outcome |
|---|---|
setup | Check the connection, ask what you want to learn, and route to the right workflow |
user-research | Plan a study or synthesize research you already have, no account needed; then field it with real people |
interview-category-shoppers | Interview category or brand shoppers recruited from the consumer panel by what they buy |
interview-brand-switchers | Hear why shoppers switched brands or traded down, recruited from the consumer panel |
map-shopper-journey | Map the path to purchase, from trip trigger to shelf or page to first use |
run-usage-and-attitudes | Usage and attitudes: occasions, habits, needs, and brand repertoire |
interview-recent-buyers | Interview people who just bought, from Shopify or any store's orders |
interview-lapsed-customers | Find customers who stopped buying and hear why |
interview-cancellations | Interview subscribers soon after they cancel, from Stripe or any billing data |
run-win-loss | Interview buyers on closed won and lost deals from HubSpot or any CRM |
investigate-drop-off | Interview the people behind a funnel drop-off from your analytics tool |
create-study-from-brief | Turn a research brief into a launch-ready interview study |
design-screeners | Recruit exactly the right respondents — panel and custom screeners, done safely |
field-a-panel | Recruit N respondents from the panel, with cost approved before any spend |
invite-your-own-participants | Interview your own customers — invites, share links, and rewards |
run-a-concept-test | Put a concept, prototype, or landing page in front of participants |
monitor-fielding | Daily digest of every fielding study — progress, quality, stalls |
analyze-completed-study | From transcripts to stakeholder-ready findings, with real quotes only |
curate-interview-quality | Review low-quality interviews and delete only the exact interviews the user confirms |
retrieve-study-results | Read an existing report and follow source references without regeneration |
search-research | Find authorized prior evidence and inspect coverage and sources |
The original eight fielding and analysis workflows retain their validated MCP behavior. Retrieval and search use current tool/schema discovery; illustrative B2/C1 adapters still require release verification. Check any named tool yourself:
npx -y @userintuition-ai/mcp describe create_study
Machine discovery
Catalog: names, descriptions, raw file URLs, documentation URLs, and SHA-256 digests. Each skill is independently fetchable; loading the catalog does not require loading every workflow.
Contributing
Maintainer workflow (catalog, tests, docs and website export) is in CONTRIBUTING.md. Release notes are in CHANGELOG.md.
Docs
- Skills overview
- Skills library (browsable version of these files)
- MCP server reference
License
MIT