MCP server
The Model Context Protocol server for Clamp. Read-only by default, write actions on Pro, works with any MCP client.
Setup
VS Code / Copilot
Add to .vscode/mcp.json in your project:
{
"servers": {
"clamp": {
"type": "stdio",
"command": "npx",
"args": ["-y", "@clamp-sh/mcp"],
"env": {
"CLAMP_API_KEY": "sk_proj..."
}
}
}
}Claude Code
claude mcp add clamp -- npx -y @clamp-sh/mcp
export CLAMP_API_KEY="sk_proj..."Claude Desktop
Add to claude_desktop_config.json:
{
"mcpServers": {
"clamp": {
"command": "npx",
"args": ["-y", "@clamp-sh/mcp"],
"env": {
"CLAMP_API_KEY": "sk_proj..."
}
}
}
}Cursor
Add to .cursor/mcp.json:
{
"mcpServers": {
"clamp": {
"command": "npx",
"args": ["-y", "@clamp-sh/mcp"],
"env": {
"CLAMP_API_KEY": "sk_proj..."
}
}
}
}API key
Get your API key from the Clamp dashboard under Settings → API Keys. Keys are scoped to a project and start with sk_proj. The MCP server auto-detects which project to query from the key — no project ID needed.
Tools
Free (all plans)
| Tool | What it does |
|---|---|
get_overview | Pageviews, visitors, sessions, bounce rate, avg duration. Period comparison included. |
get_top_pages | Most visited pages ranked by pageviews. |
get_top_referrers | Traffic sources with channel classification (organic_search, direct, referral, paid, email, organic_social). |
get_countries | Visitors by country (ISO 3166-1 alpha-2 codes). |
get_cities | Visitors by city, with optional country filter. |
get_devices | Breakdown by device type, browser, or OS. |
get_events | Custom event counts with property filtering and grouping. |
get_timeseries | Event counts over time with automatic granularity. |
Pro Pro
| Tool | What it does |
|---|---|
create_funnel | Define and immediately evaluate a multi-step conversion funnel. |
get_funnel | Retrieve a funnel with cohort filters (country, channel, device, UTM). |
create_alert | Set up metric alerts (e.g. "visitors drops_by 30% over 7d"). |
Common parameters
period — "today", "yesterday", "7d", "30d", "90d", or a custom range as "YYYY-MM-DD:YYYY-MM-DD". Defaults to "30d".
limit — Max rows returned, 1–50. Defaults to 10.
Filters — Most tools accept pathname, utm_source, utm_campaign. Referrers accept channel. Funnels accept cohort filters: country, channel, device_type, utm_source, utm_campaign.
Prompts
Pre-built analytics workflows the AI can follow. Each prompt tells the model which tools to call, in what order, and how to present the results.
| Prompt | What it produces |
|---|---|
weekly_report | Traffic summary, top pages, referrers, countries, one actionable insight. |
traffic_diagnosis | Root-cause analysis for traffic changes. Drills into channels, countries, devices, pages. |
conversion_audit | Funnel drop-off analysis with cohort segmentation. |
channel_breakdown | Traffic quality comparison across sources (volume vs engagement). |
page_performance | Deep-dive on a single page: trends, referrers, devices, engagement vs site average. |
Examples
Traffic overview
"How's my traffic this week?"
Calls get_overview with period="7d". Returns pageviews, visitors, sessions, bounce rate, avg duration, and comparison to the previous week.
"What's the bounce rate on /pricing?"
Calls get_overview with pathname="/pricing". Returns metrics for that single page, including comparison to the prior period.
"Show me traffic trends for the last 90 days"
Calls get_timeseries with period="90d". Returns daily counts over the full window.
Sources and geography
"Where are my visitors coming from?"
Calls get_top_referrers. Returns referrer hostnames with channel classification.
"Show me only organic search traffic to /blog"
Calls get_top_referrers with channel="organic_search" and pathname="/blog".
"Which countries send the most traffic?"
Calls get_countries. Returns ISO country codes with visitor and pageview counts.
Devices
"What's the mobile vs desktop split?"
Calls get_devices with group_by="device_type".
"Which browsers are my visitors using?"
Calls get_devices with group_by="browser".
Custom events
"How many signups happened this month?"
Calls get_events with name="signup" and period="30d".
"Break down signups by plan"
Calls get_events with name="signup" and group_by="plan". Returns counts per property value (e.g. free: 42, pro: 18, growth: 7).
"How many signups came from the spring campaign?"
Calls get_events with name="signup" and utm_campaign="spring-launch".
Funnels Pro
"Create a funnel from pricing page to signup to checkout"
Calls create_funnel with steps ["pageview:/pricing", "signup", "checkout_completed"]. Returns step-by-step conversion rates immediately.
"How does the pricing-to-signup funnel convert on mobile?"
Calls get_funnel with name="pricing-to-signup" and device_type="mobile".
"Compare funnel conversion for US vs Germany"
Two calls to get_funnel, one with country="US" and one with country="DE".
"What's the funnel conversion for visitors from Google?"
Calls get_funnel with channel="organic_search".
"How many mobile users from Germany signed up?"
Create a single-step funnel with steps=["signup"] and filter with country="DE" and device_type="mobile". Cohort filtering works on any funnel, including single-event funnels.
See Funnels for step syntax, cohort filters, and response shape.
Alerts Pro
"Alert me if visitors drop by 30% week over week"
Calls create_alert with metric="visitors", condition="drops_by", threshold=30, period="7d".
"Alert me if bounce rate on /pricing goes above 70%"
Calls create_alert with metric="bounce_rate", condition="above", threshold=70, pathname="/pricing".
Workflows
"Run a weekly report"
Follows the weekly_report prompt. Calls overview, top pages, referrers, countries, and timeseries, then synthesizes a structured report with trends and one actionable insight.
"Why did traffic drop last week?"
Follows the traffic_diagnosis prompt. Systematically checks channels, countries, devices, and pages to isolate the root cause.
"Audit our conversion funnel"
Follows the conversion_audit prompt. Analyzes funnel drop-offs, segments by device/country/channel, and identifies the biggest optimization opportunity.
"How is /pricing performing?"
Follows the page_performance prompt. Pulls page-specific metrics, referrers, device split, and compares engagement to the site average.