About Seline
Seline MCP connects AI clients to the Seline analytics API for privacy-focused website and product analytics. It offers hosted OAuth access or a local stdio server authenticated with a project or account API token.
Key features
- 01Inspect project and visit metrics
- 02Query pages, events, and properties
- 03Analyze funnels and visitor activity
- 04Look up individual visitor insights
Use cases
- 01Review website traffic over time
- 02Explore custom event and property data
- 03Inspect visitor journeys and funnel results
MCP support
The server exposes read-only Seline analytics tools to MCP clients. It can connect through the hosted OAuth endpoint or run locally over stdio with an API token.
Configuration
Remote servers need no local install. Pick your client, then copy the config or install in one click.
https://api.seline.com/mcpclaude mcp add --transport http getseline-seline https://api.seline.com/mcp{"mcpServers":{"getseline-seline":{"type":"http","url":"https://api.seline.com/mcp"}}}Open in Cursor{"servers":{"getseline-seline":{"type":"http","url":"https://api.seline.com/mcp"}}}Open in VS Code{"mcpServers":{"getseline-seline":{"type":"http","url":"https://api.seline.com/mcp"}}}{"mcpServers":{"getseline-seline":{"serverUrl":"https://api.seline.com/mcp"}}}Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.
Tools
Details
From the project READMESeline MCP
Seline MCP talks to the Seline public API.
You can connect in two ways:
- Hosted —
https://api.seline.com/mcp(OAuth, for ChatGPT / Claude) - Local stdio —
npx @seline-analytics/mcpwith an API key (Cursor / Claude Code / Codex)
For the local server you can use either:
- A project API token — scoped to one project
- An account API token — access to all projects on the account
With an account token, use seline_list_projects to find a project, then pass projectId on analytics tools.
Cursor setup
{
"mcpServers": {
"seline": {
"command": "npx",
"args": ["-y", "@seline-analytics/mcp", "sln_your_api_key"]
}
}
}
Then restart Cursor so it reconnects MCP servers.
Claude Code setup
Run this command:
claude mcp add seline -- npx -y @seline-analytics/mcp sln_your_api_key
Then restart Claude Code.
Codex setup
If your Codex client uses mcpServers JSON config, add:
{
"mcpServers": {
"seline": {
"command": "npx",
"args": ["-y", "@seline-analytics/mcp", "sln_your_api_key"]
}
}
}
Then restart Codex and confirm the seline_* tools appear.
Tools
Projects:
seline_list_projects->POST /api/v1/projects
Data over time:
seline_get_data->POST /api/v1/dataseline_get_charges->POST /api/v1/charges
Counts and totals:
seline_get_top_visitors->POST /api/v1/statsseline_get_pages->POST /api/v1/pagesseline_get_visit_metrics->POST /api/v1/visit-metricsseline_get_custom_events->POST /api/v1/custom-events
Events and properties:
seline_get_events->POST /api/v1/eventsseline_get_property_values->POST /api/v1/property-valuesseline_get_field_keys->POST /api/v1/event-dataseline_get_field_values->POST /api/v1/field-values
Funnels:
seline_get_funnel->POST /api/v1/funnel
Visitors:
seline_get_visitors->POST /api/v1/visitorsseline_get_visitor->POST /api/v1/visitorseline_get_visitor_events->POST /api/v1/visitor-eventsseline_get_most_active_visitors->POST /api/v1/most-active-visitorsseline_get_visitor_insight->POST /api/v1/visitor-insight
Privacy Policy
This MCP server runs locally and sends requests only to the Seline API using your API key.
- Data collected: Your API key (stored locally in your MCP client) and analytics queries you run through the tools.
- Usage: The API key authenticates read-only requests to your Seline workspace.
- Storage: Seline stores analytics per Seline Privacy Policy.
- Third parties: Requests go to Seline infrastructure only.
- Retention: Governed by Seline account and privacy policy.
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.