What it does

Time-bucketed metric over a period. metric=pageviews includes AI-mediated human browsing (ai_user_action); metric=sessions and metric=visitors are browser-only, since server-side-ingested rows carry no session or visitor id — see traffic_class_breakdown for the AI-vs-direct split. metric=visitors is always 0 on privacy_mode=strict sites (no visitor IDs computed by design) — use sessions there. metric=events trends one named event (pass event_name) — the way to chart a conversion event over time, rather than calling event_details once per day; it counts rows like pageviews, so it includes AI-mediated human browsing.

Example prompts

Ask:

"Show me pageviews day-by-day for the last 30 days on mysite.com."
"Hourly visitor count for the last 24 hours."
"Weekly trend for sessions over the past quarter."
"Is the cta_click event trending up this month?"

A typical response:

Pageviews, daily, last 30 days for mysite.com:

  2026-04-20    1,847
  2026-04-21    1,920
  2026-04-22    2,114
  2026-04-23    8,412  ← spike (HN thread)
  ...
  2026-05-17    8,114
  2026-05-18    7,420
  2026-05-19    2,890

Returns an array of { ts, value } objects.

The metric argument controls what's measured: pageviews, visitors, sessions, or events. The last one trends a single named custom event over time — pass event_name alongside it (find names via list_events). The granularity argument controls bucket size: hour, day, or week. Long periods at hour-granularity return a lot of rows. Consider day or week for ranges over 30 days.

Counts include AI-mediated human browsing. See traffic_class_breakdown to see the AI-vs-direct split if it matters for your interpretation.

Pair with compare_periods for headline-style "is this up or down?" rather than the full day-by-day list.

Arguments

Name Type Required Description / Default
site_id string required Site identifier from list_sites (8-character base32, e.g. 'wjxayhdd').
metric string
enum: pageviews, visitors, sessions, events		 required Which metric to compute over the period. pageviews includes AI-mediated human browsing; sessions and visitors are browser-only. 'events' counts occurrences of the single event named in event_name (row count, like pageviews).
period string optional Time window. Keywords: today, yesterday, last_7_days, last_30_days, last_90_days, last_12_months. Or a custom date range YYYY-MM-DD..YYYY-MM-DD (inclusive).
default: last_7_days
granularity string
enum: hour, day, week		 optional Time bucket size for the series. Use 'hour' for last_24h analysis, 'day' for week/month views, 'week' for quarter+.
default: day
event_name string optional Required when metric=events, rejected with any other metric. Case-sensitive event name — use list_events to discover the names this site actually sends. Counts match the total_count event_details reports for the same event and period (volume metric, includes AI-mediated human browsing).

How to call it directly

If you're integrating from your own code rather than a chat client, this is the JSON-RPC payload:

curl -X POST https://mcp-analytics.com/mcp \
  -H "Authorization: Bearer YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "get_timeseries",
    "arguments": {
      "site_id": "abc12345",
      "metric": "pageviews"
    }
  }
}'

Token comes from /settings after you sign up. Replace any required arguments above.

Related tools

  • breakdown: Breakdown of visits by browser, os, device_type, or country (country empty in MVP).
  • list_sites: List all sites on the authenticated account.
  • top_timezones: Top IANA timezones (Europe/Berlin, America/New_York, ...) of visitors.
  • top_referrers: Top referring hosts.
  • engagement_overview: Real reading time + scroll depth from the engagement beacon (fired on pagehide).