MCP App Store
by KeenEthics
Seal Copilot icon

Seal Copilot MCP Server for ChatGPT

By Sealmetrics SL

Overview

Seal Copilot turns your Sealmetrics data into decisions: a verdict first, the evidence with numbers, one action, and a date to check it. It diagnoses drops down to the campaign that caused them, finds revenue left on the table, audits catalog friction per product, watches the cart during the day against a learned baseline, and reviews your tracking setup. Fourteen procedures over one methodology, with playbooks for ecommerce, hotels and SaaS that it picks from your own tracking. Every figure comes from a tool result in the same session: it never invents a number, and it refuses to conclude when the sample is too small. Requires a Sealmetrics account.

How to connect

  1. ChatGPT: Open the Seal Copilot listing in the ChatGPT app directory and choose Connect. ChatGPT listing Seal Copilot exposes 44 tools here.

Tools

fetch

ChatGPT
Get a SealMetrics analytics report by its id. Id format: '<report>:<period>', e.g. 'overview:30d'.

get_browsers

ChatGPT
Get traffic broken down by browser: Chrome, Safari, Firefox, Edge, etc. Shows entrances and pageviews per browser with pagination.

get_campaigns

ChatGPT
Get traffic and conversions broken down by campaign (utm_campaign). Shows performance of individual marketing campaigns including entrances, conversions, and revenue.

get_channels

ChatGPT
Get traffic metrics grouped by channel: Paid Search, Organic Search, Social, Direct, Email, Referral, etc. Channels are automatically classified based on UTM parameters and referrer rules.

get_content_groups

ChatGPT
Get metrics grouped by content group (content_grouping). Shows pageviews and entrances per content group. Useful for understanding which sections of a site get the most traffic.

get_conversion_items_raw

ChatGPT
Returns one row per item inside a conversion (e.g. one row per product in a purchase) from /stats/conversion-items/raw. properties are always included; they carry product_id, sku, price and quantity. Accepts ranges of up to 31 days (e.g. today, yesterday, 7d, 30d, last_week, last_month, or start_date/end_date); longer ones are rejected. limit defaults to 10 and is capped at 100 rows; page returns further rows.

get_conversions

ChatGPT
Get conversions broken down by type (e.g. purchase, signup). Shows count, revenue, and average order value per conversion type. Aggregated per type; no per-event or per-product rows.

get_conversions_raw

ChatGPT
Returns raw conversion rows from /stats/conversions/raw (one row per event, with timestamp_utc and timestamp_local). Item-level data (sku, price, quantity per product) is not split into rows here. Custom properties are excluded unless include_properties=true. Accepts ranges of up to 31 days (e.g. today, yesterday, 7d, 30d, last_week, last_month, or start_date/end_date); longer ones are rejected. limit defaults to 10 and is capped at 100 rows; page returns further rows.

get_countries

ChatGPT
Get traffic broken down by country. Shows entrances, pageviews, conversions, and revenue per country. Country codes follow ISO 3166-1 alpha-2 (e.g. ES=Spain, US=United States); 'Unknown' is a real bucket for browsers whose timezone maps to no country. Country is derived from the browser's timezone, not the IP address, so treat splits as directional. The codes returned here are the values other tools accept in their country filter.

get_device_types

ChatGPT
Get traffic broken down by device type (desktop, mobile, tablet) with pagination. Shows entrances, pageviews, conversions per device type.

get_devices

ChatGPT
Get traffic broken down by device type (desktop/mobile/tablet), browser (Chrome, Safari, Firefox...), and operating system (Windows, macOS, iOS, Android...). Returns all three breakdowns in a single call.

get_doc

ChatGPT
Read one page of the official SealMetrics documentation as plain text, fetched live from docs.sealmetrics.com. Takes a documentation page path (e.g. 'getting-started/quick-start') or a full docs.sealmetrics.com URL.

get_funnel

ChatGPT
Get the funnel table broken down by UTM source/medium/campaign/term: per row, entrances, page_views, and counts of each microconversion type and conversion type plus revenue per conversion type, with period totals and the list of types seen. The counts are independent aggregates, not a sequential funnel: a row can show more conversions than add_to_cart events, and ratios between columns are not per-session dropoff. Ratios computed from it are aggregate ratios. Rows are the top limit UTM combinations by entrances (default 100, max 500); truncated: true means more exist, so a small source may be missing: the utm_source / utm_medium / utm_campaign filters (exact match, case-insensitive) narrow the rows. totals always cover the whole filtered period.

get_landing_pages

ChatGPT
Get landing page performance: entrances, bounce rate, conversions. Landing pages are the first page users see when entering the site. High bounce rates indicate poor landing page experience. Supports multi-value filters (country, device_type, browser, os, channel_group, UTMs) as arrays of strings — e.g. country=['ES','PT'] to filter to two countries. Pass include=['channel_group'] to also break down by channel group. The two are orthogonal — combine them as needed.

get_landing_pages_by_content_group

ChatGPT
Get landing page metrics grouped by content grouping. Shows aggregate performance per content group (e.g. blog, product, category).

get_marketing_playbook

ChatGPT
Returns the SealMetrics marketing analysis method as text: the steps, checks and benchmarks for a weekly, monthly or quarterly performance report and for diagnosing a channel or campaign. It contains no site data. Diagnose marketing performance for a SealMetrics site and explain why the numbers move. Generalist marketing analysis (acquisition, channels, landing pages, conversions, traffic quality) with strong bias toward causal explanation and concrete next actions for Founders, CEOs and CMOs.

get_microconversion_details

ChatGPT
Get detailed breakdown for a specific microconversion type. Shows metrics segmented by source, medium, campaign, country, device, browser, and OS.

get_microconversions

ChatGPT
Get microconversions (smaller engagement events) broken down by type: add_to_cart, newsletter_signup, pdf_download, etc. Shows count and trends per event type; aggregated, no per-event rows.

get_microconversions_raw

ChatGPT
Returns raw microconversion rows from /stats/microconversions/raw (one row per event, with timestamp_utc and timestamp_local). Use for one-row-per-event detail of microconversions (add_to_cart, newsletter_signup, etc.). Custom properties are excluded by default — pass include_properties=true to receive them. Accepts ranges of up to 31 days (e.g. today, yesterday, 7d, 30d, last_week, last_month, or start_date/end_date); longer ones are rejected. limit defaults to 10 and is capped at 100 rows; page returns further rows.

get_operating_systems

ChatGPT
Get traffic broken down by operating system: Windows, macOS, iOS, Android, Linux, etc. Shows entrances and pageviews per OS with pagination.

get_overview

ChatGPT
Get dashboard KPIs: pageviews, entrances, bounce rate, conversions, revenue. Includes time series data and optional period-over-period comparison. This is the best starting point for understanding a site's performance.

get_pages

ChatGPT
Get metrics per page URL path: pageviews and entrances. Useful for finding most popular pages and content performance. No bounce rate: bounce is only meaningful per entry page, not per page viewed. Supports multi-value filters (country, device_type, browser, os, channel_group, UTMs) as arrays of strings — e.g. device_type=['mobile'] for mobile only. include=['device'] adds a per-device breakdown; filters and include combine.

get_property_breakdown

ChatGPT
Get a complete property breakdown with pivot-table style data. Shows all values for a property key with their counts and revenue — ideal for analyzing product or category performance.

get_property_values

ChatGPT
Get property values with counts, grouped by a UTM parameter. Paginated. Use to see which values a specific property key has and how they distribute across traffic sources.

get_site

ChatGPT
Get detailed information about one site: name, domains, timezone, configuration and tracking status.

get_terms

ChatGPT
Get traffic and conversions broken down by UTM term (keyword). Shows which search keywords or ad terms drive the most traffic. Supports filtering by source, medium, and campaign.

get_top_campaigns

ChatGPT
Get the top campaigns (utm_campaign) ranked by entrances, as a compact list of the top N. No period comparison, sorting or pagination.

get_top_channels

ChatGPT
Get the top channels (Paid Search, Organic, Social, etc.) ranked by entrances, as a compact list of the top N. No pagination.

get_top_landing_pages

ChatGPT
Get the top landing pages ranked by entrances, as a compact list of the top N. Filterable by content_grouping. No period comparison, sorting, pagination or device/browser/channel/path filters.

get_top_pages

ChatGPT
Get the top pages ranked by page views, as a compact list of the top N. No period comparison, sorting, pagination or device/browser/channel/path filters.

get_top_referrers

ChatGPT
Get top referrer domains ranked by entrances. Shows which external sites send the most traffic.

get_top_sources

ChatGPT
Get the top traffic sources (utm_source) ranked by entrances, as a compact list of the top N. No period comparison, sorting or pagination. With landing_page, only the sessions that entered on that page are counted.

get_top_terms

ChatGPT
Get the top UTM terms (keywords) ranked by entrances, as a compact list of the top N. No period comparison, sorting or pagination.

get_tracking_code

ChatGPT
Get the tracking pixel code for a site and the full JavaScript API reference for implementing pageviews, conversions, microconversions, and content grouping. Returns the site-specific <script> tag plus implementation guide with examples for e-commerce, SaaS, blog, and React/Next.js.

get_traffic_mediums

ChatGPT
Get traffic broken down by medium (utm_medium): organic, cpc, email, referral, social, etc. Useful for understanding which marketing channels drive traffic.

get_traffic_sources

ChatGPT
Get traffic broken down by source (utm_source): google, facebook, twitter, direct, etc. Shows entrances, pageviews, conversions, and revenue per source.

get_troubleshooting_guide

ChatGPT
Returns the SealMetrics troubleshooting guide as text; it contains no site data. Validated symptom→cause→fix answers for the most common SealMetrics setup and data questions - tags firing before the tracker loads, lost conversions/microconversions, duplicate pageviews in SPAs or Tag Manager setups, channel rules that never match (traffic stuck in Unassigned), payment-gateway and login/SSO domains showing up as traffic sources, numbers that do not match another analytics tool, filters that seem to be ignored (country names instead of ISO codes, landing-page filters on tools that do not take one), and live browser checks of whether the tag actually loads on a page.

list_channel_rules

ChatGPT
List channel group rules configured for a site. Shows how traffic is classified into channels (Paid Search, Organic, Social, etc.) based on UTM parameters and referrer patterns.

list_microconversion_types

ChatGPT
List the microconversion type names that have events on a site in the period.

list_property_keys

ChatGPT
List the custom property keys present on conversions and/or microconversions in the period.

list_sites

ChatGPT
List the sites (web properties) this connection can read. Returns each site's ID, name and domains; the ID is the site_id the report tools take.

search_docs

ChatGPT
Search the official SealMetrics documentation (docs.sealmetrics.com), fetched live. Covers tracker installation (WordPress, WooCommerce, Shopify, Magento, PrestaShop, Google Tag Manager, React/Next.js/Vue SPAs), conversion and microconversion tracking, UTM/campaign tagging, report and metric definitions, API usage, account settings, and privacy/GDPR/consentless questions. Returns matching pages with title, path and description.

test_channel_rules

ChatGPT
Classify a source/medium/campaign combination exactly like the tracking pixel would (same RE2 semantics, lowercase matching, custom rules before defaults). Returns the resulting channel and the rule that matched. With include_inactive=true, draft (not live) custom rules are evaluated too, so a draft can be checked before it is published in the dashboard.

Example Prompts

Click any prompt to copy it.

Similar Apps

Listed on MCP App Store

Listed on MCP App Store

Add this badge to your site or README. It links to this page.

HTML

<a href="https://mcp-app-store.keenethics-labs.com/app/seal-copilot"><img src="https://mcp-app-store.keenethics-labs.com/badges/listed-on-mcp-app-store.svg" alt="Listed on MCP App Store" width="220" height="44"></a>

Markdown

[![Listed on MCP App Store](https://mcp-app-store.keenethics-labs.com/badges/listed-on-mcp-app-store.svg)](https://mcp-app-store.keenethics-labs.com/app/seal-copilot)