Summary
A cookieless analytics platform with an MCP server that installs tracking snippets directly into your codebase. The 28 tools handle framework-aware setup for Next.js, React, Vue, Django, Rails, WordPress, and about a dozen other stacks, plus OAuth onboarding so you can read dashboard data from chat. You get drop-in script snippets, event tracking without cookies or GDPR banners, and a fleet view for managing multiple sites. Useful if you're spinning up projects quickly and want GA-style insights without the compliance overhead or manual integration work. Free tier covers 10k events per month with unlimited sites.
Keep your Mac awake while Claude Code and 40+ AI agents run. Sleeps when they're idle.
One time payment $9 →
Integrate web data into your AI product. One API to scrape website & brand data.
Get API Key Now →
Agent, run crypto. Access onchain data & trade routes via 1inch.
Install now →
On Capafy, your Skill runs online 24/7 as an agent product, and you get paid every time someone uses it.
Start earning →
Tools
Public tool metadata for what this MCP can expose to an agent.
28 tools
summarize_all_sitesANSWER FIRST for any 'how is everything doing' / 'what should I look at' / 'give me a status' question across the user's whole workspace. Returns fleet totals (visitors, events, % vs prior period), status breakdown (live/active/quiet/no_events), a prioritized 'needs attention'...1 params
ANSWER FIRST for any 'how is everything doing' / 'what should I look at' / 'give me a status' question across the user's whole workspace. Returns fleet totals (visitors, events, % vs prior period), status breakdown (live/active/quiet/no_events), a prioritized 'needs attention'...
Parameters* required
date_rangestringTime window. One of: today, yesterday, last_7_days, last_14_days, last_30_days, last_90_days. Defaults to last_7_days.one of
today · yesterday · last_7_days · last_14_days · last_30_days · last_90_daysrank_sites_by_growthRank sites by signed % visitor change vs the prior period. Use for 'which sites are growing/declining', 'biggest movers', 'what's trending'. direction='up' returns growers only, 'down' returns decliners only, 'both' (default) returns the biggest movers regardless of direction....3 params
Rank sites by signed % visitor change vs the prior period. Use for 'which sites are growing/declining', 'biggest movers', 'what's trending'. direction='up' returns growers only, 'down' returns decliners only, 'both' (default) returns the biggest movers regardless of direction....
Parameters* required
limitintegerMax rows. Defaults to 10, capped at 50.
directionstringFilter to growers ('up'), decliners ('down'), or biggest absolute movers either way ('both', default).one of
up · down · bothdate_rangestringTime window. One of: today, yesterday, last_7_days, last_14_days, last_30_days, last_90_days. Defaults to last_7_days.one of
today · yesterday · last_7_days · last_14_days · last_30_days · last_90_daysidentify_dead_sitesFind sites that look abandoned: classified as 'never_installed' (site exists but has never fired an event — most likely the tracker isn't installed) or 'went_silent' (used to get traffic, last seen N+ days ago). Default quiet_days=7. Use for 'which sites are dead', 'what shoul...1 params
Find sites that look abandoned: classified as 'never_installed' (site exists but has never fired an event — most likely the tracker isn't installed) or 'went_silent' (used to get traffic, last seen N+ days ago). Default quiet_days=7. Use for 'which sites are dead', 'what shoul...
Parameters* required
quiet_daysintegerHow many days of silence count as 'dead'. Defaults to 7. Anything fresher than this is treated as alive.
detect_anomaliesFind unusually high or low traffic days across sites using per-site z-score. For each site, computes the mean and stddev of daily visitors over the range, then flags any day where |z| ≥ threshold (default 2.5, ~1-in-100). Use for 'anything weird?', 'spikes this week', 'show me...3 params
Find unusually high or low traffic days across sites using per-site z-score. For each site, computes the mean and stddev of daily visitors over the range, then flags any day where |z| ≥ threshold (default 2.5, ~1-in-100). Use for 'anything weird?', 'spikes this week', 'show me...
Parameters* required
site_idstringInternal site UUID. Get one from list_sites. Omit to scope to the entire workspace.
thresholdnumberZ-score absolute value above which to flag a day as anomalous. Defaults to 2.5 (~1-in-100). Lower = more sensitive.
date_rangestringTime window. One of: today, yesterday, last_7_days, last_14_days, last_30_days, last_90_days. Defaults to last_14_days for stable mean/stddev.one of
today · yesterday · last_7_days · last_14_days · last_30_days · last_90_daysexplain_traffic_changeANSWER FIRST for any 'why did traffic [spike/drop/change]' question on a specific site. Pulls overview + deltas, the top 5 page gainers and losers vs prior period, the top 5 referrer gainers and losers, anomaly-day callouts (z-score ≥ 2), and the daily series so you can name t...2 params
ANSWER FIRST for any 'why did traffic [spike/drop/change]' question on a specific site. Pulls overview + deltas, the top 5 page gainers and losers vs prior period, the top 5 referrer gainers and losers, anomaly-day callouts (z-score ≥ 2), and the daily series so you can name t...
Parameters* required
site_idstringInternal site UUID for the target site. Get one from list_sites.
date_rangestringTime window. One of: today, yesterday, last_7_days, last_14_days, last_30_days, last_90_days. Defaults to last_7_days.one of
today · yesterday · last_7_days · last_14_days · last_30_days · last_90_dayssetup_analytics_for_siteANSWER FIRST for any 'add analytics' / 'install analytics' / 'set up analytics' / 'add tracking to my app' request — this is the one-call install workflow that bundles site creation + framework-specific snippet + paste-target file path + post-install next steps. Strongly prefe...3 params
ANSWER FIRST for any 'add analytics' / 'install analytics' / 'set up analytics' / 'add tracking to my app' request — this is the one-call install workflow that bundles site creation + framework-specific snippet + paste-target file path + post-install next steps. Strongly prefe...
Parameters* required
domainstringThe domain you want to track (e.g. 'acme.com'). Apex domain preferred — Gizmo collects events for the whole hostname.
frameworkstringWeb framework of the user's project. Detect from package.json + config files: next.config.* → next, nuxt.config.* → nuxt, svelte.config.* → sveltekit, astro.config.* → astro, remix.config.* or @remix-run/* → remix, vite.config.* → vite, otherwise react or html.one of
next · react · vite · remix · astro · nuxtevent_hintstringOptional. If the user mentioned tracking a specific custom event (e.g. 'track signups'), pass the event name here ('signup') and the response will include a ready-to-drop example_event snippet.
list_sitesANSWER FIRST for 'what sites do I have in Gizmo?', 'list my sites', 'show me my projects', 'which domains am I tracking?'. Returns every site in the workspace with id, domain, createdAt, and current ignored status. Optional tag filter narrows to one group. Use the returned sit...1 params
ANSWER FIRST for 'what sites do I have in Gizmo?', 'list my sites', 'show me my projects', 'which domains am I tracking?'. Returns every site in the workspace with id, domain, createdAt, and current ignored status. Optional tag filter narrows to one group. Use the returned sit...
Parameters* required
tagstringOptional tag name to filter by. Omit to list every site in the workspace.
create_siteRegister a domain in Gizmo and return the install snippet. For first-time install asks ('add analytics', 'install Gizmo', 'set up tracking'), prefer setup_analytics_for_site — it bundles this call + the paste target + next-step suggestions in one response. Use create_site dire...2 params
Register a domain in Gizmo and return the install snippet. For first-time install asks ('add analytics', 'install Gizmo', 'set up tracking'), prefer setup_analytics_for_site — it bundles this call + the paste target + next-step suggestions in one response. Use create_site dire...
Parameters* required
domainstringThe domain you want to track (e.g. 'acme.com'). Apex domain preferred — Gizmo collects events for the whole hostname.
frameworkstringWeb framework of the user's project. Detect from package.json + config files: next.config.* → next, nuxt.config.* → nuxt, svelte.config.* → sveltekit, astro.config.* → astro, remix.config.* or @remix-run/* → remix, vite.config.* → vite, otherwise react or html.one of
next · react · vite · remix · astro · nuxtget_install_snippetRe-fetch the framework-specific install snippet for an existing site. Use this when the snippet was printed earlier but never installed, when the user switched frameworks, or when they ask 'where's the gizmo install snippet for foo.com?', 'how do I install Gizmo on this projec...2 params
Re-fetch the framework-specific install snippet for an existing site. Use this when the snippet was printed earlier but never installed, when the user switched frameworks, or when they ask 'where's the gizmo install snippet for foo.com?', 'how do I install Gizmo on this projec...
Parameters* required
site_idstringInternal site UUID for the target site. Get one from list_sites.
frameworkstringWeb framework of the user's project. Detect from package.json + config files: next.config.* → next, nuxt.config.* → nuxt, svelte.config.* → sveltekit, astro.config.* → astro, remix.config.* or @remix-run/* → remix, vite.config.* → vite, otherwise react or html.one of
next · react · vite · remix · astro · nuxtverify_installANSWER FIRST for any 'is it working?', 'did the install work?', 'is Gizmo tracking my site?', 'verify analytics setup', 'check if it's installed' question — typically asked right after create_site / get_install_snippet. Fetches the site's homepage and scans for the tracker scr...1 params
ANSWER FIRST for any 'is it working?', 'did the install work?', 'is Gizmo tracking my site?', 'verify analytics setup', 'check if it's installed' question — typically asked right after create_site / get_install_snippet. Fetches the site's homepage and scans for the tracker scr...
Parameters* required
site_idstringInternal site UUID for the target site. Get one from list_sites.
get_badge_snippetReturn the official "Powered by Gizmo" footer badge as a single anchor tag. Call this when the user asks to add a badge, link back, or thank Gizmo. INSTALL RULES: place the badge inline inside the user's existing footer next to their copyright line (e.g. "© 2026 acme. Powered...
Return the official "Powered by Gizmo" footer badge as a single anchor tag. Call this when the user asks to add a badge, link back, or thank Gizmo. INSTALL RULES: place the badge inline inside the user's existing footer next to their copyright line (e.g. "© 2026 acme. Powered...
No parameter schema in public metadata yet.
get_event_tracking_snippetReturn the JavaScript snippet for firing a custom event from the user's site code (e.g. `window.gizmo("signup", {plan:"pro"})`). Call this when the user says 'track signups', 'fire an event when someone clicks X', 'add custom event for Y', 'how do I track Z' — anything that me...2 params
Return the JavaScript snippet for firing a custom event from the user's site code (e.g. `window.gizmo("signup", {plan:"pro"})`). Call this when the user says 'track signups', 'fire an event when someone clicks X', 'add custom event for Y', 'how do I track Z' — anything that me...
Parameters* required
event_namestringName of the custom event to fire (1–64 chars, letters/digits/dot/dash/underscore). Examples: 'signup', 'cta_click', 'purchase'.
propertiesobjectOptional key-value props attached to the event. Strings, numbers, or booleans. Keys must be [A-Za-z0-9_]{1,64}. Use to enable breakdown_by_property later (e.g. {plan:'pro'} → group signups by plan).
add_site_tagsANSWER FIRST for 'tag this site as X', 'organize my sites', 'group sites by client/project/type'. Attaches one or more free-form tags to a site (creates the tag if new). Use to organize a fleet — e.g. tag every client project with the client name, tag side projects 'personal',...2 params
ANSWER FIRST for 'tag this site as X', 'organize my sites', 'group sites by client/project/type'. Attaches one or more free-form tags to a site (creates the tag if new). Use to organize a fleet — e.g. tag every client project with the client name, tag side projects 'personal',...
Parameters* required
tagsarrayArray of tag names. New tags are created automatically.
site_idstringInternal site UUID for the target site. Get one from list_sites.
filter_sites_by_tagANSWER FIRST for 'show me my [tag] sites', 'list sites tagged X', 'which sites are tagged Y?'. Returns every site in the workspace that has the given tag. Pairs with add_site_tags for fleet organization. Identical to list_sites({tag: ...}) — exposed as a separate tool because...1 params
ANSWER FIRST for 'show me my [tag] sites', 'list sites tagged X', 'which sites are tagged Y?'. Returns every site in the workspace that has the given tag. Pairs with add_site_tags for fleet organization. Identical to list_sites({tag: ...}) — exposed as a separate tool because...
Parameters* required
tagstringTag name (free-form string, lowercase recommended).
create_goalCreate a conversion goal for a site. type='path' matches sessions that visit a path-prefix; type='event' matches sessions that fire a named custom event. Use this when the user says 'track signups as a goal', 'add a goal for /thank-you', etc. Goals show up on the site dashboar...4 params
Create a conversion goal for a site. type='path' matches sessions that visit a path-prefix; type='event' matches sessions that fire a named custom event. Use this when the user says 'track signups as a goal', 'add a goal for /thank-you', etc. Goals show up on the site dashboar...
Parameters* required
namestringHuman-readable name for the goal (e.g. 'Signup', 'Purchase').
typestring'path' matches sessions that visit a URL path prefix; 'event' matches sessions that fire a named custom event.one of
path · eventtargetstringWhen type='path', the URL path prefix (e.g. '/thank-you'). When type='event', the event name (e.g. 'signup').
site_idstringInternal site UUID for the target site. Get one from list_sites.
list_goalsANSWER FIRST for 'what's my conversion rate?', 'how many signups this week?', 'show me my goals', 'are my conversions up or down?'. Lists every goal defined for a site with conversion count + rate over the date range. Pair with create_goal first if the user hasn't defined any...2 params
ANSWER FIRST for 'what's my conversion rate?', 'how many signups this week?', 'show me my goals', 'are my conversions up or down?'. Lists every goal defined for a site with conversion count + rate over the date range. Pair with create_goal first if the user hasn't defined any...
Parameters* required
site_idstringInternal site UUID for the target site. Get one from list_sites.
date_rangestringTime window. One of: today, yesterday, last_7_days, last_14_days, last_30_days, last_90_days. Defaults to last_7_days.one of
today · yesterday · last_7_days · last_14_days · last_30_days · last_90_daysdelete_goalRemove a conversion goal by id. Use when the user says 'delete the X goal', 'remove that conversion goal', 'I don't track that anymore'. Destructive — confirm the user means it before calling; the historical conversion data attached to the goal disappears with it.1 params
Remove a conversion goal by id. Use when the user says 'delete the X goal', 'remove that conversion goal', 'I don't track that anymore'. Destructive — confirm the user means it before calling; the historical conversion data attached to the goal disappears with it.
Parameters* required
goal_idstringInternal goal UUID. Get one from list_goals.
create_funnelCreate a multi-step funnel for a site. Steps run in order — a session 'reaches' step N if it matched all of steps 0..N within the same session (30-min idle gap). Use this when the user says 'track the pricing → signup funnel' or describes a sequence. type='path' matches a path...3 params
Create a multi-step funnel for a site. Steps run in order — a session 'reaches' step N if it matched all of steps 0..N within the same session (30-min idle gap). Use this when the user says 'track the pricing → signup funnel' or describes a sequence. type='path' matches a path...
Parameters* required
namestringHuman-readable name for the funnel (e.g. 'Pricing to signup').
stepsarrayOrdered steps. A session 'reaches' step N if it matched all of steps 0..N within the same session (30-min idle gap). Min 2, max 10.
site_idstringInternal site UUID for the target site. Get one from list_sites.
list_funnelsANSWER FIRST for 'where are users dropping off?', 'show me the funnel', 'how's my signup flow doing?', 'what's my conversion funnel?'. Lists every funnel defined for a site with per-step reached counts + drop-off rate between steps. Surface the biggest drop as the headline — t...2 params
ANSWER FIRST for 'where are users dropping off?', 'show me the funnel', 'how's my signup flow doing?', 'what's my conversion funnel?'. Lists every funnel defined for a site with per-step reached counts + drop-off rate between steps. Surface the biggest drop as the headline — t...
Parameters* required
site_idstringInternal site UUID for the target site. Get one from list_sites.
date_rangestringTime window. One of: today, yesterday, last_7_days, last_14_days, last_30_days, last_90_days. Defaults to last_7_days.one of
today · yesterday · last_7_days · last_14_days · last_30_days · last_90_daysdelete_funnelRemove a funnel by id. Use when the user says 'delete the X funnel', 'remove that flow', 'I'm not tracking that anymore'. Destructive — confirm before calling.1 params
Remove a funnel by id. Use when the user says 'delete the X funnel', 'remove that flow', 'I'm not tracking that anymore'. Destructive — confirm before calling.
Parameters* required
funnel_idstringInternal funnel UUID. Get one from list_funnels.
get_workspace_overviewAggregate metrics across all sites in the workspace (visitors, pageviews, sessions, bounce rate, top sites). Supports filters and custom date ranges. Optional dimension filters. Each filter is {dim, op, value}. Available dims: page, entry_page, exit_page, referrer, hostname, c...2 params
Aggregate metrics across all sites in the workspace (visitors, pageviews, sessions, bounce rate, top sites). Supports filters and custom date ranges. Optional dimension filters. Each filter is {dim, op, value}. Available dims: page, entry_page, exit_page, referrer, hostname, c...
Parameters* required
filtersarrayAND-joined dimension filters. Each is {dim, op, value}. See MCPFilter schema for dim/op options.
date_rangevalueOptional date range. Either {preset:'last_7_days'} OR {from:'2026-05-01', to:'2026-05-15'}. Defaults per tool — usually last_7_days.
get_site_trafficTraffic metrics for a single site (overview + top pages + top referrers). Supports filters and custom date ranges. Optional dimension filters. Each filter is {dim, op, value}. Available dims: page, entry_page, exit_page, referrer, hostname, channel, ai_source, utm_source, utm_...3 params
Traffic metrics for a single site (overview + top pages + top referrers). Supports filters and custom date ranges. Optional dimension filters. Each filter is {dim, op, value}. Available dims: page, entry_page, exit_page, referrer, hostname, channel, ai_source, utm_source, utm_...
Parameters* required
filtersarrayAND-joined dimension filters.
site_idstringInternal site UUID for the target site. Get one from list_sites.
date_rangevalueOptional date range. Either {preset:'last_7_days'} OR {from:'2026-05-01', to:'2026-05-15'}. Defaults per tool — usually last_7_days.
get_top_pagesTop page paths by event count. Supports filters (e.g. country=US) so you can answer "top pages from Germany." Pass site_id to scope to one site. Optional dimension filters. Each filter is {dim, op, value}. Available dims: page, entry_page, exit_page, referrer, hostname, channe...4 params
Top page paths by event count. Supports filters (e.g. country=US) so you can answer "top pages from Germany." Pass site_id to scope to one site. Optional dimension filters. Each filter is {dim, op, value}. Available dims: page, entry_page, exit_page, referrer, hostname, channe...
Parameters* required
limitintegerMax number of rows to return. Defaults to 20, capped at 200. Pagination via date_range for larger windows.
filtersarrayAND-joined dimension filters.
site_idstringInternal site UUID. Get one from list_sites. Omit to scope to the entire workspace.
date_rangevalueOptional date range. Either {preset:'last_7_days'} OR {from:'2026-05-01', to:'2026-05-15'}. Defaults per tool — usually last_7_days.
get_top_referrersTop referrer hostnames by event count. Supports filters (e.g. device=mobile) so you can answer "top referrers from mobile users." Pass site_id to scope to one site. Optional dimension filters. Each filter is {dim, op, value}. Available dims: page, entry_page, exit_page, referr...4 params
Top referrer hostnames by event count. Supports filters (e.g. device=mobile) so you can answer "top referrers from mobile users." Pass site_id to scope to one site. Optional dimension filters. Each filter is {dim, op, value}. Available dims: page, entry_page, exit_page, referr...
Parameters* required
limitintegerMax number of rows to return. Defaults to 20, capped at 200. Pagination via date_range for larger windows.
filtersarrayAND-joined dimension filters.
site_idstringInternal site UUID. Get one from list_sites. Omit to scope to the entire workspace.
date_rangevalueOptional date range. Either {preset:'last_7_days'} OR {from:'2026-05-01', to:'2026-05-15'}. Defaults per tool — usually last_7_days.
get_breakdownTop-N for any column-backed dimension: page, referrer, country, region, city, browser, device, language, screen, channel, ai_source, hostname. Use this for "top countries / browsers / devices / channels / AI sources / cities / etc." — anything in the dashboard's breakdown tabs...5 params
Top-N for any column-backed dimension: page, referrer, country, region, city, browser, device, language, screen, channel, ai_source, hostname. Use this for "top countries / browsers / devices / channels / AI sources / cities / etc." — anything in the dashboard's breakdown tabs...
Parameters* required
limitintegerMax number of rows to return. Defaults to 20, capped at 200. Pagination via date_range for larger windows.
filtersarrayAND-joined dimension filters.
site_idstringInternal site UUID. Get one from list_sites. Omit to scope to the entire workspace.
dimensionstringWhich built-in dimension to group by: page, referrer, country, region, city, browser, device, language, screen, channel, ai_source, or hostname.one of
page · referrer · country · region · city · browserdate_rangevalueOptional date range. Either {preset:'last_7_days'} OR {from:'2026-05-01', to:'2026-05-15'}. Defaults per tool — usually last_7_days.
query_eventsQuery raw events. Default returns the latest 100; max 1000 per call. For large analyses, paginate by date_range. Filters supported. Optional dimension filters. Each filter is {dim, op, value}. Available dims: page, entry_page, exit_page, referrer, hostname, channel, ai_source,...5 params
Query raw events. Default returns the latest 100; max 1000 per call. For large analyses, paginate by date_range. Filters supported. Optional dimension filters. Each filter is {dim, op, value}. Available dims: page, entry_page, exit_page, referrer, hostname, channel, ai_source,...
Parameters* required
limitintegerMax events to return. Defaults to 100, capped at 1000.
filtersarrayAND-joined dimension filters.
site_idstringInternal site UUID. Get one from list_sites. Omit to scope to the entire workspace.
date_rangevalueOptional date range. Either {preset:'last_7_days'} OR {from:'2026-05-01', to:'2026-05-15'}. Defaults per tool — usually last_7_days.
event_namestringCustom event name (1–64 chars, alphanumeric + dot/dash/underscore). Optional when querying all events.
get_time_seriesTime-bucketed visitors / pageviews / events for a date range. Returns one row per bucket. Bucket auto-resolves to hour for windows ≤ 48h, day otherwise — pass bucket to override. Optional site_id scopes to a single site. Filters supported. Optional dimension filters. Each filt...4 params
Time-bucketed visitors / pageviews / events for a date range. Returns one row per bucket. Bucket auto-resolves to hour for windows ≤ 48h, day otherwise — pass bucket to override. Optional site_id scopes to a single site. Filters supported. Optional dimension filters. Each filt...
Parameters* required
bucketstringBucket granularity. Auto-resolves to hour for windows ≤48h, day otherwise. Override to force a specific bucket.one of
hour · dayfiltersarrayAND-joined dimension filters.
site_idstringInternal site UUID. Get one from list_sites. Omit to scope to the entire workspace.
date_rangevalueOptional date range. Either {preset:'last_7_days'} OR {from:'2026-05-01', to:'2026-05-15'}. Defaults per tool — usually last_7_days.
breakdown_by_propertyANSWER FIRST for 'how many [X] signups by plan?', 'split [event] by [property]', 'pro vs free signups', 'sales by SKU', 'newsletter signups by source' — anything that means grouping a CUSTOM event property the user previously instrumented with get_event_tracking_snippet. Prope...5 params
ANSWER FIRST for 'how many [X] signups by plan?', 'split [event] by [property]', 'pro vs free signups', 'sales by SKU', 'newsletter signups by source' — anything that means grouping a CUSTOM event property the user previously instrumented with get_event_tracking_snippet. Prope...
Parameters* required
limitintegerMax number of rows to return. Defaults to 20, capped at 200. Pagination via date_range for larger windows.
site_idstringInternal site UUID. Get one from list_sites. Omit to scope to the entire workspace.
propertystringCustom event property name. Must match [A-Za-z0-9_.-]{1,64}. The value users sent in their tracker call (e.g. property='plan' for window.gizmo('signup', {plan:'pro'})).
date_rangevalueOptional date range. Either {preset:'last_7_days'} OR {from:'2026-05-01', to:'2026-05-15'}. Defaults per tool — usually last_7_days.
event_namestringNarrow to a single event (e.g. 'signup'). Omit to roll up across every event that has the property.
Featured
Keep your Mac awake while Claude Code and 40+ AI agents run. Sleeps when they're idle.
One time payment $9 →Integrate web data into your AI product. One API to scrape website & brand data.
Get API Key Now →Agent, run crypto. Access onchain data & trade routes via 1inch.
Install now →On Capafy, your Skill runs online 24/7 as an agent product, and you get paid every time someone uses it.
Start earning →Categories
Registryactive
Packagegizmoanalytics-mcp
TransportHTTP
UpdatedMay 17, 2026
