supermetrics
Server Details
Supermetrics MCP: discover data sources & fields, run marketing-data queries, manage campaigns.
- Status
- Healthy
- Uptime
- 100.0% over the last 24 hours
- OAuth
- Works in Glama
- Last Tested
- Transport
- Streamable HTTP · MCP 2025-11-25
- URL
- ChatGPT
- Supermetrics Marketing Agent
- Claude
- Supermetrics Marketing Analytics
TDQS
Scored across 29 tools
Most tools have distinct purposes, but several boundaries are blurred: data_query and instagram_insights both query Meta/Instagram data, and campaign_and_resource_get is a mega-tool that absorbs unrelated functions such as creative_picker and contact_supermetrics. The descriptions help, but an agent must often inspect parameters/resource_type to choose correctly.
All names use snake_case, but the conventions are mixed: many are verb_noun get_/manage_/read_, while others are noun-first discovery/style tools or place the verb last (campaign_and_resource_get). The set is readable but not fully predictable.
29 tools is heavy for a single MCP surface, even though Supermetrics covers a genuinely broad domain. Several mega-tools already consolidate functionality, but the overall count still feels near the upper limit and could be tightened.
The surface covers the major workflows: discovery, querying, campaign reads/writes, activations, dashboards, audits/monitors, social publishing, business context, team management, support, and utilities. Minor lifecycle gaps remain, such as explicit delete/removal operations for dashboards, campaigns, activations, or audiences, but agents can generally work around them.
Publisher details
- Operator
- Supermetrics · Publisher source
- Operator website
- https://supermetrics.com/ · Publisher source
- Vendor relationship
- First-party
- Documentation
- https://mcp.supermetrics.com/ · Publisher source
- Privacy policy
- Not provided
- Trust center
- http://trust.supermetrics.com/ · Publisher source
- Restrictions
- Requires a Supermetrics account. Does not require account creation in advance, as first-time users receive a 14-day trial.
Available Tools
29 toolsaccounts_discoveryARead-onlyIdempotentInspect
Discover connected accounts for a data source.
Only call if data_source_discovery config shows has_account_list=true.
RETURNS: account_id, account_name, group_name (when present), tags (when present), and any saved business context for the account: shown as context when short enough to include, otherwise as context_tags naming what context exists — when a tag is relevant to the action you're taking, fetch that context with get_business_context (ds_id + account_id, filtered by tag).
Inputs: ds_id (required), filter, compress. Returns: data, error, notes, success, +1 more
activation_audience_createADestructiveInspect
Create a new, empty audience on an ad account (ds_id + account_id) to receive an audience activation, e.g. a Meta Custom Audience or a Google Customer Match list. Call activation_audience_list first and call this only when the user asks for a new audience by name: it creates a real object on their ad platform, and Meta allows duplicate names. Returns {id, name}; pass the id in the destination's config via manage_activation_query (FA: audience_id, AW: user_list_id).
Inputs: name (required), ds_id (required), account_id (required). Returns: data, error, notes, success, +1 more
activation_audience_listARead-onlyIdempotentInspect
List the existing audiences on a destination ad account (ds_id + account_id). Works before any audience activation exists. Pass the chosen audience id in a destination's config via manage_activation_query. Read-only; returns {items: [{id, name}], total_count}, unpaginated.
Inputs: ds_id (required), account_id (required). Returns: data, error, notes, success, +1 more
activation_query_listARead-onlyIdempotentInspect
List your audience activations, or read specific ones by passing their ids; never returns data rows. Use it to recover an activation_query_id, or to check an activation's ready_to_run and missing before running it. total_count counts every match, not just this page.
Inputs: ids, limit, state, offset. Returns: data, error, notes, success, +1 more
activation_query_runADestructiveInspect
Run an audience activation: hash its email and phone identifiers and upload them to every destination. Irreversible. For an activation whose ready_to_run is true. Returns state RUNNING at once; the run goes on in the background, so follow it with activation_query_status.
Inputs: activation_query_id (required). Returns: data, error, notes, success, +1 more
activation_query_statusARead-onlyIdempotentInspect
Get the status of an audience activation: DRAFT, RUNNING, COMPLETED, PARTIALLY_FAILED or FAILED. The cheap check to poll after a run; never returns data rows or config, but includes run-level and per-destination failure reasons when a run fails or partially fails.
Inputs: activation_query_id (required). Returns: data, error, notes, success, +1 more
campaign_and_resource_getARead-onlyInspect
Get campaigns or related resources.
Unless you already fetched it this session, call get_business_context (passing ds_id and account_id) to fetch applicable guidelines — e.g. which campaigns to focus on, how to evaluate performance, reporting preferences — and reuse that result for the rest of the session.
resource_type options (options go in params; each params key names the types it applies to):
campaigns (default): list, or one campaign with campaign_id. Ad groups/ads are INSIDE campaigns — use campaign_id + campaign_detail_level, not a separate resource_type.
health_check: 30-day snapshot of one campaign. Custom dates/breakdowns: use data_query instead
forecast: the platform's own delivery estimate for defined targeting across bid/daily-budget levels (AW/FA, account currency). For FA budget planning, campaign_planner is usually the better starting point.
assets: platform creative library (use id as asset_id) plus saved uploads/AI images (saved[], use upload_ref). Lists without opening creative_picker
pages (FA Pages), posts (FA promotable posts)
keyword_ideas/keyword_volumes (AW/AC)
targeting_search: search targeting catalog
reach_estimate: audience size
audiences, recommendations, history
experiments: campaign experiments / A/B tests with status, type and date range (AW)
negative_keyword_lists: shared negative keyword lists with keywords and campaigns using them (AW)
lead_forms: compact list, id for creative.lead_gen_form_id; form_id: one form (LIA, FA). Write: manage_campaign
conversion_types: conversion actions and any pixels (definitions, not per-campaign counts). data_query for metrics
approval_status: outcome of a change sent for approval
campaign_planner: optimistic/expected/pessimistic outcomes for a Facebook Ads budget, benchmarked on similar advertisers (FA only, EUR). Not the platform's own estimate - that is forecast.
Inputs: ds_id (required), account_id (required), params, max_rows, +1 more. Returns: data, error, notes, success, +1 more
contact_supermetricsAInspect
Send feedback, create a support ticket (only when the user authorizes it), submit a sales enquiry, or reply to an existing ticket. For a common problem, first call supermetrics_guide and try the relevant steps before opening a ticket. Only ticket if that does not resolve it or it needs staff action (non-self-serve plan changes or a genuine bug). Cancelling a trial or an online subscription is not a ticket: use the cancel_trial guide. If you ticket, note which guide was tried and what still failed. When you create a support ticket, always set "category" to the option that matches the user's issue (authentication, billing, subscription, data discrepancy, or query help). For product feature requests, you can also point users to the public wishlist at https://wishlist.supermetrics.com, where they can submit and vote on ideas. Before a ticket or sales enquiry about buying or pricing, call supermetrics_guide(mode="pricing_and_purchasing"); standard plans are bought online at https://supermetrics.com/pricing. Only submit a sales enquiry for special needs: many data sources, data source accounts, users, Supermetrics API access (custom connector or API key), or other custom requirements. Only the user's own tickets are accessible, and only post a reply when the user explicitly asks. Before creating a support ticket, first call get_support_tickets to list the user's existing tickets and check whether an open one already covers the same issue. If it does, add to that ticket with type "ticket_reply" instead of creating a duplicate; only create a new ticket when none of the open tickets match.
Inputs: type (required), ds_id, company, country, +7 more. Returns: data, error, notes, success, +1 more
creative_pickerADestructiveInspect
Open the creative picker to browse, upload, AI-generate (images and video), or import ad creatives or social post media with visual preview. AI video is made only here (asset_type video). Use it whenever the user wants to add an image or video. AI images generated here are shown to the user before any campaign or post uses them. Pass asset_url or asset_prompt in params to auto-start; if the user already has an asset URL or ID, pass it directly to manage_campaign instead. Every call shows the picker; to just look up existing images, use campaign_and_resource_get resource_type assets.
Inputs: ds_id, action, params, account_id, +1 more. Returns: data, error, notes, success, +1 more
data_queryARead-onlyInspect
Query data from Google Ads, Facebook Ads, Google Analytics, LinkedIn Ads, TikTok Ads, Microsoft Ads, Shopify, HubSpot, and 200+ other platforms. Use data_source_discovery, accounts_discovery and field_discovery first. Returns rows when completed; else pass schedule_id to get_async_query_results. Team- and data-source-level context arrives with data_source_discovery/accounts_discovery, and a narrowed accounts_discovery (≤3 accounts) inlines an account's own context — apply what you already have. Only when you are querying a specific account whose context you have not seen, read it first with get_business_context (ds_id + account_id) for reporting guidelines — e.g. which campaigns to exclude, how to segment data, preferred metrics.
If the user specifies reporting preferences during the conversation (e.g. "exclude brand campaigns", "always break out Advantage+ separately", "for this client we only look at last-click attribution"), ask whether to save these as context for future sessions using manage_business_context with action 'save'.
Inputs: ds_id (required), fresh, fields, filters, +12 more. Returns: data, error, notes, success, +1 more
data_source_discoveryARead-onlyIdempotentInspect
Discover data sources or get a source's configuration. Without ds_id: lists all sources with auth status; data-out-capable sources carry data_out: true (can receive activated audiences), others omit it. With ds_id: returns config (capabilities, report_types, settings, login link if needed). With filter: narrows the list; single match returns full config automatically.
Inputs: ds_id, filter, compress. Returns: List mode returns data.sources; config mode returns data.id, data.name, etc.
field_discoveryARead-onlyIdempotentInspect
Discover available fields (metrics and dimensions) for a data source. Only call if data_source_discovery config shows has_fields=true. Fields in a query must share a common report_type if the source uses report types.
Inputs: ds_id (required), filter, compress. Returns: data, error, notes, success, +1 more
get_async_query_resultsARead-onlyIdempotentInspect
Retrieve results from a data_query using its schedule_id. Call immediately; the call waits, so it usually returns data first time. While pending, call again immediately; stop when status is 'completed' or failed. Map columns by requested_field_ids/canonical_field_ids, not the row-0 display names.
Inputs: schedule_id (required), compress. Returns: data, error, notes, success, +1 more
get_business_contextARead-onlyIdempotentInspect
Retrieve the durable business context a team has saved — guidelines, policies, and preferences — so analysis and changes follow them. Context is stored at three scopes, each read independently (a read returns one scope's entries, not a merge):
Team — applies everywhere (e.g. "report currency in EUR").
Data source — a specific platform (e.g. "for Meta, break out Advantage+ separately").
Account — a specific account (e.g. "this client reports last-click only").
Choose the scope with the ids: no ids → team level; ds_id → that data source's level; ds_id+account_id → that account's level. tags filters by category — a record matches if it carries any of the tags you pass (OR), and commas are just separators, so tags: ["a,b"] and ["a","b"] both match records tagged a or b. Omit tags to read all entries at the scope; tags with no ids searches across all scopes, so a tag lookup isn't confined to one level. Results are grouped as team/data_source/account (empty omitted); a single-source data-source list is keyed data_source:<ds_id>.
Prefer what the discovery tools already surface: data_source_discovery and accounts_discovery include team- and data-source-level context, and a narrowed accounts_discovery inlines an account's own context. Call this for a scope a discovery response did not cover — e.g. an account you are about to act on — to read tags whose context was too large to inline, or to re-read after a save.
Inputs: tags, ds_id, account_id. Returns: data, error, notes, success, +1 more
get_dashboardsARead-onlyIdempotentInspect
Read Supermetrics Studio dashboards: 'list' the dashboards the user owns or has been shared with (metadata only, no code, newest first; optional 'filter' narrows by id, title, description or tag, and 'limit'/'offset' paginate — the response carries 'total'), 'get' one dashboard's source code and metadata by ID, or 'changelog' its version history. Load the current code with 'get' before editing.
Inputs: action (required), limit, filter, offset, +1 more. Returns: data, error, notes, success, +1 more
get_support_ticketsARead-onlyIdempotentInspect
List the user's Supermetrics support tickets, or read one ticket's thread by ticket_id. Only the caller's own tickets are accessible. Check this before creating a new ticket: if an open one already covers the issue, reply to it with contact_supermetrics instead.
Inputs: ticket_id. Returns: data, error, notes, success, +1 more
get_todayARead-onlyIdempotentInspect
Get current UTC date and time. Use before data_query with custom date ranges to resolve relative date references (e.g. "last month") into specific dates.
Inputs: no named parameters. Returns: data, error, notes, success, +1 more
get_user_and_teamARead-onlyIdempotentInspect
Read your Supermetrics account and team: your profile, permissions, license details and team members (get_info); a data source's prioritised (pinned) accounts with the cap and monthly change quota (get_prioritised_accounts, ds_id); or a login link for a data source (login_data_source, ds_id). Use get_info to find license_id and team member user_ids before assigning users to a subscription.
Inputs: action (required), ds_id, include_management_scopes. Returns: data, error, notes, success, +1 more
instagram_insightsARead-onlyInspect
Instagram and Facebook (Meta) social media analytics and insights, using the same parameters and result format as data_query. Returns rows when completed; otherwise pass schedule_id to get_async_query_results. Metrics include engagement, engagement rate, reach, accounts reached, accounts engaged, followers, fans, follower growth, impressions, profile and page views, website and link clicks, top posts and best-performing content, content performance, stories, reels, video views, saves, likes, comments, shares, and audience demographics. Data sources (pass one as ds_id):
"IGI" — your connected Instagram business/creator account insights (media, reels, stories, top posts, follower demographics);
"IGPD2" — public data on any Instagram profile or hashtag (competitor benchmarking, business discovery, hashtag search);
"FB" — organic Facebook Page insights (page and post engagement, reach, fans/followers, impressions);
"FA" — Facebook & Instagram paid ad performance (Meta Ads); add the "publisher_platform" dimension to separate Facebook from Instagram placements (values include facebook and instagram; filter or segment on publisher_platform = instagram for Instagram-only ad metrics). Then use accounts_discovery to find the account IDs to pass in ds_accounts, and field_discovery to find the available field/metric IDs (and report types) for your chosen ds_id. To create or update Instagram (or Facebook) ad campaigns, use the manage_campaign tool. For any other platform (Google Ads, GA4, LinkedIn Ads, TikTok, and 170+ other data sources), use the data_query tool.
Inputs: ds_id (required), fresh, fields, filters, +12 more. Returns: data, error, notes, success, +1 more
manage_activation_queryADestructiveInspect
Create or update an audience activation: the source data to pull, which fields are email or phone identifiers, and optional destinations. Omit activation_query_id to create; provide it to update. On update, send only the parts that change: an omitted source or destinations is kept, but a source or destinations list you send replaces the stored one and must be complete ([] removes all destinations). Returns its state (activation_query_id, ready_to_run, missing). Saving stores configuration only and reads or sends no records; identifiers are SHA-256 hashed before upload, and sending happens only in a separate run step the user confirms. Discover source/accounts/fields with the data_source_discovery, accounts_discovery, and field_discovery tools first, and each destination's audience with activation_audience_list. A source that needs connector settings (e.g. BigQuery's date_column) takes them in source.settings. Confirm the setup with the user before saving. For a Google Ads (AW) destination, show the user this notice verbatim before setting its terms acceptance and consent, and set them only from their answer, never on your own: "By setting the consent status to "granted", you confirm on behalf of the organisation you represent that it complies with Google's Customer Match Policy and, for any end users in the EEA, UK or Switzerland in this audience, that it has provided all required notices and obtained valid consent to the sharing of their data with Google for advertising and ad personalization, as required by applicable law and Google's EU User Consent Policy. You instruct Supermetrics to send this consent status to Google, and Supermetrics relies on this confirmation."
Inputs: source, destinations, activation_query_id. Returns: data, error, notes, success, +1 more
manage_audits_and_monitorsADestructiveInspect
Start a one-off audit of one ad account when asked how it is doing or what's wrong; use data_query for specific metrics, custom dates, other sources or several accounts. Or create, update or run a scheduled monitor: metrics, page_check or naming_convention. Supported sources are listed on ds_id.
Inputs: action (required), name, ds_id, config, +10 more. Returns: data, error, notes, success, +1 more
manage_business_contextADestructiveInspect
Save or delete the durable business context a team applies consistently and shares with all members: reporting conventions, analysis rules, brand and tone, data governance, naming, plans, or account-specific preferences. Context is stored at team, data source, or account scope, chosen by ds_id/account_id.
save— Create/update at a scope; needscontext_textandtags. Tags identify the entry within a scope, so a scope can hold several entries (one per tag set):mode: "replace"(default) overwrites the entry with those tags,mode: "append"concatenates to it, and a different tag set is saved as a separate entry rather than overwriting.delete— Remove the single entry identified by scope (ds_id/account_id) andtags— it targets that one tag set, never the others at the scope. Omittagsto remove the untagged entry. Irreversible, so confirm first.
When to save: when the user shares a plan, guidelines, or an SOP; states a durable preference ("for this client, always…"); or corrects your approach — offer to save it so it applies in future sessions. Distill the key, reusable rules (do not store documents verbatim) and tag them by category (e.g. "data-analysis", "targeting", "naming-conventions", "brand-visuals", "brand-language"). Tags are how the entry is retrieved later, so avoid tags that merely restate the scope. Read the current context with get_business_context before changing it.
Inputs: action (required), mode, tags, ds_id, +2 more. Returns: data, error, notes, success, +1 more
manage_campaignADestructiveInspect
Create or update a campaign (AW/FA/TIK/LIA/AC/CGPTA/SCM). Provide campaign_id to update an existing campaign; omit it to create a new one (a budget is required when creating). On update only provided fields change. Also creates/updates ad groups/ads — see ad_groups (operations inferred from id presence). Partial success possible: check write_status and failures if present, otherwise notes; don't retry blindly. Creative: use asset_url (public https:// URL), asset_id, upload_ref (from creative_picker), or object_story_id (existing post). AI images: creative_picker shows them to the user and returns an upload_ref; a *_prompt field generates an image the user has not seen. Before writing to a platform, read its rules with supermetrics_guide mode: AW google_ads_capabilities, FA facebook_meta_capabilities, AC microsoft_advertising_capabilities, TIK tiktok_ads_capabilities, LIA linkedin_ads_capabilities, CGPTA chatgpt_ads_capabilities, SCM snapchat_marketing_capabilities. Platform rules are not repeated here. Use campaign_and_resource_get for full details after writing.
Inputs: ds_id (required), account_id (required), name, status, +15 more. Returns: data, error, notes, success, +3 more
manage_dashboardsADestructiveInspect
Share live dashboards in Supermetrics Studio — self-contained pages that re-query the user's data on each view, so a shared link always shows current numbers.
A new board is private: only its owner can open it. smQuery and window.fetch work only inside Studio, so a chat preview that uses them renders empty; a working chat preview has the already-fetched data hard-coded (no smQuery), while the uploaded code fetches through smQuery (see the 'code' param).
Campaign actions (pause, budget change) on a board need the campaign tools in allowed_tools and the managed accounts in allowed_accounts. Everyone the board is shared with acts on the owner's connections and can make those changes; a per-change prompt in the page exists only if the board's code includes one.
BigQuery dashboards: the response shape and allowed_accounts key differ from Supermetrics sources; supermetrics_guide(mode="dashboard_bigquery") describes them.
Actions:
'upload': Upload a new dashboard or replace an existing one. Requires title, code, allowed_tools, allowed_accounts; description is optional. The code format (HTML or JSX) is detected automatically.
'edit': Apply targeted anchor-based patches to dashboard code. Patches apply to the current code, which get_dashboards returns.
Edit patch types (pick the smallest that fits):
region replace (PREFER for changing a section's logic/content): {"from": "// #region charts", "to": "// #endregion charts", "replace": "// #region charts\n\n// #endregion charts"} — replaces the WHOLE region from 'from' to 'to', anchor lines INCLUDED. Write the COMPLETE new block in 'replace' and reproduce the 'from' and 'to' marker lines inside it (normally unchanged) so they survive; omitting a marker deletes it. Restating an anchor does NOT duplicate it. 'from' must be unique; 'to' is the first match after it.
find/replace (small unique swap): {"find": "old text", "replace": "new text"}
after/insert (a NEW element): {"after": "anchor text", "insert": "new content"}
before/insert: {"before": "anchor text", "insert": "new content"} Modifiers (find/replace only): "nth": N (target Nth occurrence), "all": true (replace all). Patches are applied sequentially and atomically.
Patchable code: When uploading, wrap each section in PAIRED region markers so edits can target them, and declare each variable/function inside the region that uses it:
HTML: …
CSS: /* #region kpi-cards /…/ #endregion kpi-cards */
JS: // #region chart-init … // #endregion chart-init
Tag key elements with data-sm attributes: data-sm="kpi-spend", data-sm="filter-date" Each anchor must be unique in the code.
Best practices:
Make it visually polished and impressive, while keeping to what the user asked for. Define colors and fonts once as CSS variables in :root (--bg, --primary, --accent, --series-1..N, --font-heading, --font-body) and have charts read them, so the Studio Design tab can retheme the board.
Design: open with one hero insight and vary panel sizes, not a grid of equal cards; pick the chart that fits the data's shape (treemap, sankey, small multiples, heatmap), not always bar/line; annotate from the data (peaks, targets, deltas); sparklines in KPI cards; inline SVG icons, not emoji.
Motion: staggered, eased entry animations; on a filter change tween existing marks into their new state instead of redrawing; count KPI numbers up; tooltips on hover; no motion under prefers-reduced-motion. A working plain board beats a broken flashy one.
Add a loading indicator (spinner or skeleton) while data is being fetched.
When it makes sense, add selectors (dropdowns, date pickers, toggles) that let viewers change query parameters such as date range, account, or metric.
Date controls: offer a genuinely flexible date range, not just a few fixed presets. smQuery's date_range_type accepts today, yesterday, this_month(_inc), last_month, this_year_inc, last_year(_inc), and the last_x_days / last_x_weeks_sun_sat / last_x_weeks_mon_sun / last_x_months / last_x_years families — substitute a number for x (e.g. last_90_days, last_3_months) and append _inc to include the current period. There is NO native quarter or year-to-date preset: deliver 'this/last quarter', 'year to date', and any viewer-picked custom range via date_range_type="custom" with start_date/end_date (YYYY-MM-DD) computed in the dashboard's own JS (derive the boundaries from new Date()). Prefer including a fully custom start/end picker alongside the presets.
Period comparison: when comparison is wanted, add ONE dashboard-wide comparison control and feed the SAME compare params into EVERY widget's smQuery — time-series charts too, not only scorecards. compare_type="prev_range" (previous period), "prev_year" (same period last year), "prev_year_weekday", or "custom" (with compare_start_date/compare_end_date, YYYY-MM-DD); compare_show="perc_change" (default), "abs_change", or "value". For a year-over-year time-series overlay, either use compare_type or issue a second smQuery for the prior period and plot it as an additional series.
Give each chart/table/KPI its own stable smQuery key (see the 'code' param) so rapidly changing a selector can't show stale data from a previous selection.
code_warningsin an 'upload' response name what makes the board render empty or wrong data or never refresh (a static board with no smQuery calls, or column mapping via requested_field_ids); the board shows that problem until a corrected upload.
Inputs: action (required), code, title, patches, +5 more. Returns: data, error, notes, success, +1 more
manage_social_mediaADestructiveInspect
Manage social media comments, publishing and scheduling on Instagram (ds_id IGI, default), LinkedIn pages (ds_id LIP), TikTok (ds_id TIKBA, account_id = username) and Google Business Profile (GMB); writes may return a link the user must approve; to analyse, query ds_id IGI (get media_id there), IGPD2, TIKBA (video_id is the media_id) or LIP (update_id is the number in a share- or ugcPost- media_id; include today, e.g. last_7_days_inc, for new posts). LinkedIn media_id: share-, ugcPost- or the post URN; comment ids come from list_comments, pass them unchanged. GMB account_id is a location from accounts_discovery; reviews are its comments; an owner reply shares its review's id, and delete removes only the reply. Workflows: supermetrics_guide mode social_publishing. AI images: creative_picker shows them to the user and returns an upload_ref; media_prompt(s) generate images the user has not seen. Actions and their parameters (all take ds_id): list_comments(account_id, media_id (omit on GMB; on TikTok omit for the latest posts); limit, cursor) | comment(account_id, message; media_id OR comment_id to reply) | moderate_comment(account_id, comment_id; hidden true/false (Instagram, TikTok) OR delete true) | publish(account_id, content_type IMAGE/VIDEO/CAROUSEL/STORY on Instagram, TEXT/LINK/IMAGE/CAROUSEL on LinkedIn, IMAGE/CAROUSEL/VIDEO on TikTok (always a job; TikTok reviews the post), TEXT/IMAGE on GMB; caption, alt_text, title, link_url; media_url OR upload_ref (creative_picker uploads, or AI-generates with preview; existing saved images: campaign_and_resource_get resource_type assets, ds_id IGI, without opening the picker) OR media_prompt (an AI image the user has not seen), CAROUSEL: media_urls/media_prompts, STORY sequence: frames, on_failure; style, reference_image; platform_options (Instagram, TikTok, GMB); publish_at, timezone, spacing_minutes; best Instagram time: IGI online_followers by hour; space one account's scheduled posts 5+ minutes apart, as Instagram throttles bursts) | update_post(account_id, media_id, comments_enabled (not TikTok) OR scheduled_id with publish_at/timezone/caption/alt_text/title/platform_options) | cancel_scheduled(scheduled_id) | status(job_id [+wait_seconds] OR scheduled_id OR approval_ref OR quota+account_id (Instagram); none lists scheduled posts, on every platform without ds_id: account_id, state, limit, cursor). Scheduled posts, jobs and approvals are found by id, without ds_id. every write except cancel_scheduled takes validate_only; repeating the same publish or comment within 15 minutes returns the first result instead of writing twice. Follow each result's next_action.
Inputs: action (required), ds_id, limit, quota, +33 more. Returns: data, error, notes, success, +1 more
manage_user_and_teamADestructiveInspect
Change your Supermetrics team: invite a member (invite_member, email + role; sent only after the user approves it on a link you show them), assign team users to a subscription (add_user_to_subscription, license_id + user_ids), or replace a data source's prioritised (pinned) accounts (set_prioritised_accounts, ds_id + account_ids). Prioritised-account changes consume the monthly change quota. Read the current state with get_user_and_team first to find license_id, member user_ids and the existing pins.
Inputs: action (required), role, ds_id, email, +3 more. Returns: data, error, notes, success, +1 more
marketing_utilitiesARead-onlyIdempotentInspect
Free marketing utilities: no Supermetrics subscription or connected data source needed.
Actions:
utm_build: tag each of the urls with source/medium/campaign/term/content, correctly encoded (macros like {campaignid} kept), validated against your team's conventions.
utm_parse: check the urls against your team's conventions and GA4 channel rules. For one campaign used in several places (an event's booth QR code, printed cards, follow-up email, social post), build one link per placement with a different utm_content, so reports show which placement worked. utm_build and utm_parse return suggested_url (the link with casing/separator fixed), fixed (what changed in it) and unresolved (issues that need a decision, e.g. an unapproved source). Show suggested_url and fixed in the same reply as the check, then ask only about unresolved; never wait to be asked for a fixed link.
pagespeed_check: PageSpeed Insights for up to 10 urls (run in parallel; 30s-2min): scores, lab and field vitals, top fixes; detail="full" for every finding. Returns within ~45s; if checks are still running it returns status="running" and a job_id: poll with {"action":"pagespeed_check","job_id":...} about every 30s. Results are cached for 24h (cached=true, checked_at); pass refresh=true only after the page has changed.
link_check: follow up to 50 urls through redirects; do UTM and click-id params (gclid, fbclid...) reach the landing page, and at which hop are they lost. Also broken pages, loops, tagging mistakes.
web_vitals_trend: ~6 months of real-user LCP/INP/CLS/FCP/TTFB for up to 50 sites (bare domain) or pages.
safety_check: up to 50 urls, domains or email addresses (e.g. a cold-pitch sender) and every host they redirect through: Web Risk malware/phishing flags, invalid or expiring certificates, domains expiring soon or registered in the last 30 days, and senders with no working website.
link_preview: how up to 50 urls render when shared on LinkedIn, Facebook, X and Slack; flags missing or small og:image, missing description, truncated titles.
email_dns: SPF, DKIM, DMARC, MX and BIMI for up to 50 domains (urls may be domains, emails or URLs).
char_limits: exact lengths of up to 200 texts against Google Ads, Meta, LinkedIn and X limits (counted as each platform counts). Use it whenever copy must fit a limit; do not count characters yourself. Google asset fields (path, long_headline, business_name, callout, sitelink_text, sitelink_description) are checked only when named in field.
Results: every result has input (as sent), url, status (ok|warning|critical|unknown) and findings[] of {code, severity, message, suggestion?, ...extras}; unknown results carry error {code, message} (e.g. quota_exhausted, daily_limit, no_data, url_throttled). Each batch has a summary of counts.
UTM conventions come from business context (tags='utm-conventions'). A JSON object is enforced; every key is optional; lists are exact, case-sensitive matches: {"approved_sources":[...],"approved_mediums":[...],"blocklist":{"utm_campaign":["test"]}, "regex":{"utm_campaign":"\d{4}"},"max_length":{"utm_campaign":60},"casing":"lowercase"|"as_is", "separator":"-"|"_","require_all_core":true} When utm_source is a known ad network (google, bing, facebook/instagram, tiktok, linkedin, snapchat, pinterest), utm_build returns network {medium, campaign_macro, id_macro, click_id} and both actions flag macros written in another network's syntax, which that network would leave unfilled. utm results report conventions_source: team (JSON enforced), team_free_text (returned as team_conventions for you to apply) or defaults (lowercase, hyphens).
Inputs: action (required), term, urls, field, +11 more. Returns: data, error, notes, success, +1 more
read_audits_and_monitorsARead-onlyIdempotentInspect
Read monitors and audits. Monitors are scheduled checks: metrics, page_check and naming_convention; supported sources are listed on ds_id. An audit is a one-off review of one ad account. The metric and dimension ids in action="schema" config_example are from field_discovery — never guess them. Operations: list (no arguments — your monitors); get (monitor_id — one monitor's details, including its editable configuration); schema (monitor_type and ds_id — the fields a monitor of that type accepts); audit (run_token from manage_audits_and_monitors(action='audit') — waits for the audit and returns its report). Omit action and it is inferred: run_token implies audit, monitor_id implies get, monitor_type implies schema, otherwise list.
Inputs: ds_id, action, run_token, monitor_id, +1 more. Returns: data, error, notes, success, +1 more
supermetrics_guideARead-onlyIdempotentInspect
Explain what Supermetrics can do, what's new, or how to fix a common problem.
Use mode="tour" (the default) for what the user can do; mode="whats_new" for recent changes. Not for looking up data sources, accounts or fields. Call tour/whats_new at most once per conversation.
TROUBLESHOOTING: when the user reports a common problem, fetch the matching guide and walk them through it before offering a support ticket. Connecting / access: connect_data_source (sign-in / "no accounts found"), google_connectors (GA4/Search Console/YouTube/etc. access), meta_instagram (Facebook/Instagram setup), microsoft_connectors (Microsoft Ads/Bing Webmaster), linkedin_connectors (Ads vs Company Pages), tiktok_connectors (Ads vs Organic), x_twitter_connectors (X Ads vs X Organic), amazon_connectors (Ads/DSP/Seller/Vendor), data_warehouse (BigQuery/Snowflake), reconnect (expired connection). Licensing and accounts: license_and_trials (LICENSE_NOT_FOUND / trial expired), cancel_trial (cancel a trial or subscription, "will I be charged?"), data_source_on_plan (LICENSE_DATA_SOURCE_NOT_AVAILABLE), destination_on_plan (AI Chats destination missing), prioritized_accounts ("not a prioritised account"), manage_users_and_accounts (adding users/accounts and making a new account usable), pricing_and_purchasing (buying questions: plans, how accounts are counted, AI credits, billing). Data and queries: no_data (empty results), data_discrepancies (numbers don't match the platform), ga4_data (GA4 sampling/thresholding/partial data), query_errors (date/field/account errors), quotas_and_limits (quota exceeded). Campaigns: write_access. Social posts: social_publishing (publishing, scheduling, comments with manage_social_media). Each returns steps the user can follow without Supermetrics staff.
HOW TO: google_sheets — fetch before putting Supermetrics data into a Google Sheet. It builds the sheet so the Supermetrics add-on refreshes it automatically on a schedule (or on demand). dashboard_bigquery — fetch before building a Studio dashboard on BigQuery data.
CAPABILITIES: before writing to a platform, fetch its capability guide (one per connector, see mode) for the operations, platform_settings keys, targeting fields and platform rules (limits, refusals) that connector applies — kept with the connector code, so it beats guessing from an error.
Inputs: mode. Returns: data, error, notes, success, +1 more
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Related MCP Connectors
MCP server for querying and analyzing data from ad platforms, analytics tools, and spreadsheets
Query marketing data across 20+ ad platforms; with write access, create, schedule and share reports.
Query selected Google and Bing analytics, search, and advertising reports through MCP.
Marketing-measurement MCP: dashboards, cross-channel queries, integrations, and knowledge files.
Related MCP Servers
- AlicenseAqualityBmaintenanceEnables MCP-compatible clients to query advertising data, analyze spend, conversions, and audiences, compare campaigns, generate dashboards, and manage campaigns with optional read-only safeguards.28Apache 2.0
- AlicenseAqualityBmaintenanceEnables MCP clients to read and analyze Facebook (Meta) Ads marketing data, including campaigns, ad sets, ads, insights, and 190 metrics, with optional write tools that preview changes before applying.412Apache 2.0
- AlicenseNot gradedqualityDmaintenanceUnified MCP server for managing Meta Ads, LinkedIn Ads, Google Ads, GA4, and Search Console with 89 read/write tools, multi-account support, OAuth setup, and safe dry-run mutations.MIT
- AlicenseCqualityCmaintenanceMCP server for Salesforce marketing and revenue ops teams. 47 tools covering leads, contacts, accounts, campaigns, campaign members, tasks, and 17 reporting tools including campaign ROI, lead-source attribution, pipeline-by-campaign, multi-touch campaign influence, MQL trend, forecast summary, and the native SFDC Reports API.471-
Glama MCP Gateway
Add one secure layer between your agents and this server.