"Google Campaign Manager 360" matching MCP tools:

• create_linkedin_ads_creative

  Hermoso

  Create the AD ITSELF on LinkedIn — a CREATIVE inside an existing campaign. A LinkedIn campaign holds no copy and no visual, so until this runs the campaign CANNOT show an impression no matter what its status says; say that rather than calling a campaign "live". Two ways in: pass postUrn to sponsor a post that already exists (LinkedIn’s "boost this post"), or pass text and/or imageUrl / videoUrl to author a DIRECT SPONSORED CONTENT post — a real post by a company Page the user administers that is NEVER shown on the Page’s feed and exists only as an ad. Created DRAFT: it spends nothing until you activate it with set_linkedin_ads_status(level:"creative", status:"ACTIVE", confirm:true). Creating it ACTIVE under an already-live campaign starts REAL AD SPEND on the very next auction and therefore requires confirm:true. The whole tree — group, campaign and every creative — is read back from LinkedIn before you are told anything exists.

  Connector

• create_google_ads_conversion_action

  Hermoso

  Create a CONVERSION ACTION — the thing that tells Google what counts as a result on this account. This is a PREREQUISITE, not a nicety: MAXIMIZE_CONVERSIONS, MAXIMIZE_CONVERSION_VALUE, TARGET_CPA, TARGET_ROAS and every Performance Max campaign are undeliverable without one, because Google has nothing to optimise toward. type WEBPAGE (a purchase / lead / signup on the site — the normal choice), UPLOAD_CLICKS or UPLOAD_CALLS; every other Google conversion type (Firebase, Google Analytics 4, Floodlight, store visits) is READ-ONLY and is created in those products, not here. Set category to what actually happened (PURCHASE, SUBMIT_LEAD_FORM, SIGNUP, BOOK_APPOINTMENT…) and defaultValueUsd when a conversion has a known worth — TARGET_ROAS has nothing to maximise without a value. Created ENABLED and counted in "conversions" by default, because a conversion action that is neither records nothing. It CANNOT SERVE AN AD and cannot spend a cent, so it needs no confirmation. A WEBPAGE action records NOTHING until its Google tag is installed on the site — say that when you report it.

  Connector

• create_google_ads_performance_max_campaign

  Hermoso

  Build a PERFORMANCE MAX campaign — Google’s cross-surface campaign type (Search, YouTube, Display, Discover, Gmail, Maps) and the one Google pushes hardest at small advertisers. ALWAYS created PAUSED; it spends NOTHING until you enable it with set_google_ads_status(confirm:true). PMax has NO manual bidding and NO keywords: it bids only on conversions, so the account MUST already have a conversion action — check with list_google_ads_conversion_actions, because this REFUSES rather than build a campaign that cannot optimise. Creative lives in an ASSET GROUP, and Google’s minimums are enforced before anything is sent: 3–15 headlines (≤30 chars), 1–5 longHeadlines (≤90), 2–5 descriptions (≤90), one businessName (≤25), at least one LOGO (1:1), one MARKETING_IMAGE (1.91:1) and one SQUARE_MARKETING_IMAGE (1:1) — upload the images with upload_google_ads_asset first and pass their asset resource names. A YouTube video is optional (Google generates one from the asset group if you omit it). Brand guidelines: since Google Ads API v21 they are ON by default for new PMax campaigns, which means the businessName and LOGO assets are linked to the CAMPAIGN (CampaignAsset), not to the asset group — Hermoso does that for you. Leave brandGuidelinesEnabled alone unless the user wants the older asset-group layout, and pass false for that. Budget, campaign, location/language targeting, the asset group and every asset link go up in ONE ATOMIC operation — if any part is rejected, nothing at all is created — and the whole tree is READ BACK from Google before you are told it exists. Print the returned note verbatim; if it says the campaign cannot serve, say that instead of calling it finished. RETAIL/SHOPPING: pass merchantCenterId (from list_merchant_accounts) to make it a Shopping-feed Performance Max — it then advertises the WHOLE feed (one root listing group); feedLabel narrows it to a single feed. Partitioning the feed by brand/category/custom label is NOT built and is refused by name, so a caller can never believe they narrowed it when they did not.

  Connector

• delete_google_ads_object

  Hermoso

  PERMANENTLY remove a Google Ads object. Google has no HTTP delete — removal is a `remove` operation that puts the object in the terminal REMOVED state, which cannot be undone or re-enabled, so treat it as a delete. Levels: "campaign" + campaignId · "adGroup" + adGroupId · "ad" + adGroupId AND adId · "keyword" + adGroupId AND keywordId · "conversionAction" + conversionActionId · "campaignAsset"/"adGroupAsset" + the LINK’s full resourceName (get it from google_ads_report over campaign_asset / ad_group_asset — an asset id alone does not identify a link). THERE IS DELIBERATELY NO "asset" LEVEL: Google publishes no operation that deletes an Asset, only its links, so removing a link unlinks the asset and leaves it in the library. CALL IT WITHOUT confirm FIRST — nothing is removed and you get the object’s real name, status, LIFETIME SPEND and child counts read live from Google; show the user exactly that. A target with children, live delivery or real spend additionally needs confirmName (its exact name) and confirmChildren (the child count from that read-back). Removing a CAMPAIGN also removes its campaign-owned budget, and the note says whether it did. Removing the last ENABLED conversion action makes every smart-bidding campaign on the account undeliverable — the refusal says so. To stop delivery without removing, use set_google_ads_status(status:"PAUSED").

  Connector

• set_geo_targeting

  Growomat

  Set a campaign's WHOLE geo-targeting set in one call — named areas and radius circles in a single include/exclude list, plus the campaign-level presence-vs-interest choice — the easy front door over update_campaign's geoTargets. Place names in include/exclude are resolved server-side (countries → geo registry → Google suggest); radius entries take lat/lng or an `around` place name (geocoded). Radius rows can only be INCLUDED (neither platform supports negative proximity — carve out places with an excluded area instead), and a set where every row is negative is rejected (ONLY_EXCLUSIONS). An empty set is legal: no geo criteria = ads can serve anywhere. locationIntent defaults to PRESENCE (people actually in the locations) — deliberately NOT the platforms' wider search-interest default. Use dry_run:true to get the resolved rows + the plain-language effective-targeting sentence (exactly what the UI shows) WITHOUT mutating the campaign. Scoped to the active Space.

  Connector

• acc_list_projects

  ACC MCP

  Enumerate every ACC and BIM 360 project the authenticated APS app can see by walking all accessible hubs and their project lists. When to use: The agent needs to discover project IDs before calling any other tool (e.g. the user says 'show me my projects' or 'find issues in the Tower project' and no project_id is known yet). Also useful to confirm hub membership for a project. When NOT to use: Do not call this repeatedly in a loop — cache the result; if the user already supplied a project_id starting with 'b.', skip discovery. APS scopes: data:read account:read. No write scope needed. Rate limits: APS default ~50 req/min per app per endpoint; BIM 360 hubs endpoints are pageable (limit 200). This tool fans out 1 hubs call + N project calls (one per hub) so call it sparingly on tenants with many hubs. Errors: 401 (APS token expired — refresh and retry once); 403 (app not provisioned in the BIM 360/ACC account — ask user to have an account admin add the APS client_id); 404 (rare, indicates hub deleted mid-call); 429 (rate limit — back off 60s); 5xx (ACC upstream — retry with jitter). Side effects: None. Read-only and idempotent.

  Connector

• campaign_results

  agent-cold-email

  Outcome counts for ONE campaign. Input: campaignId (from launch_campaign). Returns { campaignId, sent, reply, bounce, complaint, unsubscribe, failed, soft_bounce } — bounce = HARD only, soft_bounce separate, opens not tracked. 404 if unknown. Use metrics for account-wide totals, list_campaigns for every campaign at once.

  Connector

• metrics

  agent-cold-email

  Account-wide outcome totals across ALL campaigns: { sent, reply, bounce, complaint, unsubscribe, failed, soft_bounce } — same shape as campaign_results but summed tenant-wide (bounce = hard only, opens not tracked). Use campaign_results for one campaign, list_campaigns per-campaign, or account for billing/quota.

  Connector

• set_google_ads_targeting

  Hermoso

  ADD locations and languages to an existing Google Ads campaign. THIS ADDS; IT DOES NOT REPLACE — Google campaign criteria are a list, this call only ever creates entries, and there is no remove operation here. So a campaign already targeting the United States that you "change to Canada" ends up targeting BOTH and still spending in the US; the read-back names every pre-existing location and language it kept, and you MUST relay that rather than reporting the new total as the answer. Removing targeting is done in Google Ads (Campaign ▸ Settings ▸ Locations). Pass locations by NAME ("United States", "California", "Toronto") — they are resolved to Google's geo target ids for you; excludedLocations adds a NEGATIVE criterion (the reliable way to stop serving somewhere from here); languages takes ISO codes ("en","fr"). A campaign with NO location targeting runs WORLDWIDE, which is the most expensive default in Google Ads. Changing a LIVE campaign's targeting moves real spend immediately, so that needs confirm:true.

  Connector

• create_microsoft_ads_campaign

  Hermoso

  Build a campaign on a connected Microsoft Advertising (Bing Ads) account. ALWAYS created Paused — it spends NOTHING until you activate it with set_microsoft_ads_status(confirm:true). Microsoft’s object graph is campaign → ad group → responsive search ad → keywords, so a campaign ON ITS OWN CANNOT SERVE AN IMPRESSION: pass adGroup{name, ad{headlines,descriptions,finalUrls}, keywords[]} and this builds the whole tree. Microsoft has NO atomic multi-object write (unlike Google), so the levels are created in sequence and the campaign is DELETED again if anything below it is rejected — you never inherit a half-built campaign. Microsoft requires 3–15 headlines (≤30 chars) and 2–4 descriptions (≤90 chars); expanded text ads can no longer be created at all. dailyBudget is in the ACCOUNT’S currency, not necessarily USD. LOCATION TARGETING: pass locations[] (country / region / city names, ISO country codes, or numeric Microsoft location ids). A Microsoft campaign has NO geo targeting unless it is set, and Microsoft does not require any — so if you pass none, the campaign IS CREATED and serves WORLDWIDE (Microsoft’s own default), and the returned note says so loudly. That is safe at this stage because the campaign is Paused and spends nothing; it is NOT safe to activate without telling the user, so relay the warning. Nothing is created when a location you DID name cannot be resolved (call microsoft_ads_geo_search to disambiguate, then pass the id). Pass worldwide:true to record that everywhere was deliberate and suppress the nudge. The locations are written and READ BACK inside the same rollback as the rest of the tree, so a campaign is either targeted as asked or does not exist. Everything is READ BACK from Microsoft before you are told it exists; print the returned note verbatim, and if it says the campaign cannot serve yet, say that rather than calling it a finished ad.

  Connector

• comms_create_campaign

  Coordinalo — Service Business Operations

  Create a new email campaign with HTML body to send to a segmented audience. Supports variable substitution: {nombre}, {apellido}, {nombre_completo}, {email}, {telefono}, {organizacion}. Use audienceType "predefined" with audienceId "active"/"inactive"/"new"/"withPhone"/"withoutPhone", or "adhoc" with custom filters. On send, links in the body are auto-tagged with UTMs (utm_campaign = stable slug from campaign name) and a legal footer with unsubscribe link is appended. Returns campaign ID and recipient count. Campaign starts as draft — use comms_send_campaign to execute. Requires confirm: true.

  Connector

• audience_360

  Autario Data Analytics Platform

  Audience 360 | the caller's OWN audience report over their connected Google Search Console + GA4 + social (Facebook Page, Instagram, TikTok) connector data, computed deterministically server-side (the exact numbers the user sees in the app | nothing re-derived, nothing estimated). Use this FIRST for any interpretation question about a user's traffic/audience ("why is my AI traffic falling", "which queries are rising", "which pages do AI assistants cite", "how is my funnel doing") | it is far more token-efficient and more faithful than rebuilding KPIs from raw connector tables. Pick only the sections you need: overview (funnel stages + audience segments), channels (weekly channel mix + AI-share shift + brand-vs-generic clicks), queries (top brand/generic queries + 28d risers/fallers + high-impression-low-click opportunities), content (per-page sessions x engagement joined with search demand + AI-cited pages), audience (countries, devices, new-vs-returning, totals), conversions (GA4 key events), social (connected Facebook Page / Instagram / TikTok reach, follower trends, top posts, post-format engagement + IG follower demographics), health (report-vs-API cross-checks). Lists are capped and weekly series bounded; every truncation is marked with an omitted count. Filter with range/channel/countries to sharpen the question. Requires the caller's own autario account (API key or OAuth) with the Audience 360 app connected | see get_app_context("audience-360") for the data map behind it.

  Connector

• suggest_content_campaign

  ProductClank

  Preview a content campaign for a product BEFORE launching it. FREE — nothing is created and no credits are charged. Returns an AI-drafted campaign (title, description, call-to-action) plus whether the user can afford to launch it (1000 credits). Use this to show the user what the campaign would look like and get their approval, then call create_content_campaign. Requires a product_id from search_products; write the campaign_message brief from what you know about the product.

  Connector

• set_linkedin_ads_status

  Hermoso

  Turn a LinkedIn campaign group, campaign or CREATIVE (the ad itself) on (ACTIVE) or off (PAUSED). Pass level:"campaign" + campaignId, level:"campaignGroup" + campaignGroupId, or level:"creative" + creativeId. All three tiers must be ACTIVE for an ad to serve — activating only the campaign leaves a DRAFT creative sitting there showing nothing. ACTIVATING STARTS REAL AD SPEND — you MUST first show the user the campaign and its budget, get an explicit yes, then call with status:"ACTIVE" and confirm:true. Pausing is always safe. The resulting status is READ BACK from LinkedIn along with its servingStatuses before you are told it took: LinkedIn defers validation on drafts, so activation is exactly where a hidden problem surfaces, and the note reports what LinkedIn actually says rather than what was requested.

  Connector

• google_business_account

  Hermoso

  Read the Google Business Profile ACCOUNT that owns one of the brand’s listings — the account name, its type (a personal Google account, a location group, a user group or an organization), the connected user’s role on it (primary owner / owner / manager / site manager), the account’s verification state and the permission level. Use it to answer “can we actually edit this listing?” and “whose account is it on?” before offering an edit that Google would refuse anyway. It reads exactly ONE account — the parent of a listing already shared with this brand — and never lists the other accounts the connected Google login can reach; that roster belongs to the account picker (list_connector_accounts). Read-only, 0 credits. Needs Google Business Profile connected.

  Connector

• create_google_ads_campaign

  Hermoso

  Build a campaign on a connected Google Ads account. ALWAYS created PAUSED — it spends NOTHING until you enable it with set_google_ads_status(confirm:true). Google's object graph is campaign → ad group → ad, so a campaign ON ITS OWN CANNOT SERVE AN IMPRESSION: pass adGroup{name, ad{headlines,descriptions,finalUrls}, keywords[]} and this builds budget + campaign + location/language targeting + ad group + ad + keywords in ONE ATOMIC operation (if any part is rejected, nothing at all is created — no half-built campaign to clean up). Also here: bidding strategy, locations by NAME ("United States", "Toronto" — resolved for you), languages, and start/end dates. Google requires 3–15 headlines (≤30 chars) and 2–4 descriptions (≤90 chars) on a search ad. Everything is READ BACK from Google before you are told it exists; print the returned note verbatim, and if it says the campaign cannot serve yet, say that rather than calling it a finished ad.

  Connector

• google-maps.place_types

  Vee3

  List supported Google Maps place type values for search filters. Returns place_types as a string array. Use a value with place_type on google-maps.search or google-maps.nearby_search. Cost = 1 token.

  Connector

• google-trends.categories

  Vee3

  List all Google Trends category and subcategory labels you can pass to other Google Trends tools in the category field. Returns cat (array of category names, including All categories) and msg. Use this before interest-over-time or interest-by-region calls when filtering by category. Cost = 5 tokens.

  Connector

• google-trends.regions

  Vee3

  List all countries and subregions you can pass to other Google Trends tools in the country and region fields. Returns geo.countries: each country name maps to country (label) and regions (array of subregion names). Also returns msg. Use this before interest-over-time or interest-by-region calls when filtering by geography. Pair with google-trends.categories when filtering by category. Cost = 5 tokens.

  Connector

• acc_project_summary

  ACC MCP

  Fetch the full ACC project metadata record (name, type, status, dates, extension attributes) for a single project via APS Data Management. If hub_id is omitted the tool picks the first accessible hub, which may be wrong on multi-hub tenants. When to use: The user asks 'tell me about project X' or an agent needs project metadata (start/end dates, type, Forma/BIM 360 flavor) before deciding which downstream tool to call. When NOT to use: Do not use as a cheap existence check — prefer acc_list_projects which returns hub_id with every project and is one call regardless of tenant size. APS scopes: data:read account:read. Forma / BIM 360 hubs endpoints only require data:read. Rate limits: APS default ~50 req/min per endpoint; BIM 360 hubs endpoints pageable (limit 200). Cache results for the session. Errors: 401 (APS token expired — refresh); 403 (user lacks project view or app not in account); 404 (project not in the chosen hub — supply the correct hub_id, or call acc_list_projects first); 422 (malformed project_id — confirm 'b.' prefix); 429 (rate limit — back off 60s); 5xx (ACC upstream — retry). Side effects: None. Read-only and idempotent.

  Connector