Google Ads
<!-- BEGIN:skill-intro -->
Tools for working with a Google Ads account against the
Google Ads API (
https://googleads.googleapis.com/v23/
). Reads are expressed in
GAQL (Google Ads Query Language) against the search endpoint; writes go through the per-resource mutate endpoints. 13 tools across account navigation, reads and reporting, and campaign / budget / conversion-tracking management. Google Ads organizes accounts as a hierarchy: a
manager (MCC) account can operate
client accounts beneath it. Most tools take the operating account's
, plus an optional
(the manager account) when the operating account is reached through a manager.
<!-- legal:disclaimer -->
Independent, unofficial connector for Google Ads. Not affiliated with, endorsed by, or sponsored by Google Ads. "Google Ads" is a trademark of its owner, used only to identify the service this connector works with.
<!-- /legal:disclaimer -->
<!-- END:skill-intro -->
When to use this
<!-- BEGIN:skill-use-cases -->
- Resolve which account to act on — list the accounts the connection can access, then (when access is through a manager) the client accounts beneath it.
- Read campaigns, ad groups, and ads — list them with status and budget, or run an arbitrary GAQL query for anything the structured reads don't cover.
- Build performance reports — pick a resource, the metrics to measure, and a date range.
- Manage campaigns and budgets — pause, enable, or remove a campaign; create or adjust a daily budget.
- Set up conversion tracking — list or create conversion actions (including the offline-conversion action).
<!-- END:skill-use-cases -->
Setup
This is an
agentskills.io skill.
If the connector has not been installed as a skill yet, install it first with
npx skills add zapier/connectors --skill google-ads
(or your harness's own skill-install mechanism), then continue here. Installing the skill copies these files, not dependencies. Before running the CLI, a local MCP server, or
auth commands, run
here once. Importing the published package as a dependency in your own project instead? That
already resolves everything — see
.
The connector runs on Node.js 22.18+. Pick the reference that matches how you're running it, and load it before doing anything else:
| You have... | Load |
|---|
| An MCP-aware client — tools may already be loaded (e.g. ), or you can register a local server yourself (or guide the user to) | |
| Terminal / subprocess access (you can run ) | |
| Only your own code, importing this package as a dependency | |
| No tool access, no terminal, no ability to import this package — you write your own code that calls the Google Ads API directly (e.g. a code-execution sandbox) | references/use-as-recipe.md
|
Scripts
<!-- BEGIN:skill-connections-note? -->
All scripts use the single connection
. Customer-scoped scripts take the operating account's
(digits only) and an optional
(the manager account, when access is through a manager).
<!-- END:skill-connections-note -->
<!-- BEGIN:skill-scripts-table -->
| Script | Script name | Connections | Description |
|---|
scripts/listAccessibleCustomers.ts
| | | List the accounts the connection can directly access (the account-resolution entry point). |
scripts/listCustomerClients.ts
| | | List the client (operating) accounts beneath a manager account. |
| | | Run an arbitrary GAQL query — the full read surface. |
scripts/listSearchableFields.ts
| | | List the selectable / filterable / sortable fields for a resource (compose a query). |
| | | List campaigns with status, channel type, budget, and dates. |
| | | List ad groups, optionally scoped to one campaign. |
| | | List ads, optionally scoped to one ad group. |
scripts/listConversionActions.ts
| | | List the conversion actions configured on the account. |
| | | Build a performance report: resource + metrics + segments over a date range. |
scripts/setCampaignStatus.ts
| | | Pause, enable, or remove a campaign. |
scripts/createCampaignBudget.ts
| | | Create a daily campaign budget (amount in micros). |
scripts/updateCampaignBudget.ts
| | | Update an existing budget's amount, name, or delivery method. |
scripts/createConversionAction.ts
| | | Create a conversion action (e.g. for offline tracking). |
Learn a script's input contract before calling it — never guess field names, casing, or types. Run
on a script (
./scripts/<name>.ts --help
or
node cli.js run <name> --help
); it renders the
as JSON Schema and lists the connection flag and resolvers. Guessing the payload just produces a
and wastes a round-trip.
<!-- END:skill-scripts-table -->
<!-- BEGIN:disambiguation-and-refusals? -->
Disambiguation & refusals
- Money is in micros. Budgets, bids, and report cost metrics () are 1,000,000 × the currency amount (e.g. $50.00 → ). The one exception is a conversion action's default value, which is plain currency. Don't report a of 5,000,000 as "$5,000,000".
- Act on ids, not names. Writes (, ) take ids. Resolve a name to an id first with / ; if two campaigns share a name, list them with a distinguishing field (id, status, channel type) and confirm which one before acting — never silently pick.
- Unsupported — decline, don't substitute. This connector does not upload offline conversions or add members to a Customer Match audience (Google routes new API integrations to the separate Data Manager API for those), and does not create full campaigns or manage keywords / targeting / ad creatives. If asked, say it's unsupported rather than substituting another tool and reporting success. sets up the conversion action; it does not upload conversions.
<!-- END:disambiguation-and-refusals -->
Auth
Every shape passes auth as one connection
selector, not the secret — a
string. Every connector accepts
(Zapier-managed auth — routes through Zapier's auth, retries, and governance layer); some also accept one or more direct-token resolvers (naming and count vary per connector) — check this connector's own resolvers rather than assuming. The
prefix is optional; a bare value goes to the first resolver that claims it — a UUID-shaped bare value always claims
. Each script declares the connections it needs and the resolvers each accepts. The exact syntax for passing a connection (and how to see this connector's resolver list) differs by shape — see the reference you loaded above.
Checking what's already configured first? Don't dump environment values to do it —
or
prints the value along with the name, leaking a live credential into the transcript if one is set. Check names only (
env | cut -d= -f1 | grep -i <name>
) or test a known name directly (
).
<!-- BEGIN:skill-auth-notes? operational behavior that differs by WHICH resolver is used — a safety gate only one path enforces, scopes/permissions that differ between resolvers, a billing/plan difference tied to the auth path, or a feature only available (or unavailable) on one resolver. Not for describing how to obtain or pass a credential — that's references/use-without-zapier.md's job. Leave this region empty (unfilled) if every resolver behaves identically. -->
<!-- END:skill-auth-notes -->
No connection yet? Pick one — and follow the reference's own flow to obtain it; never just ask the user for a connection id or token as if they already have one memorized:
| Load |
|---|
| Pass the credential directly | references/use-without-zapier.md
|
| Route it through a Zapier connection | references/use-with-zapier.md
|
Output format
Every script returns a
envelope:
- — the script's result (the shape its declares; see the reference you loaded above for how to inspect a script's exact schema in your shape).
meta.outputDataValidation
— what validating did:
{ skipped: false, droppedPaths: null }
— validated, nothing removed.
{ skipped: false, droppedPaths: [...], instruction }
— validated, but those paths were stripped from : fields the script returned from the API that the doesn't declare. If you need them, re-run with output validation skipped.
- — validation was bypassed; is the raw, unchecked script output.
Reading dropped fields / . To receive the raw, unvalidated result, opt out of output validation (the exact syntax differs by shape — see the reference you loaded above). Input validation is never skipped.
Trimming the result / . To shrink a large result down to the fields you need, pass a jq expression that post-processes
(again, exact syntax per shape). The jq runs against
only, NOT the
envelope, so write it rooted at
(run the script's
— or your shape's equivalent — to see its output schema). The transformed value replaces
,
is preserved, and the result is NOT re-validated against the output schema.
<!-- BEGIN:skill-references-table -->
References
Load the matching reference file before working in that area:
| Reference | Covers | Load it when |
|---|
| references/google-ads-api-gotchas.md | Auth headers, account hierarchy, GAQL, micros, mutate semantics, errors, rate limits, conversion tracking, versioning. | Before composing a GAQL query, working with money fields (micros), setting campaign status, or interpreting a Google Ads API error. |
<!-- END:skill-references-table -->