GCP Plugin

Plugin ID: gcp | Auth: OAuth2 (with PKCE) | Scopes: compute.readonly, devstorage.read_only, bigquery.readonly, iam.readonly, cloudplatformprojects.readonly

The GCP plugin uses a dedicated gcp OAuth provider (separate from the google provider used by Google Workspace) to keep cloud-platform scopes isolated.

Tools

gcp.projects.list

Name: List Projects List GCP projects accessible to the authenticated user. Input:

Parameter	Type	Required	Default	Description
`filter`	`string`	No	—	Filter expression (e.g. `"name:my-project"`)
`pageSize`	`integer`	No	`50`	Max results
`pageToken`	`string`	No	—	Pagination token

Output:

{
  "projects": [
    { "projectId": "my-project-123", "name": "My Project", "projectNumber": "123456789", "lifecycleState": "ACTIVE" }
  ],
  "nextPageToken": null
}

gcp.compute.instances.list

Name: List VM Instances List Compute Engine VM instances in a project and zone. Input:

Parameter	Type	Required	Default	Description
`project`	`string`	Yes	—	GCP project ID
`zone`	`string`	Yes	—	Compute zone (e.g. `"us-central1-a"`)
`maxResults`	`integer`	No	`50`	Max results
`pageToken`	`string`	No	—	Pagination token

Output:

{
  "instances": [
    { "id": "123", "name": "web-server-1", "status": "RUNNING", "machineType": "e2-medium", "zone": "us-central1-a" }
  ],
  "nextPageToken": null
}

gcp.compute.instances.get

Name: Get VM Instance Get details of a specific Compute Engine VM instance. Input:

Parameter	Type	Required	Description
`project`	`string`	Yes	GCP project ID
`zone`	`string`	Yes	Compute zone
`instance`	`string`	Yes	Instance name

Output:

{
  "id": "123", "name": "web-server-1", "status": "RUNNING", "machineType": "e2-medium",
  "networkInterfaces": [{}], "disks": [{}], "creationTimestamp": "2026-01-01T00:00:00Z"
}

gcp.compute.instances.start

Name: Start VM Instance | Destructive: Yes Start a stopped Compute Engine VM instance. Returns a long-running operation. Requires user approval. Input:

Parameter	Type	Required	Description
`project`	`string`	Yes	GCP project ID
`zone`	`string`	Yes	Compute zone
`instance`	`string`	Yes	Instance name

Output:

{ "operation": "op-123", "status": "RUNNING", "targetLink": "https://..." }

gcp.compute.instances.stop

Name: Stop VM Instance | Destructive: Yes Stop a running Compute Engine VM instance. Requires user approval. Input:

Parameter	Type	Required	Description
`project`	`string`	Yes	GCP project ID
`zone`	`string`	Yes	Compute zone
`instance`	`string`	Yes	Instance name

Output:

{ "operation": "op-456", "status": "RUNNING", "targetLink": "https://..." }

gcp.storage.buckets.list

Name: List Buckets List Cloud Storage buckets in a project. Input:

Parameter	Type	Required	Default	Description
`project`	`string`	Yes	—	GCP project ID
`maxResults`	`integer`	No	`50`	Max results
`pageToken`	`string`	No	—	Pagination token

Output:

{
  "buckets": [
    { "id": "my-bucket", "name": "my-bucket", "location": "US", "storageClass": "STANDARD", "timeCreated": "2026-01-01T00:00:00Z" }
  ],
  "nextPageToken": null
}

gcp.storage.objects.list

Name: List Objects List objects in a Cloud Storage bucket. Input:

Parameter	Type	Required	Default	Description
`bucket`	`string`	Yes	—	Bucket name
`prefix`	`string`	No	—	Object name prefix filter
`delimiter`	`string`	No	—	Delimiter for hierarchy (usually `"/"`)
`maxResults`	`integer`	No	`100`	Max results
`pageToken`	`string`	No	—	Pagination token

Output:

{
  "objects": [
    { "name": "data/file.csv", "size": "1024", "contentType": "text/csv", "updated": "2026-04-01T00:00:00Z" }
  ],
  "prefixes": ["data/subdir/"],
  "nextPageToken": null
}

gcp.storage.objects.get_metadata

Name: Get Object Metadata Get metadata for a Cloud Storage object (does not download the content). Input:

Parameter	Type	Required	Description
`bucket`	`string`	Yes	Bucket name
`object`	`string`	Yes	Object name (path within bucket)

Output:

{ "name": "data/file.csv", "bucket": "my-bucket", "size": "1024", "contentType": "text/csv", "md5Hash": "abc123", "updated": "2026-04-01T00:00:00Z", "metadata": null }

gcp.bigquery.datasets.list

Name: List Datasets List BigQuery datasets in a project. Input:

Parameter	Type	Required	Default	Description
`project`	`string`	Yes	—	GCP project ID
`maxResults`	`integer`	No	`50`	Max results
`pageToken`	`string`	No	—	Pagination token

Output:

{
  "datasets": [
    { "datasetId": "analytics", "projectId": "my-project", "location": "US" }
  ],
  "nextPageToken": null
}

gcp.bigquery.tables.list

Name: List Tables List tables in a BigQuery dataset. Input:

Parameter	Type	Required	Default	Description
`project`	`string`	Yes	—	GCP project ID
`datasetId`	`string`	Yes	—	BigQuery dataset ID
`maxResults`	`integer`	No	`50`	Max results
`pageToken`	`string`	No	—	Pagination token

Output:

{
  "tables": [
    { "tableId": "events", "type": "TABLE", "creationTime": "1700000000000" }
  ],
  "nextPageToken": null
}

gcp.bigquery.query

Name: Run Query Run a read-only SQL query against BigQuery. Input:

Parameter	Type	Required	Default	Description
`project`	`string`	Yes	—	GCP project ID
`query`	`string`	Yes	—	Standard SQL query
`maxResults`	`integer`	No	`100`	Max rows
`useLegacySql`	`boolean`	No	`false`	Use legacy SQL syntax

Output:

{
  "totalRows": "42",
  "schema": [{ "name": "user_id", "type": "STRING" }],
  "rows": [["user_123"]],
  "jobComplete": true
}

gcp.iam.service_accounts.list

Name: List Service Accounts List service accounts in a GCP project. Input:

Parameter	Type	Required	Default	Description
`project`	`string`	Yes	—	GCP project ID
`pageSize`	`integer`	No	`50`	Max results
`pageToken`	`string`	No	—	Pagination token

Output:

{
  "accounts": [
    { "email": "sa@my-project.iam.gserviceaccount.com", "displayName": "My SA", "disabled": false, "uniqueId": "123" }
  ],
  "nextPageToken": null
}

Getting Started

Guides

Plugins

Tools

gcp.projects.list

gcp.compute.instances.list

gcp.compute.instances.get

gcp.compute.instances.start

gcp.compute.instances.stop

gcp.storage.buckets.list

gcp.storage.objects.list

gcp.storage.objects.get_metadata

gcp.bigquery.datasets.list

gcp.bigquery.tables.list

gcp.bigquery.query

gcp.iam.service_accounts.list

​Tools

​gcp.projects.list

​gcp.compute.instances.list

​gcp.compute.instances.get

​gcp.compute.instances.start

​gcp.compute.instances.stop

​gcp.storage.buckets.list

​gcp.storage.objects.list

​gcp.storage.objects.get_metadata

​gcp.bigquery.datasets.list

​gcp.bigquery.tables.list

​gcp.bigquery.query

​gcp.iam.service_accounts.list

Tools

gcp.projects.list

gcp.compute.instances.list

gcp.compute.instances.get

gcp.compute.instances.start

gcp.compute.instances.stop

gcp.storage.buckets.list

gcp.storage.objects.list

gcp.storage.objects.get_metadata

gcp.bigquery.datasets.list

gcp.bigquery.tables.list

gcp.bigquery.query

gcp.iam.service_accounts.list