---
title: "Google Compute Engine MCP server for AI agents - actionlake"
description: "Give Claude 1,018 Google Compute Engine actions through one MCP server: instances. Your agent never holds the credential. 100,000 calls a month free."
canonical: https://actionlake.com/connectors/google-compute-engine
updated: 2026-10-22
---

Connectors · Developer & infrastructure

# Google Compute Engine MCP server for AI agents.

actionlake is an MCP gateway: one remote MCP server that connects Claude and other MCP clients to Google Compute Engine and the rest of your systems. It gives your agent 1,018 Google Compute Engine actions (466 reads, 402 writes and 150 that wait for a confirmation) and holds the Google Compute Engine credential itself, so the agent never does. It paces Google Compute Engine's limits and records every call. Connecting Google Compute Engine needs your own Google Compute Engine OAuth app and one admin sign-in.

[Connect Google Compute Engine free](https://actionlake.com/console?start=sandbox) [See pricing](https://actionlake.com/pricing)

100,000 free calls a month. No card, no email, no password: press Start a free sandbox on the next screen and keep the key it shows you.

`claude mcp add --transport http actionlake https://connect.actionlake.com/mcp`

the first call

```
do {
  action: "gcp_compute_engine.firewall_policies.list",
  intent: "check which Google Compute Engine account is connected"
}
# Firewall policies: list · read · 1 upstream call
```

- 1,018 actions in one signed file
- 466 read look, never change
- 402 write create and update
- 150 wait for a yes destructive, confirmed first

Set up

## Connect Google Compute Engine in three steps.

1. 01
   ### Add actionlake to Claude.
   Run the command once in Claude Code; its consent page can create a free sandbox: 100,000 calls a month, no card. In the Claude app, add a custom connector with the URL below and paste your key on the consent page. Creating a new sandbox from the Claude app is not yet tested by us.
   `https://connect.actionlake.com/mcp`
2. 02
   ### Connect Google Compute Engine.
   OAuth 2.0, authorization code with PKCE. You register your own OAuth app with Google Compute Engine, an admin signs in once, and you paste the app's client id and secret and the tokens from that sign-in; actionlake renews the access token from then on. actionlake stores what you paste write-only, sealed under your tenant's key.
   Before the first sign-in, Google Compute Engine needs an OAuth app of your own, registered in its developer portal. actionlake has no Google Compute Engine app to lend you.
   - `access_token` Access token from one sign-in with your own OAuth app
   - `refresh_token` Refresh token from the same sign-in; the access token is renewed from it
   - `client_id` Google OAuth client id (Google Cloud console > APIs & Services > Credentials; enable the Compute Engine API on the…
   - `client_secret` Google OAuth client secret
   ```
   account { op: "connections.requirements", service: "gcp_compute_engine" }
   # free: exactly what to paste, and what is found for you
   ```
3. 03
   ### Make the first call. Check before a write.
   Ask in plain words; Claude picks the action. Before anything changes, `check` validates the call and prices it, for free.
   ```
   do    { action: "gcp_compute_engine.firewall_policies.list" }
   # firewall policies: list · read · 1 upstream call
   check { action: "gcp_compute_engine.instances.insert", input: { … } }
   # free: inputs, permissions, cost and confirmations, nothing sent
   ```

Google Compute Engine at a glance

**Sign-in**

OAuth 2.0, authorization code with PKCE. You register your own OAuth app with Google Compute Engine, an admin signs in once, and you paste the app's client id and secret and the tokens from that sign-in; actionlake renews the access token from then on.

**You provide**

`access_token` `refresh_token` `client_id` `client_secret`

**Limits we pace to**

1,200 calls a minute, per connection

**API covered**

Google Compute Engine (VMs, disks, networks, load balancing) API compute v1 (Discovery revision 20261007)`compute.googleapis.com/compute/v1`

**Built from**

Google Compute Engine's documentation, read 22 October 2026[googleapis.com](https://www.googleapis.com/discovery/v1/apis/compute/v1/rest) [cloud.google.com](https://cloud.google.com/compute/docs/reference/rest/v1)

**Service file**

Version 1.0.0, signed (Ed25519)

**Cached reads**

454 of 466 reads, for up to 5 minutes; a write clears what it affects

**Shared entities**

1, so ask can line Google Compute Engine up with your other systems

**Data**

Processed and stored in Sydney, Australia. No US region yet

Capabilities

## What an AI agent can do in Google Compute Engine.

1,018 actions in 131 areas. Every one carries its effect, so a key can allow the reads and hold the rest.

- **466** read look, never change
- **402** write create and update
- **150** destructive wait for a confirmation

1. ### Instances
   51
   11 read · 33 write · 7 destructive
   Instances: list · Instances: insert · Instances: get · Instances: update
2. ### Instance group managers
   24
   6 read · 13 write · 5 destructive
   Instance group managers: aggregated list · Instance group managers: create instances · Instance group managers: insert · Instance group managers: list
3. ### Region instance group managers
   23
   5 read · 13 write · 5 destructive
   Region instance group managers: create instances · Region instance group managers: insert · Region instance group managers: list · Region instance group managers: list errors
4. ### Routers
   22
   12 read · 7 write · 3 destructive
   Routers: aggregated list · Routers: insert · Routers: list · Routers: list bgp routes
5. ### Network firewall policies
   21
   8 read · 9 write · 4 destructive
   Network firewall policies: aggregated list · Network firewall policies: insert · Network firewall policies: list · Network firewall policies: test iam permissions
6. ### Disks
   20
   5 read · 11 write · 4 destructive
   Disks: list · Disks: insert · Disks: get · Disks: delete
7. ### Firewall policies
   18
   7 read · 8 write · 3 destructive
   Firewall policies: insert · Firewall policies: list · Firewall policies: list associations · Firewall policies: test iam permissions
8. ### Region disks
   18
   4 read · 10 write · 4 destructive
   Region disks: bulk insert · Region disks: create snapshot · Region disks: insert · Region disks: list

**123 more areas · 821 actions**

- Region network firewall policies **18**
- Backend services **17**
- Organization security policies **16**
- Node groups **15**
- Backend buckets **14**
- Projects **14**
- Networks **13**
- Target pools **13**
- Region backend services **12**
- Security policies **12**
- Subnetworks **12**
- Images **11**
- Reservations **11**
- Target https proxies **11**
- Instance groups **10**
- Interconnect groups **10**
- Region security policies **10**
- Storage pools **10**
- Target SSL proxies **10**
- URL maps **10**
- Instant snapshots **9**
- Interconnect attachment groups **9**
- Network attachments **9**
- Network endpoint groups **9**
- Region backend buckets **9**
- Region snapshots **9**
- Resource policies **9**
- Service attachments **9**
- Snapshots **9**
- Addresses **8**
- Autoscalers **8**
- Forwarding rules **8**
- Health checks **8**
- Instance templates **8**
- Interconnects **8**
- Licenses **8**
- Machine images **8**
- Node templates **8**
- Public delegated prefixes **8**
- Region composite health checks **8**
- Region health sources **8**
- Region instant snapshots **8**
- Reservation sub blocks **8**
- Target TCP proxies **8**
- VPN gateways **8**
- Firewalls **7**
- Future reservations **7**
- Global addresses **7**
- Global forwarding rules **7**
- Global network endpoint groups **7**
- HTTP health checks **7**
- Https health checks **7**
- Instant snapshot groups **7**
- Interconnect attachments **7**
- Packet mirrorings **7**
- Public advertised prefixes **7**
- Region autoscalers **7**
- Region health aggregation policies **7**
- Region health check services **7**
- Region health checks **7**
- Region instant snapshot groups **7**
- Region network endpoint groups **7**
- Region target https proxies **7**
- Region URL maps **7**
- Rollouts **7**
- SSL policies **7**
- Target HTTP proxies **7**
- Target instances **7**
- External VPN gateways **6**
- Global VM extension policies **6**
- Region notification endpoints **6**
- Region SSL policies **6**
- Reservation blocks **6**
- Target VPN gateways **6**
- VPN tunnels **6**
- Cross site networks **5**
- Global operations **5**
- Global public delegated prefixes **5**
- Instance group manager resize requests **5**
- Network edge security services **5**
- Region commitments **5**
- Region instance group manager resize requests **5**
- Region instance groups **5**
- Region target HTTP proxies **5**
- Reservation slots **5**
- Routes **5**
- SSL certificates **5**
- Target GRPC proxies **5**
- Wire groups **5**
- Zone VM extension policies **5**
- License codes **4**
- Region instance templates **4**
- Region operations **4**
- Region SSL certificates **4**
- Region target TCP proxies **4**
- Rollout plans **4**
- Zone operations **4**
- Accelerator types **3**
- Advice **3**
- Disk types **3**
- Global organization operations **3**
- Hosts **3**
- Machine types **3**
- Node types **3**
- Preview features **3**
- Storage pool types **3**
- Global frontend settings **2**
- Image views **2**
- Instance settings **2**
- Interconnect locations **2**
- Interconnect remote locations **2**
- Managed rulesets **2**
- Network profiles **2**
- Region disk types **2**
- Region snapshot settings **2**
- Regions **2**
- Reliability risks **2**
- Snapshot settings **2**
- Zones **2**
- Image family views **1**
- Project views **1**
- Region instances **1**
- Region zones **1**

Top actions

## The Google Compute Engine API, shaped for agents.

The ids your agent types, what each does, and the one name it shares with other systems. 49 of Google Compute Engine's 56 action names are shared with at least one other connector, so an agent learns them once.

| Action | What it does | Effect | Same name in |
| --- | --- | --- | --- |
| `gcp_compute_engine.instances.list` Instances: list | Retrieves the list of instances contained within the specified zone. | read | `infra.instance.list` 18 connectors |
| `gcp_compute_engine.instances.insert` Instances: insert | Creates an instance resource in the specified project using the data included in the request. | write | `infra.instance.create` 10 connectors |
| `gcp_compute_engine.instances.delete` Instances: delete | Deletes the specified Instance resource. | destructive | `infra.instance.delete` 13 connectors |
| `gcp_compute_engine.regions.list` Regions: list | Retrieves the list of region resources available to the specified project. | read | `infra.region.list` 16 connectors |
| `gcp_compute_engine.snapshots.list` Snapshots: list | Retrieves the list of Snapshot resources contained within the specified project. | read | `infra.snapshot.list` 12 connectors |
| `gcp_compute_engine.disks.list` Disks: list | Retrieves a list of persistent disks contained within the specified zone. | read | `infra.volume.list` 10 connectors |
| `gcp_compute_engine.addresses.list` Addresses: list | Retrieves a list of addresses contained within the specified region. | read | `infra.ip_address.list` 9 connectors |
| `gcp_compute_engine.snapshots.insert` Snapshots: insert | Creates a snapshot in the specified project using the data included in the request. | write | `infra.snapshot.create` 9 connectors |
| `gcp_compute_engine.disks.insert` Disks: insert | Creates a persistent disk in the specified project using the data in the request. | write | `infra.volume.create` 8 connectors |
| `gcp_compute_engine.networks.list` Networks: list | Retrieves the list of networks available to the specified project. | read | `infra.network.list` 6 connectors |
| `gcp_compute_engine.addresses.insert` Addresses: insert | Creates an address resource in the specified project by using the data included in the request. | write | `infra.ip_address.create` 6 connectors |
| `gcp_compute_engine.networks.insert` Networks: insert | Creates a network in the specified project using the data included in the request. | write | `infra.network.create` 5 connectors |

Try it

## Prompts to paste into Claude.

Each one names the Google Compute Engine actions it will call, so you know what it reads and what it changes before you send it.

1. Read read
   Firewall policies: list in Google Compute Engine and tell me what stands out.
   calls `gcp_compute_engine.firewall_policies.list`
2. Write write
   Instances: insert in Google Compute Engine. Show me exactly what you will send before you save it.
   calls `gcp_compute_engine.instances.insert`

Governance for Google Compute Engine

## Destructive actions wait for a yes.

150 Google Compute Engine actions are destructive. The first 10:

- Instances: delete `gcp_compute_engine.instances.delete`
- Disks: delete `gcp_compute_engine.disks.delete`
- Snapshots: delete `gcp_compute_engine.snapshots.delete`
- Networks: delete `gcp_compute_engine.networks.delete`
- Addresses: delete `gcp_compute_engine.addresses.delete`
- Images: delete `gcp_compute_engine.images.delete`
- Subnetworks: delete `gcp_compute_engine.subnetworks.delete`
- Firewalls: delete `gcp_compute_engine.firewalls.delete`
- Reservations: delete `gcp_compute_engine.reservations.delete`
- Instance templates: delete `gcp_compute_engine.instance_templates.delete`

A destructive action runs only with `confirm: true`. Under the `human` policy the agent cannot confirm for itself: the run stops until a named person approves, in the console or from a signed link in an email. The cap on a key is refused before the call is sent, never after.

do gcp_compute_engine.instances.delete confirm_policy: **human**

Instances: delete?

**intent:** the user asked me to instances: delete

run suspended · waiting for a named person · key _gcp-compute-engine-agent_ cannot confirm itself

In the console, or from an emailed signed link.

a key for this connector

```
account {
  op: "keys.create", name: "gcp-compute-engine-agent",
  allowed_actions: ["gcp_compute_engine.*"], cap_calls: 2000,
  confirm_policy: "human_for_writes",
  approval_rules: [
    { match: "gcp_compute_engine.instances.delete", approver: "lead@example.com" },
    { match: "gcp_compute_engine.disks.delete", approver: "lead@example.com" },
    { match: "gcp_compute_engine.snapshots.delete", approver: "lead@example.com" }
  ]
}
```

Illustrative. Scoped to Google Compute Engine only, capped at 2,000 upstream calls, every write confirmed, and the destructive ones approved by a named person.

Cost

## What Google Compute Engine costs on actionlake.

Every Google Compute Engine action is one upstream call, the only thing actionlake bills. Searching, `check`, `cost`, `account` and any call Google Compute Engine refused with 401, 403, 429 or a 5xx are free. Paid plans start at US$49 a month for 250,000 calls, with overage from US$0.30 down to US$0.18 per 1,000.

- 1 upstream call per action all 1,018 Google Compute Engine actions
- 148 actions page each page of a long list is one call
- 454 reads can be cached a cache hit costs nothing
- 100,000 calls a month, free paid plans from US$49 a month for 250,000

[See pricing](https://actionlake.com/pricing)

Action index

## Every Google Compute Engine action.

1,018 actions by area: the id your agent calls, what it does, and its effect.

**Instances 51**

- `gcp_compute_engine.instances.get` Instances: get read
- `gcp_compute_engine.instances.insert` Instances: insert write
- `gcp_compute_engine.instances.list` Instances: list read
- `gcp_compute_engine.instances.update` Instances: update write
- `gcp_compute_engine.instances.add_access_config` Instances: add access config write
- `gcp_compute_engine.instances.add_network_interface` Instances: add network interface write
- and 45 more in this area

**Instance group managers 24**

- `gcp_compute_engine.instance_group_managers.aggregated_list` Instance group managers: aggregated list read
- `gcp_compute_engine.instance_group_managers.create_instances` Instance group managers: create instances write
- `gcp_compute_engine.instance_group_managers.insert` Instance group managers: insert write
- `gcp_compute_engine.instance_group_managers.list` Instance group managers: list read
- `gcp_compute_engine.instance_group_managers.abandon_instances` Instance group managers: abandon instances destructive
- `gcp_compute_engine.instance_group_managers.apply_updates_to_instances` Instance group managers: apply updates to instances write
- and 18 more in this area

**Region instance group managers 23**

- `gcp_compute_engine.region_instance_group_managers.create_instances` Region instance group managers: create instances write
- `gcp_compute_engine.region_instance_group_managers.insert` Region instance group managers: insert write
- `gcp_compute_engine.region_instance_group_managers.list` Region instance group managers: list read
- `gcp_compute_engine.region_instance_group_managers.list_errors` Region instance group managers: list errors read
- `gcp_compute_engine.region_instance_group_managers.abandon_instances` Region instance group managers: abandon instances destructive
- `gcp_compute_engine.region_instance_group_managers.apply_updates_to_instances` Region instance group managers: apply updates to instances write
- and 17 more in this area

**Routers 22**

- `gcp_compute_engine.routers.aggregated_list` Routers: aggregated list read
- `gcp_compute_engine.routers.insert` Routers: insert write
- `gcp_compute_engine.routers.list` Routers: list read
- `gcp_compute_engine.routers.list_bgp_routes` Routers: list bgp routes read
- `gcp_compute_engine.routers.delete` Routers: delete destructive
- `gcp_compute_engine.routers.delete_named_set` Routers: delete named set destructive
- and 16 more in this area

**Network firewall policies 21**

- `gcp_compute_engine.network_firewall_policies.aggregated_list` Network firewall policies: aggregated list read
- `gcp_compute_engine.network_firewall_policies.insert` Network firewall policies: insert write
- `gcp_compute_engine.network_firewall_policies.list` Network firewall policies: list read
- `gcp_compute_engine.network_firewall_policies.test_iam_permissions` Network firewall policies: test iam permissions read
- `gcp_compute_engine.network_firewall_policies.add_association` Network firewall policies: add association write
- `gcp_compute_engine.network_firewall_policies.add_packet_mirroring_rule` Network firewall policies: add packet mirroring rule write
- and 15 more in this area

**Disks 20**

- `gcp_compute_engine.disks.delete` Disks: delete destructive
- `gcp_compute_engine.disks.get` Disks: get read
- `gcp_compute_engine.disks.insert` Disks: insert write
- `gcp_compute_engine.disks.list` Disks: list read
- `gcp_compute_engine.disks.add_resource_policies` Disks: add resource policies write
- `gcp_compute_engine.disks.aggregated_list` Disks: aggregated list read
- and 14 more in this area

**Firewall policies 18**

- `gcp_compute_engine.firewall_policies.insert` Firewall policies: insert write
- `gcp_compute_engine.firewall_policies.list` Firewall policies: list read
- `gcp_compute_engine.firewall_policies.list_associations` Firewall policies: list associations read
- `gcp_compute_engine.firewall_policies.test_iam_permissions` Firewall policies: test iam permissions read
- `gcp_compute_engine.firewall_policies.add_association` Firewall policies: add association write
- `gcp_compute_engine.firewall_policies.add_rule` Firewall policies: add rule write
- and 12 more in this area

**Region disks 18**

- `gcp_compute_engine.region_disks.bulk_insert` Region disks: bulk insert write
- `gcp_compute_engine.region_disks.create_snapshot` Region disks: create snapshot write
- `gcp_compute_engine.region_disks.insert` Region disks: insert write
- `gcp_compute_engine.region_disks.list` Region disks: list read
- `gcp_compute_engine.region_disks.add_resource_policies` Region disks: add resource policies write
- `gcp_compute_engine.region_disks.delete` Region disks: delete destructive
- and 12 more in this area

**Region network firewall policies 18**

- `gcp_compute_engine.region_network_firewall_policies.add_association` Region network firewall policies: add association write
- `gcp_compute_engine.region_network_firewall_policies.insert` Region network firewall policies: insert write
- `gcp_compute_engine.region_network_firewall_policies.list` Region network firewall policies: list read
- `gcp_compute_engine.region_network_firewall_policies.test_iam_permissions` Region network firewall policies: test iam permissions read
- `gcp_compute_engine.region_network_firewall_policies.add_rule` Region network firewall policies: add rule write
- `gcp_compute_engine.region_network_firewall_policies.clone_rules` Region network firewall policies: clone rules write
- and 12 more in this area

**Backend services 17**

- `gcp_compute_engine.backend_services.aggregated_list` Backend services: aggregated list read
- `gcp_compute_engine.backend_services.insert` Backend services: insert write
- `gcp_compute_engine.backend_services.list` Backend services: list read
- `gcp_compute_engine.backend_services.list_usable` Backend services: list usable read
- `gcp_compute_engine.backend_services.add_signed_url_key` Backend services: add signed url key write
- `gcp_compute_engine.backend_services.delete` Backend services: delete destructive
- and 11 more in this area

**Organization security policies 16**

- `gcp_compute_engine.organization_security_policies.insert` Organization security policies: insert write
- `gcp_compute_engine.organization_security_policies.list` Organization security policies: list read
- `gcp_compute_engine.organization_security_policies.list_associations` Organization security policies: list associations read
- `gcp_compute_engine.organization_security_policies.list_preconfigured_expression_sets` Organization security policies: list preconfigured expression sets read
- `gcp_compute_engine.organization_security_policies.add_association` Organization security policies: add association write
- `gcp_compute_engine.organization_security_policies.add_rule` Organization security policies: add rule write
- and 10 more in this area

**Node groups 15**

- `gcp_compute_engine.node_groups.aggregated_list` Node groups: aggregated list read
- `gcp_compute_engine.node_groups.insert` Node groups: insert write
- `gcp_compute_engine.node_groups.list` Node groups: list read
- `gcp_compute_engine.node_groups.list_nodes` Node groups: list nodes read
- `gcp_compute_engine.node_groups.add_nodes` Node groups: add nodes write
- `gcp_compute_engine.node_groups.delete` Node groups: delete destructive
- and 9 more in this area

**Backend buckets 14**

- `gcp_compute_engine.backend_buckets.aggregated_list` Backend buckets: aggregated list read
- `gcp_compute_engine.backend_buckets.insert` Backend buckets: insert write
- `gcp_compute_engine.backend_buckets.list` Backend buckets: list read
- `gcp_compute_engine.backend_buckets.list_usable` Backend buckets: list usable read
- `gcp_compute_engine.backend_buckets.add_signed_url_key` Backend buckets: add signed url key write
- `gcp_compute_engine.backend_buckets.delete` Backend buckets: delete destructive
- and 8 more in this area

**Projects 14**

- `gcp_compute_engine.projects.disable_xpn_host` Projects: disable xpn host write
- `gcp_compute_engine.projects.disable_xpn_resource` Projects: disable xpn resource write
- `gcp_compute_engine.projects.enable_xpn_host` Projects: enable xpn host write
- `gcp_compute_engine.projects.list_xpn_hosts` Projects: list xpn hosts read
- `gcp_compute_engine.projects.enable_xpn_resource` Projects: enable xpn resource write
- `gcp_compute_engine.projects.get` Projects: get read
- and 8 more in this area

**Networks 13**

- `gcp_compute_engine.networks.get` Networks: get read
- `gcp_compute_engine.networks.insert` Networks: insert write
- `gcp_compute_engine.networks.list` Networks: list read
- `gcp_compute_engine.networks.patch` Networks: patch write
- `gcp_compute_engine.networks.add_peering` Networks: add peering write
- `gcp_compute_engine.networks.cancel_request_remove_peering` Networks: cancel request remove peering write
- and 7 more in this area

**Target pools 13**

- `gcp_compute_engine.target_pools.aggregated_list` Target pools: aggregated list read
- `gcp_compute_engine.target_pools.insert` Target pools: insert write
- `gcp_compute_engine.target_pools.list` Target pools: list read
- `gcp_compute_engine.target_pools.test_iam_permissions` Target pools: test iam permissions read
- `gcp_compute_engine.target_pools.add_health_check` Target pools: add health check write
- `gcp_compute_engine.target_pools.add_instance` Target pools: add instance write
- and 7 more in this area

**Region backend services 12**

- `gcp_compute_engine.region_backend_services.insert` Region backend services: insert write
- `gcp_compute_engine.region_backend_services.list` Region backend services: list read
- `gcp_compute_engine.region_backend_services.list_usable` Region backend services: list usable read
- `gcp_compute_engine.region_backend_services.test_iam_permissions` Region backend services: test iam permissions read
- `gcp_compute_engine.region_backend_services.delete` Region backend services: delete destructive
- `gcp_compute_engine.region_backend_services.get` Region backend services: get read
- and 6 more in this area

**Security policies 12**

- `gcp_compute_engine.security_policies.aggregated_list` Security policies: aggregated list read
- `gcp_compute_engine.security_policies.insert` Security policies: insert write
- `gcp_compute_engine.security_policies.list` Security policies: list read
- `gcp_compute_engine.security_policies.list_preconfigured_expression_sets` Security policies: list preconfigured expression sets read
- `gcp_compute_engine.security_policies.add_rule` Security policies: add rule write
- `gcp_compute_engine.security_policies.delete` Security policies: delete destructive
- and 6 more in this area

**Subnetworks 12**

- `gcp_compute_engine.subnetworks.get` Subnetworks: get read
- `gcp_compute_engine.subnetworks.insert` Subnetworks: insert write
- `gcp_compute_engine.subnetworks.list` Subnetworks: list read
- `gcp_compute_engine.subnetworks.patch` Subnetworks: patch write
- `gcp_compute_engine.subnetworks.aggregated_list` Subnetworks: aggregated list read
- `gcp_compute_engine.subnetworks.delete` Subnetworks: delete destructive
- and 6 more in this area

**Images 11**

- `gcp_compute_engine.images.get` Images: get read
- `gcp_compute_engine.images.insert` Images: insert write
- `gcp_compute_engine.images.list` Images: list read
- `gcp_compute_engine.images.patch` Images: patch write
- `gcp_compute_engine.images.delete` Images: delete destructive
- `gcp_compute_engine.images.deprecate` Images: deprecate write
- and 5 more in this area

**Reservations 11**

- `gcp_compute_engine.reservations.get` Reservations: get read
- `gcp_compute_engine.reservations.insert` Reservations: insert write
- `gcp_compute_engine.reservations.list` Reservations: list read
- `gcp_compute_engine.reservations.update` Reservations: update write
- `gcp_compute_engine.reservations.aggregated_list` Reservations: aggregated list read
- `gcp_compute_engine.reservations.delete` Reservations: delete destructive
- and 5 more in this area

**Target https proxies 11**

- `gcp_compute_engine.target_https_proxies.aggregated_list` Target https proxies: aggregated list read
- `gcp_compute_engine.target_https_proxies.get` Target https proxies: get read
- `gcp_compute_engine.target_https_proxies.insert` Target https proxies: insert write
- `gcp_compute_engine.target_https_proxies.list` Target https proxies: list read
- `gcp_compute_engine.target_https_proxies.delete` Target https proxies: delete destructive
- `gcp_compute_engine.target_https_proxies.patch` Target https proxies: patch write
- and 5 more in this area

**Instance groups 10**

- `gcp_compute_engine.instance_groups.aggregated_list` Instance groups: aggregated list read
- `gcp_compute_engine.instance_groups.insert` Instance groups: insert write
- `gcp_compute_engine.instance_groups.list` Instance groups: list read
- `gcp_compute_engine.instance_groups.list_instances` Instance groups: list instances read
- `gcp_compute_engine.instance_groups.add_instances` Instance groups: add instances write
- `gcp_compute_engine.instance_groups.delete` Instance groups: delete destructive
- and 4 more in this area

**Interconnect groups 10**

- `gcp_compute_engine.interconnect_groups.create_members` Interconnect groups: create members write
- `gcp_compute_engine.interconnect_groups.insert` Interconnect groups: insert write
- `gcp_compute_engine.interconnect_groups.list` Interconnect groups: list read
- `gcp_compute_engine.interconnect_groups.test_iam_permissions` Interconnect groups: test iam permissions read
- `gcp_compute_engine.interconnect_groups.delete` Interconnect groups: delete destructive
- `gcp_compute_engine.interconnect_groups.get` Interconnect groups: get read
- and 4 more in this area

**Region security policies 10**

- `gcp_compute_engine.region_security_policies.add_rule` Region security policies: add rule write
- `gcp_compute_engine.region_security_policies.get` Region security policies: get read
- `gcp_compute_engine.region_security_policies.insert` Region security policies: insert write
- `gcp_compute_engine.region_security_policies.list` Region security policies: list read
- `gcp_compute_engine.region_security_policies.delete` Region security policies: delete destructive
- `gcp_compute_engine.region_security_policies.get_rule` Region security policies: get rule read
- and 4 more in this area

**Storage pools 10**

- `gcp_compute_engine.storage_pools.aggregated_list` Storage pools: aggregated list read
- `gcp_compute_engine.storage_pools.insert` Storage pools: insert write
- `gcp_compute_engine.storage_pools.list` Storage pools: list read
- `gcp_compute_engine.storage_pools.list_disks` Storage pools: list disks read
- `gcp_compute_engine.storage_pools.delete` Storage pools: delete destructive
- `gcp_compute_engine.storage_pools.get` Storage pools: get read
- and 4 more in this area

**Target SSL proxies 10**

- `gcp_compute_engine.target_ssl_proxies.get` Target ssl proxies: get read
- `gcp_compute_engine.target_ssl_proxies.insert` Target ssl proxies: insert write
- `gcp_compute_engine.target_ssl_proxies.list` Target ssl proxies: list read
- `gcp_compute_engine.target_ssl_proxies.test_iam_permissions` Target ssl proxies: test iam permissions read
- `gcp_compute_engine.target_ssl_proxies.delete` Target ssl proxies: delete destructive
- `gcp_compute_engine.target_ssl_proxies.set_backend_service` Target ssl proxies: set backend service write
- and 4 more in this area

**URL maps 10**

- `gcp_compute_engine.url_maps.aggregated_list` Url maps: aggregated list read
- `gcp_compute_engine.url_maps.insert` Url maps: insert write
- `gcp_compute_engine.url_maps.list` Url maps: list read
- `gcp_compute_engine.url_maps.test_iam_permissions` Url maps: test iam permissions read
- `gcp_compute_engine.url_maps.delete` Url maps: delete destructive
- `gcp_compute_engine.url_maps.get` Url maps: get read
- and 4 more in this area

**Instant snapshots 9**

- `gcp_compute_engine.instant_snapshots.aggregated_list` Instant snapshots: aggregated list read
- `gcp_compute_engine.instant_snapshots.insert` Instant snapshots: insert write
- `gcp_compute_engine.instant_snapshots.list` Instant snapshots: list read
- `gcp_compute_engine.instant_snapshots.test_iam_permissions` Instant snapshots: test iam permissions read
- `gcp_compute_engine.instant_snapshots.delete` Instant snapshots: delete destructive
- `gcp_compute_engine.instant_snapshots.get` Instant snapshots: get read
- and 3 more in this area

**Interconnect attachment groups 9**

- `gcp_compute_engine.interconnect_attachment_groups.get` Interconnect attachment groups: get read
- `gcp_compute_engine.interconnect_attachment_groups.insert` Interconnect attachment groups: insert write
- `gcp_compute_engine.interconnect_attachment_groups.list` Interconnect attachment groups: list read
- `gcp_compute_engine.interconnect_attachment_groups.test_iam_permissions` Interconnect attachment groups: test iam permissions read
- `gcp_compute_engine.interconnect_attachment_groups.delete` Interconnect attachment groups: delete destructive
- `gcp_compute_engine.interconnect_attachment_groups.get_iam_policy` Interconnect attachment groups: get iam policy read
- and 3 more in this area

Network attachments **9** Network endpoint groups **9** Region backend buckets **9** Region snapshots **9** Resource policies **9** Service attachments **9** Snapshots **9** Addresses **8** Autoscalers **8** Forwarding rules **8** Health checks **8** Instance templates **8** Interconnects **8** Licenses **8** Machine images **8** Node templates **8** Public delegated prefixes **8** Region composite health checks **8** Region health sources **8** Region instant snapshots **8** Reservation sub blocks **8** Target TCP proxies **8** VPN gateways **8** Firewalls **7** Future reservations **7** Global addresses **7** Global forwarding rules **7** Global network endpoint groups **7** HTTP health checks **7** Https health checks **7** Instant snapshot groups **7** Interconnect attachments **7** Packet mirrorings **7** Public advertised prefixes **7** Region autoscalers **7** Region health aggregation policies **7** Region health check services **7** Region health checks **7** Region instant snapshot groups **7** Region network endpoint groups **7** Region target https proxies **7** Region URL maps **7** Rollouts **7** SSL policies **7** Target HTTP proxies **7** Target instances **7** External VPN gateways **6** Global VM extension policies **6** Region notification endpoints **6** Region SSL policies **6** Reservation blocks **6** Target VPN gateways **6** VPN tunnels **6** Cross site networks **5** Global operations **5** Global public delegated prefixes **5** Instance group manager resize requests **5** Network edge security services **5** Region commitments **5** Region instance group manager resize requests **5** Region instance groups **5** Region target HTTP proxies **5** Reservation slots **5** Routes **5** SSL certificates **5** Target GRPC proxies **5** Wire groups **5** Zone VM extension policies **5** License codes **4** Region instance templates **4** Region operations **4** Region SSL certificates **4** Region target TCP proxies **4** Rollout plans **4** Zone operations **4** Accelerator types **3** Advice **3** Disk types **3** Global organization operations **3** Hosts **3** Machine types **3** Node types **3** Preview features **3** Storage pool types **3** Global frontend settings **2** Image views **2** Instance settings **2** Interconnect locations **2** Interconnect remote locations **2** Managed rulesets **2** Network profiles **2** Region disk types **2** Region snapshot settings **2** Regions **2** Reliability risks **2** Snapshot settings **2** Zones **2** Image family views **1** Project views **1** Region instances **1** Region zones **1**

This index shows a sample of each area. All 1,018 actions, with their inputs and prices, are in the [catalogue](https://actionlake.com/console/catalogue), which a free sandbox key can search.

Questions

## The Google Compute Engine MCP server, answered.

**Is there a Google Compute Engine MCP server for Claude?**

Yes. actionlake is an MCP gateway: a remote MCP server that gives Claude, or another MCP client, 1,018 Google Compute Engine actions across instances, instance group managers, region instance group managers and routers, through the same eight tools it uses for every other connector. actionlake is independent; it is not made or endorsed by Google Compute Engine.

**What do I need to connect Google Compute Engine?**

Your own Google Compute Engine OAuth app, registered with Google Compute Engine (its client id and client secret), and the access and refresh tokens from one admin sign-in. actionlake has no Google Compute Engine app of its own, so this is not a one-step consent. actionlake renews the access token as it expires. The free call account { op: "connections.requirements", service: "gcp_compute_engine" } lists exactly what to paste.

**Can an AI agent delete or change things in Google Compute Engine?**

Only if its key allows it, and only with an explicit confirmation. 150 Google Compute Engine actions are destructive, including instances: delete, disks: delete and snapshots: delete. Each runs only with confirm: true. Under the human confirm policy, or an approval rule on the key, the run waits until a named person approves it.

**Will an AI agent use up my Google Compute Engine API limit?**

actionlake paces every call to the limits in the Google Compute Engine service file: 1,200 calls a minute per connection. Repeated reads are answered from cache (454 Google Compute Engine actions can be), which never reaches Google Compute Engine.

**What does a Google Compute Engine action cost?**

One upstream call, the only unit actionlake bills. A list that pages bills one call per page (148 Google Compute Engine actions page). Cache hits, searches, dry runs and calls Google Compute Engine refused with 401, 403, 429 or a 5xx are free. The Free plan includes 100,000 calls a month with no card. Paid plans start at US$49 a month for 250,000 calls, with overage from US$0.30 down to US$0.18 per 1,000.

**Which Google Compute Engine API does it cover?**

The Google Compute Engine (VMs, disks, networks, load balancing) API compute v1 (Discovery revision 20261007) at compute.googleapis.com/compute/v1. It is built from Google Compute Engine's documentation as read on 22 October 2026. 1,018 actions in 131 areas, each one in the signed service file (version 1.0.0).

**Does actionlake keep my Google Compute Engine data, and where?**

actionlake processes and stores data in Sydney, Australia; there is no US region yet. Reads may be cached for up to 5 minutes, and a write clears the reads it affects. Each run's request and response is kept, sealed, for your plan's trace window: 1 day on Free, 3 days on Starter, 7 days on Pro, 14 days on Team, 30 days on Business and 30 days on Enterprise. Your Google Compute Engine credential is sealed under your tenant's key, never logged and never returned. Your traffic is not used to train models.

**Which AI clients can use it?**

Claude Code, tested by us end to end, and the Claude app as a custom connector, tested by us with an existing key. Other MCP clients that support remote servers over Streamable HTTP (MCP revisions 2025-06-18 and 2025-11-25) and OAuth dynamic client registration, or a bearer header, should connect the same way; we have not tested them. Or plain REST, under the same action names.

## More connectors like Google Compute Engine

Shares the most action names with

- [Fly.io Machines 19](https://actionlake.com/connectors/fly-io-machines)
- [Hetzner Cloud 15](https://actionlake.com/connectors/hetzner-cloud)
- [Akamai Cloud (Linode) 8](https://actionlake.com/connectors/akamai-cloud-linode)
- [Render 8](https://actionlake.com/connectors/render)
- [Coolify 5](https://actionlake.com/connectors/coolify)
- [Google Cloud SQL Admin 4](https://actionlake.com/connectors/google-cloud-sql-admin)

Joins with, on shared entities

- [NetBox 1](https://actionlake.com/connectors/netbox)
- [DigitalOcean 1](https://actionlake.com/connectors/digitalocean)
- [Vultr 1](https://actionlake.com/connectors/vultr)
- [Hostinger 1](https://actionlake.com/connectors/hostinger)
- [Azure Compute (Virtual Machines) 1](https://actionlake.com/connectors/azure-compute-virtual-machines)
- [SUSE Rancher Manager 1](https://actionlake.com/connectors/suse-rancher-manager)

Browse

- [All Developer & infrastructure connectors](https://actionlake.com/connectors/category/developer-and-infrastructure)
- [All connectors](https://actionlake.com/connectors)
- [How actionlake compares](https://actionlake.com/compare)

actionlake is independent. It is not made or endorsed by Google Compute Engine; the name is the vendor's own. Counts are from the Google Compute Engine service file, version 1.0.0.

## Give your agent Google Compute Engine today.

[Connect Google Compute Engine free](https://actionlake.com/console?start=sandbox) [See pricing](https://actionlake.com/pricing)

Start free - 100,000 calls/month, no card. No signup. Hard caps on every key. Cached answers are free.

---

Source: https://actionlake.com/connectors/google-compute-engine
More: https://actionlake.com/llms.txt