Connectors · AI

Google Gemini API (AI Studio) MCP server for AI agents.

actionlake is an MCP gateway: one remote MCP server that connects Claude and other MCP clients to Google Gemini API (AI Studio) and the rest of your systems. It gives your agent 83 Google Gemini API (AI Studio) actions (39 reads, 32 writes and 12 that wait for a confirmation) and holds the Google Gemini API (AI Studio) credential itself, so the agent never does. It paces Google Gemini API (AI Studio)'s limits and records every call.

Connect Google Gemini API (AI Studio) free See 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

83actionsin one signed file
39readlook, never change
32writecreate and update
12wait for a yesdestructive, confirmed first

Set up

Connect Google Gemini API (AI Studio) in three steps.

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
02
Connect Google Gemini API (AI Studio).
API key. No consent screen: your agent can connect it for you, but the key then passes through your chat. To keep it out of the model's context, paste it in the console instead. actionlake stores what you paste write-only, sealed under your tenant's key.
- api_keyGemini API key from Google AI Studio (sent as x-goog-api-key)
```
account { op: "connections.requirements", service: "google_gemini" }
# free: exactly what to paste, and what is found for you
```

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: "google_gemini.files.list" }# list files · read · 1 upstream callcheck { action: "google_gemini.models.predict", input: { … } }# free: inputs, permissions, cost and confirmations, nothing sent

Google Gemini API (AI Studio) at a glance

Sign-in: API key. No consent screen: your agent can connect it for you, but the key then passes through your chat. To keep it out of the model's context, paste it in the console instead.
You provide: api_key
Limits we pace to: at most 16 at a time, per connection
API covered: Google Gemini API (AI Studio) API v1beta (discovery revision 20261020)generativelanguage.googleapis.com
Built from: Google Gemini API (AI Studio)'s documentation, read 22 October 2026generativelanguage.googleapis.com
Service file: Version 1.0.0, signed (Ed25519)
Cached reads: 37 of 39 reads, for up to an hour; a write clears what it affects
Shared entities: 1, so ask can line Google Gemini API (AI Studio) 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 Gemini API (AI Studio).

83 actions in 21 areas. Every one carries its effect, so a key can allow the reads and hold the rest.

39readlook, never change
32writecreate and update
12destructivewait for a confirmation

Models
18
9 read · 9 write · 0 destructive
Generate Answer models · Generate Content models · Generate Message models · Generate Text models
Tuned models
11
2 read · 7 write · 2 destructive
Generate Content tuned models · Generate Text tuned models · Stream Generate Content tuned models · Async Batch Embed Content tuned models
Batches
6
2 read · 2 write · 2 destructive
List batches · Get batches · Cancel batches · Delete batches
Cached contents
5
2 read · 2 write · 1 destructive
Create cached contents · List cached contents · Get cached contents · Update cached contents
Corpus permissions
5
2 read · 2 write · 1 destructive
Create corpus permissions · List corpus permissions · Get corpus permissions · Update corpus permissions
File search stores
5
2 read · 2 write · 1 destructive
List file search stores · Create file search stores · Get file search stores · Delete file search stores
Tuned model permissions
5
2 read · 2 write · 1 destructive
Create tuned model permissions · List tuned model permissions · Get tuned model permissions · Update tuned model permissions
Corpora
4
2 read · 1 write · 1 destructive
List corpora · Create corpora · Get corpora · Delete corpora

13 more areas · 24 actions

Environments 4
Files 4
File search store documents 3
Dynamic 2
Model operations 2
Tuned model operations 2
Auth tokens 1
Corpus operations 1
Environment files 1
File search store operations 1
File search store upload operations 1
Generated file operations 1
Generated files 1

Top actions

The Google Gemini API (AI Studio) API, shaped for agents.

The ids your agent types, what each does, and the one name it shares with other systems. 25 of Google Gemini API (AI Studio)'s 25 action names are shared with at least one other connector, so an agent learns them once.

Action	What it does	Effect	Same name in
`google_gemini.files.list`List files	Lists the metadata for `File`s owned by the requesting project.	read	`content.file.list`77 connectors
`google_gemini.files.get`Get files	Gets the metadata for the given `File`.	read	`content.file.get`74 connectors
`google_gemini.models.predict`Predict models	Performs a prediction request.	write	`analytics.inference.create`17 connectors
`google_gemini.models.generate_text`Generate Text models	Generates a response from the model given an input message.	write	`analytics.inference.create`17 connectors
`google_gemini.files.delete`Delete files	Deletes the `File`.	destructive	`content.file.delete`68 connectors
`google_gemini.batches.list`List batches	Lists operations that match the specified filter in the request.	read	`automation.background_job.list`34 connectors
`google_gemini.model_operations.list`List model operations	Lists operations that match the specified filter in the request.	read	`automation.background_job.list`34 connectors
`google_gemini.auth_tokens.create`Create auth tokens	Creates a token that can be used to constrain the behavior of a BidiGenerateContent session.	write	`identity.api_key.create`100 connectors
`google_gemini.models.get`Get models	Gets information about a specific `Model` such as its version number, token limits,…	read	`infra.ml_model.get`13 connectors
`google_gemini.tuned_models.get`Get tuned models	Gets information about a specific TunedModel.	read	`infra.ml_model.get`13 connectors
`google_gemini.corpora.list`List corpora	Lists all `Corpora` owned by the user.	read	`infra.search_index.list`5 connectors
`google_gemini.file_search_stores.list`List file search stores	Lists all `FileSearchStores` owned by the user.	read	`infra.search_index.list`5 connectors

Try it

Prompts to paste into Claude.

Each one names the Google Gemini API (AI Studio) actions it will call, so you know what it reads and what it changes before you send it.

Readread
Which Google Gemini API (AI Studio) files changed this week, and who changed them?
callsgoogle_gemini.files.list
Writewrite
Predict models in Google Gemini API (AI Studio). Show me exactly what you will send before you save it.
callsgoogle_gemini.models.predict

Governance for Google Gemini API (AI Studio)

Destructive actions wait for a yes.

12 Google Gemini API (AI Studio) actions are destructive. These are all of them:

Delete filesgoogle_gemini.files.delete
Cancel batchesgoogle_gemini.batches.cancel
Delete batchesgoogle_gemini.batches.delete
Delete tuned modelsgoogle_gemini.tuned_models.delete
Delete corporagoogle_gemini.corpora.delete
Delete file search storesgoogle_gemini.file_search_stores.delete
Delete environmentsgoogle_gemini.environments.delete
Transfer Ownership tuned modelsgoogle_gemini.tuned_models.transfer_ownership
Delete cached contentsgoogle_gemini.cached_contents.delete
Delete corpus permissionsgoogle_gemini.corpus_permissions.delete
Delete tuned model permissionsgoogle_gemini.tuned_model_permissions.delete
Delete file search store documentsgoogle_gemini.file_search_store_documents.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.

dogoogle_gemini.files.deleteconfirm_policy: human

Delete files?

intent: the user asked me to delete files

run suspended · waiting for a named person · key google-gemini-agent cannot confirm itself

In the console, or from an emailed signed link.

a key for this connector

account {  op: "keys.create", name: "google-gemini-agent",  allowed_actions: ["google_gemini.*"], cap_calls: 2000,  confirm_policy: "human_for_writes",  approval_rules: [    { match: "google_gemini.files.delete", approver: "[email protected]" },    { match: "google_gemini.batches.cancel", approver: "[email protected]" },    { match: "google_gemini.batches.delete", approver: "[email protected]" }  ]}

Illustrative. Scoped to Google Gemini API (AI Studio) only, capped at 2,000 upstream calls, every write confirmed, and the destructive ones approved by a named person.

Cost

What Google Gemini API (AI Studio) costs on actionlake.

Every Google Gemini API (AI Studio) action is one upstream call, the only thing actionlake bills. Searching, check, cost, account and any call Google Gemini API (AI Studio) 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.

1upstream call per actionall 83 Google Gemini API (AI Studio) actions
83actions pageeach page of a long list is one call
37reads can be cacheda cache hit costs nothing
100,000calls a month, freepaid plans from US$49 a month for 250,000

See pricing

Action index

Every Google Gemini API (AI Studio) action.

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

Models18

google_gemini.models.async_batch_embed_contentAsync Batch Embed Content modelswrite
google_gemini.models.batch_embed_contentsBatch Embed Contents modelsread
google_gemini.models.batch_embed_textBatch Embed Text modelsread
google_gemini.models.batch_generate_contentBatch Generate Content modelswrite
google_gemini.models.count_message_tokensCount Message Tokens modelsread
google_gemini.models.count_text_tokensCount Text Tokens modelsread
google_gemini.models.count_tokensCount Tokens modelsread
google_gemini.models.embed_contentEmbed Content modelsread
google_gemini.models.embed_textEmbed Text modelsread
google_gemini.models.generate_answerGenerate Answer modelswrite
google_gemini.models.generate_contentGenerate Content modelswrite
google_gemini.models.generate_messageGenerate Message modelswrite
google_gemini.models.generate_textGenerate Text modelswrite
google_gemini.models.getGet modelsread
google_gemini.models.listList modelsread
google_gemini.models.predictPredict modelswrite
google_gemini.models.predict_long_runningPredict Long Running modelswrite
google_gemini.models.stream_generate_contentStream Generate Content modelswrite

Tuned models11

google_gemini.tuned_models.async_batch_embed_contentAsync Batch Embed Content tuned modelswrite
google_gemini.tuned_models.batch_generate_contentBatch Generate Content tuned modelswrite
google_gemini.tuned_models.createCreate tuned modelswrite
google_gemini.tuned_models.deleteDelete tuned modelsdestructive
google_gemini.tuned_models.generate_contentGenerate Content tuned modelswrite
google_gemini.tuned_models.generate_textGenerate Text tuned modelswrite
google_gemini.tuned_models.getGet tuned modelsread
google_gemini.tuned_models.listList tuned modelsread
google_gemini.tuned_models.stream_generate_contentStream Generate Content tuned modelswrite
google_gemini.tuned_models.transfer_ownershipTransfer Ownership tuned modelsdestructive
google_gemini.tuned_models.updateUpdate tuned modelswrite

Batches6

google_gemini.batches.cancelCancel batchesdestructive
google_gemini.batches.deleteDelete batchesdestructive
google_gemini.batches.getGet batchesread
google_gemini.batches.listList batchesread
google_gemini.batches.update_embed_content_batchUpdate Embed Content Batch batcheswrite
google_gemini.batches.update_generate_content_batchUpdate Generate Content Batch batcheswrite

Cached contents5

google_gemini.cached_contents.createCreate cached contentswrite
google_gemini.cached_contents.deleteDelete cached contentsdestructive
google_gemini.cached_contents.getGet cached contentsread
google_gemini.cached_contents.listList cached contentsread
google_gemini.cached_contents.updateUpdate cached contentswrite

Corpus permissions5

google_gemini.corpus_permissions.createCreate corpus permissionswrite
google_gemini.corpus_permissions.deleteDelete corpus permissionsdestructive
google_gemini.corpus_permissions.getGet corpus permissionsread
google_gemini.corpus_permissions.listList corpus permissionsread
google_gemini.corpus_permissions.updateUpdate corpus permissionswrite

File search stores5

google_gemini.file_search_stores.createCreate file search storeswrite
google_gemini.file_search_stores.deleteDelete file search storesdestructive
google_gemini.file_search_stores.getGet file search storesread
google_gemini.file_search_stores.import_fileImport File file search storeswrite
google_gemini.file_search_stores.listList file search storesread

Tuned model permissions5

google_gemini.tuned_model_permissions.createCreate tuned model permissionswrite
google_gemini.tuned_model_permissions.deleteDelete tuned model permissionsdestructive
google_gemini.tuned_model_permissions.getGet tuned model permissionsread
google_gemini.tuned_model_permissions.listList tuned model permissionsread
google_gemini.tuned_model_permissions.updateUpdate tuned model permissionswrite

Corpora4

google_gemini.corpora.createCreate corporawrite
google_gemini.corpora.deleteDelete corporadestructive
google_gemini.corpora.getGet corporaread
google_gemini.corpora.listList corporaread

Environments4

google_gemini.environments.createCreate environmentswrite
google_gemini.environments.deleteDelete environmentsdestructive
google_gemini.environments.getGet environmentsread
google_gemini.environments.listList environmentsread

Files4

google_gemini.files.deleteDelete filesdestructive
google_gemini.files.getGet filesread
google_gemini.files.listList filesread
google_gemini.files.registerRegister fileswrite

File search store documents3

google_gemini.file_search_store_documents.deleteDelete file search store documentsdestructive
google_gemini.file_search_store_documents.getGet file search store documentsread
google_gemini.file_search_store_documents.listList file search store documentsread

Dynamic2

google_gemini.dynamic.generate_contentGenerate Content dynamicwrite
google_gemini.dynamic.stream_generate_contentStream Generate Content dynamicwrite

Model operations2

google_gemini.model_operations.getGet model operationsread
google_gemini.model_operations.listList model operationsread

Tuned model operations2

google_gemini.tuned_model_operations.getGet tuned model operationsread
google_gemini.tuned_model_operations.listList tuned model operationsread

Auth tokens1

google_gemini.auth_tokens.createCreate auth tokenswrite

Corpus operations1

google_gemini.corpus_operations.getGet corpus operationsread

Environment files1

google_gemini.environment_files.listList environment filesread

File search store operations1

google_gemini.file_search_store_operations.getGet file search store operationsread

File search store upload operations1

google_gemini.file_search_store_upload_operations.getGet file search store upload operationsread

Generated file operations1

google_gemini.generated_file_operations.getGet generated file operationsread

Generated files1

google_gemini.generated_files.listList generated filesread

Questions

The Google Gemini API (AI Studio) MCP server, answered.

Is there a Google Gemini API (AI Studio) MCP server for Claude?

Yes. actionlake is an MCP gateway: a remote MCP server that gives Claude, or another MCP client, 83 Google Gemini API (AI Studio) actions across models, tuned models, batches and cached contents, through the same eight tools it uses for every other connector. actionlake is independent; it is not made or endorsed by Google Gemini API (AI Studio).

What do I need to connect Google Gemini API (AI Studio)?

Gemini API key from Google AI Studio (sent as x-goog-api-key). The free call account { op: "connections.requirements", service: "google_gemini" } lists exactly what to paste.

Can an AI agent delete or change things in Google Gemini API (AI Studio)?

Only if its key allows it, and only with an explicit confirmation. 12 Google Gemini API (AI Studio) actions are destructive, including delete files, cancel batches and delete batches. 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 Gemini API (AI Studio) API limit?

actionlake paces every call to the limits in the Google Gemini API (AI Studio) service file: at most 16 at a time per connection. Repeated reads are answered from cache (37 Google Gemini API (AI Studio) actions can be), which never reaches Google Gemini API (AI Studio).

What does a Google Gemini API (AI Studio) action cost?

One upstream call, the only unit actionlake bills. A list that pages bills one call per page (83 Google Gemini API (AI Studio) actions page). Cache hits, searches, dry runs and calls Google Gemini API (AI Studio) 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 Gemini API (AI Studio) API does it cover?

The Google Gemini API (AI Studio) API v1beta (discovery revision 20261020) at generativelanguage.googleapis.com. It is built from Google Gemini API (AI Studio)'s documentation as read on 22 October 2026. 83 actions in 21 areas, each one in the signed service file (version 1.0.0).

Does actionlake keep my Google Gemini API (AI Studio) data, and where?

actionlake processes and stores data in Sydney, Australia; there is no US region yet. Reads may be cached for up to an hour, 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 Gemini API (AI Studio) 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.

Give your agent Google Gemini API (AI Studio) today.

Connect Google Gemini API (AI Studio) free See pricing

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

Google Gemini API (AI Studio) MCP server for AI agents.

Connect Google Gemini API (AI Studio) in three steps.

Add actionlake to Claude.

Connect Google Gemini API (AI Studio).

Make the first call. Check before a write.

What an AI agent can do in Google Gemini API (AI Studio).

Models

Tuned models

Batches

Cached contents

Corpus permissions

File search stores

Tuned model permissions

Corpora