Google Gemini MCP server for your AI

Generates multiple embedding vectors from the input `Content` which consists of a batch of strings represented as `EmbedContentRequest` objects.

Runs a model's tokenizer on input `Content` and returns the token count.

Enqueues a batch of `GenerateContent` requests for batch processing.

Enqueues a batch of `EmbedContent` requests for batch processing.

Generates a text embedding vector from the input `Content` using the specified [Gemini Embedding model](https://ai.google.dev/gemini-api/docs/models/gemini#text-embedding).

Generates a model response given an input `GenerateContentRequest`.

Create or edit an image with Gemini's image models ("Nano Banana").

Same as Predict but returns an LRO.

Gets information about a specific `Model` such as its version number, token limits, [parameters](https://ai.google.dev/gemini-api/docs/models/generative-models#model-parameters) and other metadata.

Lists the [`Model`s](https://ai.google.dev/gemini-api/docs/models/gemini) available through the Gemini API.

Cancel a running Batch API job.

Creates an empty `FileSearchStore`.

Delete a Batch API job record (it stops tracking it; it does not cancel a running job).

Deletes the `File`.

Deletes a `FileSearchStore`.

Download the content of a file Gemini GENERATED: most usefully a Batch API results file (the batch's responsesFile, JSON Lines, one result per line), which comes back as text under `raw`.

Get a Batch API job's state and, once it succeeded, its results (inlined responses or a responsesFile).

Gets the metadata for the given `File`.

Gets information about a specific `FileSearchStore`.

Imports a `File` from File Service to a `FileSearchStore`.

List this project's Batch API jobs (generation and embeddings) with their state.

Lists all `FileSearchStores` owned by the user.

Lists the metadata for `File`s owned by the requesting project.

Upload a file to the Gemini Files API so a prompt can reference it (PDFs, images, audio, video, text, code).

Add a document to a File Search store in one step: the connector uploads the bytes, and Gemini chunks, embeds and indexes them for retrieval with the fileSearch tool.

Creates CachedContent resource.

Deletes CachedContent resource.

Reads CachedContent resource.

Lists CachedContents.

Updates CachedContent resource (only expiration is updatable).

Cancels an interaction by id.

Creates a new interaction.

Create a custom text-to-speech voice.

Deletes the interaction by id.

Retrieves the full details of a single interaction based on its `Interaction.id`.

Lists all Agents.

List the voices available for text-to-speech: your custom voices first, then Google's prebuilt catalog (Puck, Charon, Kore, ...).

Deletes a `Document`.

Poll ANY long-running Gemini operation by the `name` the starting call returned: video generation (generate_videos), batch jobs (create_batch, create_embedding_batch: names like batches/abc), File Search uploads and imports, and tuning jobs.

Lists all `Document`s in a `Corpus`.

Generates a [streamed response](https://ai.google.dev/gemini-api/docs/text-generation?lang=python#generate-a-text-stream) from the model given an input `GenerateContentRequest`.

Creates a new trigger that will invoke the specified agent on the given cron schedule.

Creates a new Webhook.

Deletes a trigger.

Deletes a Webhook.

Gets details of a single trigger.

Gets a specific Webhook.

Lists executions for a trigger.

Lists triggers for a project.

Lists all Webhooks.

Sends a ping event to a Webhook.

Generates a new signing secret for a Webhook.

Runs a trigger immediately.

Updates a trigger.

Updates an existing Webhook.

Creates an environment.

Deletes an environment.

Gets an environment.

Retrieves file metadata or directory contents from an environment's snapshot.

Lists environments.

Registers a Google Cloud Storage files with FileService.

Updates a batch of GenerateContent requests for batch processing.

Updates a batch of EmbedContent requests for batch processing.

Write a file into an agent environment's workspace (the sandbox an Antigravity agent works in).

Creates a credential.

Deletes a credential.

Gets metadata of a single credential (no secret fields).

Lists credentials for a project.

Updates a credential.

Creates a new Agent (Typed version for SDK).

Deletes an Agent.

Delete a stored custom voice.

Gets a specific Agent.

Get one custom voice (created with create_voice and store = true) by id, including its description and, for a prompted voice, its sample audio.

Delete a finished long-running operation record (stable v1 API).

Gets information about a specific `Document`.

List the project's long-running operations (stable v1 API).

Generates a model response given an input `GenerateContentRequest`.

Generates a [streamed response](https://ai.google.dev/gemini-api/docs/text-generation?lang=python#generate-a-text-stream) from the model given an input `GenerateContentRequest`.

Poll one model operation, such as a Veo video generation, by its full name.

List the long-running operations (such as Veo video generations) started on one model.

Creates a token that can be used to constrain the behavior of a BidiGenerateContent session.

Downloads media from a `FileSearchStore`.

Poll a File Search import operation by its full name.

Poll a File Search upload operation by its full name.

Poll the operation producing a generated file by its full name.

Lists the generated files owned by the requesting project.