Claude MarketplacesSubmit
pavelguzenfeld avatar

Gemini

pavelguzenfeld/gemini-mcp

Connects Claude to Google's Gemini API so you can bounce questions off a second model without leaving your workflow. Exposes four tools: gemini_ask for one-off questions, gemini_analyze for targeted code review or text analysis, gemini_chat for multi-turn conversations with history, and gemini_models to list what's available. Supports all current Gemini models including the 1M token context window on 2.5-pro. Handles rate limits with retry logic and lets you override the model per call. Useful when you want a different perspective on architecture decisions, need massive context windows, or want to compare how two frontier models approach the same problem.

tools
8
auth
Secret needed
transport
STDIO
env var
1
last updated
Mar 10, 2026
View on GitHub

Tools

8 tools

Public tool metadata for what this MCP can expose to an agent.

GEMINI_COUNT_TOKENSCounts the number of tokens in text using Gemini tokenization. Useful for estimating costs, checking input limits, and optimizing prompts before making API calls.2 params

Counts the number of tokens in text using Gemini tokenization. Useful for estimating costs, checking input limits, and optimizing prompts before making API calls.

Parameters* required
modelstring
Model to use for token counting. Examples: 'gemini-1.5-flash', 'gemini-1.5-pro'default: gemini-1.5-flash
textstring
Text to count tokens for
GEMINI_EMBED_CONTENTGenerates text embeddings using Gemini embedding models. Converts text into numerical vectors for semantic search, similarity comparison, clustering, and classification tasks.4 params

Generates text embeddings using Gemini embedding models. Converts text into numerical vectors for semantic search, similarity comparison, clustering, and classification tasks.

Parameters* required
modelstring
Embedding model to use. Examples: 'text-embedding-004', 'embedding-001'default: text-embedding-004
task_typestring
Task type: 'RETRIEVAL_QUERY', 'RETRIEVAL_DOCUMENT', 'SEMANTIC_SIMILARITY', 'CLASSIFICATION', 'CLUSTERING'
textstring
Text to generate embeddings for
titlestring
Optional title for the content (for document embeddings)
GEMINI_GENERATE_CONTENTGenerates text content from prompts using Gemini models. Supports various models like Gemini Flash and Pro with configurable temperature, token limits, and safety settings for diverse text generation tasks.9 params

Generates text content from prompts using Gemini models. Supports various models like Gemini Flash and Pro with configurable temperature, token limits, and safety settings for diverse text generation tasks.

Parameters* required
max_output_tokensinteger
Maximum number of tokens to generate
modelstring
Model to use. Examples: 'gemini-1.5-flash', 'gemini-1.5-pro', 'gemini-2.0-flash-exp'default: gemini-1.5-flash
promptstring
Text prompt for content generation
safety_settingsarray
Safety filter settings
stop_sequencesarray
Sequences where generation should stop
system_instructionstring
System instruction to guide the model's behavior
temperaturenumber
Controls randomness (0.0 to 2.0)
top_kinteger
Top-k sampling parameter
top_pnumber
Nucleus sampling parameter (0.0 to 1.0)
GEMINI_GENERATE_IMAGEGenerates images from text prompts using Gemini 2.5 Flash Image Preview model (Nano Banana). Supports creative image generation with customizable parameters like aspect ratio, safety settings, and optional local file saving. Generated images are automatically uploaded to S3 an...9 params

Generates images from text prompts using Gemini 2.5 Flash Image Preview model (Nano Banana). Supports creative image generation with customizable parameters like aspect ratio, safety settings, and optional local file saving. Generated images are automatically uploaded to S3 an...

Parameters* required
max_output_tokensinteger
Maximum number of tokens to generate (max 32,768)
modelstring
Model to use. Use 'gemini-2.5-flash-image-preview' for image generationdefault: gemini-2.5-flash-image-preview
promptstring
Text prompt for image generation
safety_settingsarray
Safety filter settings
save_pathstring
Optional local path to save the generated image
system_instructionstring
System instruction to guide image generation behavior
temperaturenumber
Controls randomness (0.0 to 2.0)
top_kinteger
Top-k sampling parameter
top_pnumber
Nucleus sampling parameter (0.0 to 1.0)
GEMINI_GENERATE_VIDEOSGenerates videos from text prompts using Google's Veo models. Creates high-quality video content. Returns operation ID for tracking progress. After this, call GEMINI_WAIT_FOR_VIDEO to download the video using the operation ID.4 params

Generates videos from text prompts using Google's Veo models. Creates high-quality video content. Returns operation ID for tracking progress. After this, call GEMINI_WAIT_FOR_VIDEO to download the video using the operation ID.

Parameters* required
extrasobject
Additional parameters passed through to API
modelstring
Model to use. Examples: 'veo-3.0-generate-preview', 'veo-3.0-fast-generate-preview', 'veo-2.0-generate-001'default: veo-3.0-generate-preview
person_generationstring
Controls person generation in videos. Values: 'allow_adult' or 'dont_allow'. IMPORTANT: Veo 3 models in EU/UK/CH/MENA regions ONLY support 'allow_adult'. Veo 2 models support both values in all reg...
promptstring
Text prompt for Veo video generation
GEMINI_GET_VIDEOS_OPERATIONChecks the status of a Veo video generation operation. Use the operation name from GenerateVideos to track progress and get the download URL when complete.1 params

Checks the status of a Veo video generation operation. Use the operation name from GenerateVideos to track progress and get the download URL when complete.

Parameters* required
operation_namestring
Operation resource name returned by predictLongRunning
GEMINI_LIST_MODELSLists available Gemini and Veo models with their capabilities and limits. Useful for discovering supported models and their features before making generation requests.1 params

Lists available Gemini and Veo models with their capabilities and limits. Useful for discovering supported models and their features before making generation requests.

Parameters* required
filter_prefixstring
Filter models by name prefix (client-side). Leave empty to get all models.default:
GEMINI_WAIT_FOR_VIDEOPolls a Veo video generation operation until completion, then downloads and returns the video as a FileDownloadable with public URL.1 params

Polls a Veo video generation operation until completion, then downloads and returns the video as a FileDownloadable with public URL.

Parameters* required
operation_namestring
The operation name from video generation (e.g., 'models/...')

Sponsors

Promote your product hereReach 125k+ visitors monthlyPromote your product hereReach 125k+ visitors monthlyAgent CaffeineSponsoredClose your lid, come back to finished workSnugSponsoredThe clipboard manager for the AI eraPromote your product hereReach 125k+ visitors monthlyWaxSponsoredBlock anything with supervision, without erasing your iPhonePromote your product hereReach 125k+ visitors monthlyknobsSponsoredSee your site on any phone, any day, in any language

Configuration

GEMINI_API_KEY*secret

Google Gemini API key

Readme

gemini-mcp

npm MCP Registry Node.js License: MIT

Lightweight MCP server that exposes Google Gemini as tools for Claude Code (or any MCP client).

Use Gemini for second opinions, large-context analysis, code review, or anything where a different model perspective helps.

Tools

ToolDescription
gemini_askAsk Gemini a question or give it a task
gemini_analyzeSend code/text for analysis with a specific instruction
gemini_chatMulti-turn conversation with full history
gemini_modelsList available Gemini models

Quick Start

1. Get an API key

Go to Google AI Studio and create a free API key.

2. Install

Option A — Clone (recommended for Claude Code)

View the full README on GitHub

Details

Repository
github.com/pavelguzenfeld/gemini-mcp
Category
AI & LLM Tools
Package
claude-gemini-mcp
Transport
STDIO
Auth
Required
Registry
MCP registry, active

More from pavelguzenfeld

Related AI & LLM Tools MCP servers

View all
1pipeworx-io avatarGemini Cryptopipeworx-io/mcp-gemini-cryptoGemini Exchange MCP (public endpoints).0stars
2pipeworx-io avatarHuggingfacepipeworx-io/mcp-huggingfaceHugging Face Hub MCP — models, datasets, spaces0stars
3slashml-ai avatarSlashML (/ml) Persistent Agent Memory & Code Localizationslashml-ai/ml-memoryPersistent conversational memory, behavioral code search, and bug localization for AI agents.0stars
4sumanaddanki avatarSuma Memorysumanaddanki/suma-mcpPersistent memory for Claude Code and Cursor. Stop re-explaining your project every session.0stars
5j0hanz avatarMemory MCP Serverj0hanz/memory-mcpSQLite-backed MCP server for persistent memory, full-text retrieval, and graph traversal.0stars
6jacobbarglof avatarthron-memoryjacobbarglof/thron-memoryOne private memory for all your AI apps, on your phone and computer. Part of ThronOS.0stars