Gemini MCP Server
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| type | No | Service account field (paste JSON approach). Set to 'service_account'. | |
| auth_uri | No | Auth URI (paste JSON approach). | |
| client_id | No | Service account client ID (paste JSON approach). | |
| token_uri | No | Token URI (paste JSON approach). | |
| project_id | No | GCP project ID (paste JSON approach). | |
| HTTPS_PROXY | No | Optional proxy URL for users behind proxy/VPN. | |
| private_key | No | Private key (paste JSON approach). | |
| client_email | No | Service account email (paste JSON approach). | |
| GEMINI_API_KEY | No | API key from Google AI Studio for AI Studio authentication mode. | |
| private_key_id | No | Private key ID (paste JSON approach). | |
| universe_domain | No | Universe domain (paste JSON approach). Often 'googleapis.com'. | |
| GOOGLE_CLOUD_PROJECT | No | GCP project ID for explicit Vertex AI mode. | |
| client_x509_cert_url | No | Client cert URL (paste JSON approach). | |
| GOOGLE_CLOUD_LOCATION | No | Region for Vertex AI (default: 'global'). | |
| GOOGLE_CREDENTIALS_JSON | No | Entire service account JSON as a single string (explicit Vertex AI mode). | |
| GOOGLE_GENAI_USE_VERTEXAI | No | Set to 'true' to enable explicit Vertex AI mode. | |
| auth_provider_x509_cert_url | No | Auth provider cert URL (paste JSON approach). | |
| GOOGLE_APPLICATION_CREDENTIALS | No | File path to service account JSON key (explicit Vertex AI mode). |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| gemini_multimodal_queryA | Query using images + text for multimodal understanding. Analyze designs, diagrams, screenshots, or any visual content with natural language questions. |
| gemini_analyze_contentA | Analyze code, documents, or data. Supports file path or direct content input. Provides summarization, code review, explanation, optimization, and debugging. Auto-detects content type and programming language. TIP: This tool supports PARALLEL calls - analyze multiple files simultaneously for faster results. |
| gemini_analyze_codebaseA | Analyze entire codebase using 1M token context. Supports directory path, file paths, or file contents. Provides architecture overview, identifies patterns, security issues, performance bottlenecks, and dependency problems. |
| gemini_brainstormA | Generate creative ideas and solutions. Provides multiple ideas with pros/cons and feasibility assessment. Supports reading project context files to generate ideas that fit your project. |
| gemini_searchA | Search the web using Gemini's built-in Google Search grounding. Features:
Usage Tips:
|
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 5 tools
Each tool targets a distinct capability: multimodal query, content analysis, codebase analysis, brainstorming, and web search. No two tools have overlapping purposes, and descriptions are clear about their unique inputs and outputs.
All tools share the consistent 'gemini_' prefix and use descriptive action words (query, analyze, brainstorm, search). However, 'gemini_multimodal_query' phrases the action as a noun-verb combination, while others are more straightforward verb-object patterns, so there is a minor deviation.
With only 5 tools, the server is well-scoped and avoids unnecessary bloat. Each tool covers a distinct major feature of the Gemini API, making the set easy to navigate and understand.
The tool surface covers the primary Gemini capabilities: multimodal understanding, content analysis, codebase analysis, ideation, and web search. A pure text generation tool is missing, but the existing tools can handle summarization, explanation, and brainstorming, so there are no critical dead ends.