Gemini MCP Server
I've been running this Gemini MCP server in my Claude Desktop setup for months. It's one of the few I leave on permanently - not because Gemini replaces Claude, but because grounded search, image generation, SVG diagrams, and video are things Gemini does well. Having them as tools inside Claude beats switching browser tabs. Its 13 documented tools cover gemini, image, generate.
People connecting media tools to Claude, Cursor, VS Code, or another MCP client. The project is written in TypeScript.
VERIFIED ACTIVE
LAST COMMIT 2026-08-17 · ★ 30 · #17 OF 155 MAINTAINED MEDIA · VERIFIED 2026-08-25
Apache-2.0 · TypeScript servers · how we verify → /methodology
01 · Install Gemini
before you install - you'll need
Set GEMINI_API_KEY, GEMINI_IMAGE_OUTPUT_DIR before connecting.
Claude Code
claude mcp add houtini-ai-gemini-mcp -- npx -y @houtini/gemini-mcp Claude Desktop / Cursor / VS Code - add to config
{
"mcpServers": {
"houtini-ai-gemini-mcp": {
"command": "npx",
"args": [
"-y",
"@houtini/gemini-mcp"
]
}
}
} Same JSON for Cursor. For VS Code, rename the top-level key from `mcpServers` to `servers`.
Using another client? Same JSON, different key
Claude Desktop · mcpServers
Cursor · mcpServers
VS Code · servers
Windsurf · mcpServers
Zed · context_servers
Cline · mcpServers
Roo Code · mcpServers
Continue · mcpServers
LibreChat · mcpServers
Gemini CLI · mcpServers
Codex CLI · mcp_servers
Full setup guides: every client.
02 · Evidence
Security posture
What to check before giving this server access to your agent - from the registry, GitHub, and our own probes. We don't score safety; we show what's verifiable.
runs as local process (stdio) - runs on your machine with your user's permissions
license Apache-2.0 - declared in the repository
npm package @houtini/gemini-mcp - the @houtini scope differs from the repo owner (houtini-ai); confirm it's the project's official package
registry namespace io.github.houtini-ai is GitHub-verified and matches the repo owner
03 · What Gemini can do
Prose above is summarized from the project's README and registry record - no invented capabilities.
What you can build
With this server connected, an agent can list available models from the Gemini API, edit images with natural-language instructions. Multi-turn continuity via thought signatures, and read a local image file and return base64 for any image tool.
Capability map
Tools grouped from the project's README - what Gemini lets an agent do.
Gemini
5 tools - e.g. Chat with Gemini 3.1 Pro. Google Search grounding on by default. Supports thinking_level; Multi-step iterative research with Google Search. Synthesises comprehensive reports; Lists available models from…
gemini_chat · gemini_deep_research · gemini_list_models · gemini_help · gemini_prompt_assistant
Generate
4 tools - e.g. Image generation with optional search grounding. Full-res saved to disk; Video generation with Veo 3.1 - 4-8 seconds at up to 4K with native audio; Production-ready SVG: diagrams, illustrations, icons…
generate_image · generate_video · generate_svg · generate_landing_page
Image
4 tools - e.g. Edit images with natural-language instructions. Multi-turn continuity via thought signatures; Fast image descriptions using Gemini 3 Flash; Structured extraction and analysis using Gemini 3.1 Pro
edit_image · describe_image · analyze_image · load_image_from_path
Latest releases
v2.6.1 · 2026-08-08
Two fixes for the two things new users hit first. · A wrong GEMINI_API_KEY used to take the process down hard · Not "log an error and exit" - a native crash. On Windows it surfaced as a libuv assertion and exit code…
04 · Who maintains Gemini
gemini is maintained by houtini-ai. We track 9 MCP servers from houtini-ai - 7 actively maintained, 215 combined GitHub stars, oldest repo from Sep 2025. Full record: all servers from houtini-ai.
- lm Connect Claude to any OpenAI-compatible LLM endpoint and offload routine work to a local model. ★ 111
- seo-audit-console Technical SEO in Claude: GSC + crawl + DataForSEO - audit, market sizing, content recon. ★ 12
- geo-analyzer Scores content on the signals that get it cited by ChatGPT, Perplexity and AI Overviews ★ 21
- google-knowledge-graph Search Google's Knowledge Graph for structured information about real-world entities ★ 11
- seo-crawler-mcp Crawl and analyse websites for SEO errors using Crawlee with SQLite storage ★ 16
- fmp Access financial data and market information via Financial Modeling Prep API ★ 2
05 · Facts
- repository
- github.com/houtini-ai/gemini-mcp
- category
- media - ranked #17 of 155 actively-maintained media servers as of 2026-08-25.
- release cadence
- 1 release in the last 90 days (latest 2026-08-08)
- registry
- io.github.houtini-ai/gemini (active, first published 2026-01-29 · 2 versions)
- packages
- npm:@houtini/gemini-mcp
06 · Gemini FAQ
Is Gemini still maintained?
Yes - as of 2026-08-25, its last commit was 2026-08-17 and it shipped 1 release in the last 90 days. We re-verify nightly.
What can Gemini do?
With this server connected, an agent can list available models from the Gemini API, edit images with natural-language instructions. Multi-turn continuity via thought signatures, and read a local image file and return base64 for any image tool.
How do I install Gemini?
Run `npx -y @houtini/gemini-mcp`. The README documents 2 environment variables (GEMINI_API_KEY, GEMINI_IMAGE_OUTPUT_DIR) to set first. Set GEMINI_API_KEY, GEMINI_IMAGE_OUTPUT_DIR before connecting. You can also paste the ready-made client config above.
Does Gemini run locally?
Yes - it's a stdio server: it runs on your machine (via npx) with your user's permissions. Your data stays local unless the server itself calls external APIs.
07 · Alternatives to Gemini
Alternatives to Gemini
Maintained media servers if Gemini isn't the fit.
- Openshorts Turn long videos into viral vertical shorts and publish them to TikTok, Instagram and YouTube. ★ 3,410 · 2026-08-25
- Claude Real Video Let any LLM watch a video locally - and search everything it has ever watched. ★ 2,061 · 2026-08-22
- Voicemode Natural voice conversations for AI assistants - STT/TTS via MCP ★ 1,337 · 2026-08-09
- Nova3 D Structured, part-aware 3D generation for AI agents. Named-part GLB, preview URL, Blender script. ★ 682 · 2026-08-12
- Comfyui MCP MCP server + Claude Code plugin for ComfyUI: run workflows, generate images, manage models & VRAM. ★ 671 · 2026-08-25
- Yutu The AI-powered toolkit that grows your YouTube channel on autopilot ★ 620 · 2026-08-23
More media MCP servers · Pixel Surgeon · Drawthings · Florence2 · Secondhand MCP
More TypeScript MCP servers · Google Knowledge Graph · Seo Audit Console · MERX - TRON Resource Exchange · Grok MCP · Hqbase · see all