Search for Google images, view results, and download them directly within your IDE.
Query for Google images from your editor/IDE and view search results as visual previews in your conversation. Then download and persist the selected image(s) to your local project.
The unique feature of this MCP is that it displays search results directly in the chat history as actual images, not just text descriptions or links. This visual integration currently works with Cline and Windsurf only.
Follow these steps to set up your Google Images Search MCP:
*
(asterisk) to search the entire webAdd the following configuration to your MCP settings:
{
"mcpServers": {
"googleImagesSearch": {
"command": "npx",
"args": ["-y", "@srigi/mcp-google-images-search"],
"env": {
"API_KEY": "your-google-api-key-here",
"SEARCH_ENGINE_ID": "your-search-engine-id-here"
},
"autoApprove": ["search_image", "persist_image"]
}
}
}
Replace your-google-api-key-here
and your-search-engine-id-here
with the values from steps 1 and 2.
Here's how to use the MCP once configured:
Search for images: Ask your AI assistant to search for images
Search for image of F-22 in-air
Get more results: Request additional search results
Show me 3 more images
Save an image: Ask to save a specific result to your project
Save the 3rd image to the "assets" folder
The MCP will display the search results as actual images in your chat history, and you can easily save any of them to your local project directory.
Feel free to join development of this MCP. Quality contribution is welcomed.
For development, you need to start the dev build and use a different MCP configuration that points to the continuously recompiled source files:
.envrc (example)
to .envrc
API_KEY
and SEARCH_ENGINE_ID
in .envrc
pnpm install
pnpm dev
This starts TypeScript compiler in watch mode, continuously recompiling changes.
Use this MCP configuration for development (replace the path with your actual project path):
{
"mcpServers": {
"googleImagesSearch": {
"command": "node",
"args": ["/absolute/path/to/project/src/index.js"],
"env": {
"API_KEY": "your-google-api-key-here",
"SEARCH_ENGINE_ID": "your-search-engine-id-here"
},
"autoApprove": ["search_image", "persist_image"]
}
}
}
For debugging and testing, you can start the MCP inspector:
pnpx @modelcontextprotocol/inspector
Open the inspector URL that includes the MCP_PROXY_AUTH_TOKEN
(reported on the terminal). In the inspector's UI fill the same Command: and Arguments: as in above MCP configuration, then click "▷ Connect". Finally click "List Tools" to see the MPC's tools, and invoke them with desired arguments.
Check the log file for debugging information.
Note: When you make changes to the source code, the dev build will automatically recompile, but you may need to restart the MCP server in your IDE to apply the changes.
Search for images using Google Custom Search API.
Parameters:
query
(string, required): Search query for imagescount
(number, optional): Number of results to return (1-10, default: 2)safe
(string, optional): Safe search setting - 'off', 'medium', 'high' (default: 'off')startIndex
(number, optional): Starting index for pagination (default: 1)Download and save images from URLs to your local project directory.
Parameters:
link
(string, required): URL to the full-quality imagepath
(string, required): Relative path where to save the image (can be folder or folder/filename.ext)Features:
add example usage screencast
open GH issue for image support in the chat history:
allow specifying of the output file in the chat
configurable logging severity
log directory in user profile folder (AppData\Roaming, Library/Application Support) or configurable
image fetch timeout & retry
Fetches user data and event information from the Connpass platform using the Connpass and Gemini APIs.
Best people search engine that reduces the time spent on talent discovery.
Provides AI assistants with direct access to Mastra.ai's complete knowledge base.
Search for IP addresses using Chaitin's IP Intelligence API.
Kagi search API integration
Search and retrieve academic papers from the arXiv repository via its API.
An enhanced MCP server for SearXNG web searching, utilizing a category-aware web-search, web-scraping, and includes a date/time retrieval tool.
All-in-One SEO & Web Intelligence Toolkit API from FetchSERP.
Search and get up-to-date information about NPM, Cargo, PyPi, and NuGet packages.
A Model Context Protocol (MCP) server providing access to Google Programmable Search Engine (PSE) and Custom Search Engine (CSE).