MCP Search Server - Web search proxy for AI assistants using Gemini
Project description
MCP Search Server
A Model Context Protocol (MCP) server that provides web search capabilities for AI assistants using Gemini models.
Features
- 🔍 Web search via Gemini's grounding API
- 📝 Three-section output: Summary + Search Results + Sources
- 🚀 Works with OpenCode, Zed, Claude Desktop, and any MCP-compatible tool
- ⚡ Result caching for faster repeated queries
- 🌐 Language-adaptive (responds in user's query language)
Installation
pip install mcp-search-server
Configuration
Environment Variables
| Variable | Required | Default | Description |
|---|---|---|---|
MCP_SEARCH_URL |
⚠️ Yes | http://localhost:8046/v1/messages |
Anthropic-compatible API endpoint |
MCP_SEARCH_API_KEY |
⚠️ Yes | - | API key for the endpoint |
MCP_SEARCH_MODEL |
No | gemini-3-flash |
Default search model |
MCP_SEARCH_CACHE_TTL |
No | 300 |
Cache TTL in seconds |
OpenCode Configuration
Add to your opencode.json:
{
"mcp": {
"mcp-search": {
"type": "local",
"command": ["python", "-m", "mcp_search"],
"env": {
"MCP_SEARCH_URL": "http://localhost:8046/v1/messages",
"MCP_SEARCH_API_KEY": "sk-your-api-key"
}
}
}
}
Zed Configuration
Add to your settings.json:
{
"language_models": {
"mcp": {
"mcp-search": {
"command": ["python", "-m", "mcp_search"],
"env": {
"MCP_SEARCH_URL": "http://localhost:8046/v1/messages",
"MCP_SEARCH_API_KEY": "sk-your-api-key"
}
}
}
}
}
Claude Desktop Configuration
Add to your claude_desktop_config.json:
{
"mcpServers": {
"mcp-search": {
"command": "python",
"args": ["-m", "mcp_search"],
"env": {
"MCP_SEARCH_URL": "http://localhost:8046/v1/messages",
"MCP_SEARCH_API_KEY": "sk-your-api-key"
}
}
}
}
Usage
Once configured, the AI assistant can use the mcp_search tool:
mcp_search(query="Who is Kanru Hua", max_results=5)
Parameters
| Parameter | Type | Required | Description |
|---|---|---|---|
query |
string | ✅ | Search keywords |
max_results |
integer | No | Number of results (default: 8) |
model |
string | No | Search model to use |
Output Format
The search returns three sections:
- Summary: Brief overview answering the query
- Search Results: Numbered list of results with title, URL, and description
- Sources: Complete list of source citations
License
MIT
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file mcp_gemini_search-0.1.0.tar.gz.
File metadata
- Download URL: mcp_gemini_search-0.1.0.tar.gz
- Upload date:
- Size: 9.2 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.14.2
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
8164d81693bbb69f26a1d326a4b3acf17584a4914657054be68ce871c2f535f8
|
|
| MD5 |
5b10e726a3564eb033c0e8af07856bb8
|
|
| BLAKE2b-256 |
3e19e0745e832e6b8c1a9dd9e07e51fa52971acddc6983392ac7981730536bee
|
File details
Details for the file mcp_gemini_search-0.1.0-py3-none-any.whl.
File metadata
- Download URL: mcp_gemini_search-0.1.0-py3-none-any.whl
- Upload date:
- Size: 7.0 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.14.2
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
fe0a406a909fc61f07aa786f0810bf7b745eee2a20a2f015b868fa4c16761789
|
|
| MD5 |
cfa76415366e3fa6669f64c394e56f0c
|
|
| BLAKE2b-256 |
888bda36af453814cf4e89df02a1eaa9c21edb4484097807611829e7153cfbac
|