Stateless MCP server that proxies research queries to Gemini CLI, reducing agent context/model usage
Connect Runs on your machine
claude mcp add gemini-researcher -e GEMINI_API_KEY=<GEMINI_API_KEY> -e PROJECT_ROOT=<PROJECT_ROOT> -e RESPONSE_CHUNK_SIZE_KB=<RESPONSE_CHUNK_SIZE_KB> -e CACHE_TTL_MS=<CACHE_TTL_MS> -e DEBUG=<DEBUG> -e GOOGLE_APPLICATION_CREDENTIALS=<GOOGLE_APPLICATION_CREDENTIALS> -e GOOGLE_CLOUD_PROJECT=<GOOGLE_CLOUD_PROJECT> -e VERTEX_AI_PROJECT=<VERTEX_AI_PROJECT> -- npx -y [email protected]{
"mcpServers": {
"gemini-researcher": {
"command": "npx",
"args": [
"-y",
"[email protected]"
],
"env": {
"GEMINI_API_KEY": "<GEMINI_API_KEY>",
"PROJECT_ROOT": "<PROJECT_ROOT>",
"RESPONSE_CHUNK_SIZE_KB": "<RESPONSE_CHUNK_SIZE_KB>",
"CACHE_TTL_MS": "<CACHE_TTL_MS>",
"DEBUG": "<DEBUG>",
"GOOGLE_APPLICATION_CREDENTIALS": "<GOOGLE_APPLICATION_CREDENTIALS>",
"GOOGLE_CLOUD_PROJECT": "<GOOGLE_CLOUD_PROJECT>",
"VERTEX_AI_PROJECT": "<VERTEX_AI_PROJECT>"
}
}
}
}{
"mcpServers": {
"gemini-researcher": {
"command": "npx",
"args": [
"-y",
"[email protected]"
],
"env": {
"GEMINI_API_KEY": "<GEMINI_API_KEY>",
"PROJECT_ROOT": "<PROJECT_ROOT>",
"RESPONSE_CHUNK_SIZE_KB": "<RESPONSE_CHUNK_SIZE_KB>",
"CACHE_TTL_MS": "<CACHE_TTL_MS>",
"DEBUG": "<DEBUG>",
"GOOGLE_APPLICATION_CREDENTIALS": "<GOOGLE_APPLICATION_CREDENTIALS>",
"GOOGLE_CLOUD_PROJECT": "<GOOGLE_CLOUD_PROJECT>",
"VERTEX_AI_PROJECT": "<VERTEX_AI_PROJECT>"
}
}
}
}{
"servers": {
"gemini-researcher": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"[email protected]"
],
"env": {
"GEMINI_API_KEY": "<GEMINI_API_KEY>",
"PROJECT_ROOT": "<PROJECT_ROOT>",
"RESPONSE_CHUNK_SIZE_KB": "<RESPONSE_CHUNK_SIZE_KB>",
"CACHE_TTL_MS": "<CACHE_TTL_MS>",
"DEBUG": "<DEBUG>",
"GOOGLE_APPLICATION_CREDENTIALS": "<GOOGLE_APPLICATION_CREDENTIALS>",
"GOOGLE_CLOUD_PROJECT": "<GOOGLE_CLOUD_PROJECT>",
"VERTEX_AI_PROJECT": "<VERTEX_AI_PROJECT>"
}
}
}
}Replace the values in angle brackets with your own. Keys stay on your machine; we never see them.
Tools
This server runs locally, so its tools are listed by your MCP client once it starts.
Environment variables
| Name | Description |
|---|---|
GEMINI_API_KEYOptional, secret | Gemini API key (optional if you already authenticated Gemini CLI via "gemini" login) |
PROJECT_ROOTOptional | Override the project root directory used for path validation (defaults to current working directory) |
RESPONSE_CHUNK_SIZE_KBOptional | Chunk size threshold (KB) for large responses (default: 10) |
CACHE_TTL_MSOptional | Chunk cache TTL in milliseconds (default: 3600000 / 1 hour) |
DEBUGOptional | Enable debug logging (set to "true" or "1") |
GOOGLE_APPLICATION_CREDENTIALSOptional | Vertex AI / Google auth: path to a service account JSON credentials file |
GOOGLE_CLOUD_PROJECTOptional | Vertex AI / Google auth: GCP project ID (used by some auth configurations) |
VERTEX_AI_PROJECTOptional | Vertex AI / Google auth: Vertex AI project identifier (used by some auth configurations) |
Details
| Package | [email protected] (NPM, stdio) |
| First published | 26 Jan 2026 |
| Last updated | 26 Jan 2026 |
| Registry name | io.github.capyBearista/gemini-researcher |
More ai and media servers See all
Listing data from the official MCP registry. Status and tools from NestDaddy's daily check. Browse all MCP servers.