MCP server for Gemini AI image processing - 基于 Gemini 2.5 Flash 的图片处理 MCP 服务
Setup
Choose your client, copy the config, then fill in any required environment variables.
claude mcp add nanobananamcp --env GEMINI_API_KEY='<GEMINI_API_KEY>' --env GEMINI_MCP_OUTPUT_DIR='<GEMINI_MCP_OUTPUT_DIR>' -- uvx gemini-mcp~/.codex/config.toml
[mcp_servers.nanobananamcp]
command = "uvx"
args = ["gemini-mcp"]
env = { GEMINI_API_KEY = "<GEMINI_API_KEY>", GEMINI_MCP_OUTPUT_DIR = "<GEMINI_MCP_OUTPUT_DIR>" }
# codex mcp add nanobananamcp --env GEMINI_API_KEY='<GEMINI_API_KEY>' --env GEMINI_MCP_OUTPUT_DIR='<GEMINI_MCP_OUTPUT_DIR>' -- uvx gemini-mcp~/.cursor/mcp.json
{
"mcpServers": {
"nanobananamcp": {
"command": "uvx",
"args": [
"gemini-mcp"
],
"env": {
"GEMINI_API_KEY": "<GEMINI_API_KEY>",
"GEMINI_MCP_OUTPUT_DIR": "<GEMINI_MCP_OUTPUT_DIR>"
}
}
}
}.vscode/mcp.json
{
"servers": {
"nanobananamcp": {
"type": "stdio",
"command": "uvx",
"args": [
"gemini-mcp"
],
"env": {
"GEMINI_API_KEY": "<GEMINI_API_KEY>",
"GEMINI_MCP_OUTPUT_DIR": "<GEMINI_MCP_OUTPUT_DIR>"
}
}
}
}~/.gemini/settings.json
{
"mcpServers": {
"nanobananamcp": {
"command": "uvx",
"args": [
"gemini-mcp"
],
"env": {
"GEMINI_API_KEY": "<GEMINI_API_KEY>",
"GEMINI_MCP_OUTPUT_DIR": "<GEMINI_MCP_OUTPUT_DIR>"
}
}
}
}claude_desktop_config.json
{
"mcpServers": {
"nanobananamcp": {
"command": "uvx",
"args": [
"gemini-mcp"
],
"env": {
"GEMINI_API_KEY": "<GEMINI_API_KEY>",
"GEMINI_MCP_OUTPUT_DIR": "<GEMINI_MCP_OUTPUT_DIR>"
}
}
}
}.trae/mcp.json
{
"mcpServers": {
"nanobananamcp": {
"command": "uvx",
"args": [
"gemini-mcp"
],
"env": {
"GEMINI_API_KEY": "<GEMINI_API_KEY>",
"GEMINI_MCP_OUTPUT_DIR": "<GEMINI_MCP_OUTPUT_DIR>"
}
}
}
}Environment variables
GEMINI_API_KEYGEMINI_MCP_OUTPUT_DIR
Replace the placeholders with your own credentials and keep them out of version control.