Uses Gemini's large context window to store and cache context for other AI tools such as Cursor.
Setup
Choose your client, copy the config, then fill in any required environment variables.
claude mcp add gemini-context --env GEMINI_API_KEY='<GEMINI_API_KEY>' -- node dist/mcp-server.js~/.codex/config.toml
[mcp_servers.gemini-context]
command = "node"
args = ["dist/mcp-server.js"]
env = { GEMINI_API_KEY = "<GEMINI_API_KEY>" }
# codex mcp add gemini-context --env GEMINI_API_KEY='<GEMINI_API_KEY>' -- node dist/mcp-server.js~/.cursor/mcp.json
{
"mcpServers": {
"gemini-context": {
"command": "node",
"args": [
"dist/mcp-server.js"
],
"env": {
"GEMINI_API_KEY": "<GEMINI_API_KEY>"
}
}
}
}.vscode/mcp.json
{
"servers": {
"gemini-context": {
"type": "stdio",
"command": "node",
"args": [
"dist/mcp-server.js"
],
"env": {
"GEMINI_API_KEY": "<GEMINI_API_KEY>"
}
}
}
}~/.gemini/settings.json
{
"mcpServers": {
"gemini-context": {
"command": "node",
"args": [
"dist/mcp-server.js"
],
"env": {
"GEMINI_API_KEY": "<GEMINI_API_KEY>"
}
}
}
}claude_desktop_config.json
{
"mcpServers": {
"gemini-context": {
"command": "node",
"args": [
"dist/mcp-server.js"
],
"env": {
"GEMINI_API_KEY": "<GEMINI_API_KEY>"
}
}
}
}.trae/mcp.json
{
"mcpServers": {
"gemini-context": {
"command": "node",
"args": [
"dist/mcp-server.js"
],
"env": {
"GEMINI_API_KEY": "<GEMINI_API_KEY>"
}
}
}
}Environment variables
GEMINI_API_KEY
Replace the placeholders with your own credentials and keep them out of version control.