Manage secrets directly from your AI-powered editor
Connect HLOS to Cursor via MCP (Model Context Protocol) and let your AI assistant help manage secrets without ever exposing the actual values.
Connect HLOS to Cursor in just a few steps.
Install the server globally or run with npx
npm install -g hlos-mcp-serverAdd HLOS to your Cursor MCP configuration
code ~/.cursor/config.jsonLog in to connect your HLOS account
hlos loginAsk Cursor to manage your secrets!
"Hey Cursor, what secrets do I need for this project?"Add the following to your Cursor MCP configuration file at ~/.cursor/config.json:
{
"mcpServers": {
"hlos": {
"command": "npx",
"args": ["-y", "hlos-mcp-server"],
"env": {
"HLOS_API_URL": "https://hlos.ai/api"
}
}
}
}Cursor never sees your actual secret values. The MCP server only exposes metadata like secret names, providers, and health status. You enter values directly in the HLOS dashboard.
Once connected, Cursor can use these HLOS tools to help manage your secrets:
list_secretsList secrets with metadata (names only, never values)
check_secret_existsVerify if a secret is configured in your space
get_secret_infoGet detailed metadata about a secret
suggest_missing_secretsAnalyze code for missing environment variables
create_secretRequest secret creation (requires user approval)
check_healthSecret health dashboard with rotation warnings
"What secrets do I need for this Next.js project?""Check if STRIPE_SECRET_KEY is configured""Show me the health status of my secrets""Help me add a new OpenAI API key"