Firebase MCP Server
Experimental Firebase MCP server that gives AI coding agents access to Firestore, Auth, security rules, Cloud Messaging, and project management through the Firebase CLI.
Last reviewed Feb 28, 2026
What it does
The Firebase MCP Server (experimental) connects AI coding agents to your Firebase projects. Agents can manage projects and apps, explore and query Cloud Firestore data, manage authentication users, read security rules, send Cloud Messaging, and access crash analytics. It runs via the Firebase CLI and requires prior firebase login authentication.
Available tools
| Tool | What it does |
|---|---|
list_projects |
List Firebase projects in your account |
get_project |
Get project configuration and metadata |
list_apps |
List apps (iOS, Android, Web) for a project |
firestore_list_collections |
List collections in a Firestore database |
firestore_query |
Run read queries against Firestore |
auth_list_users |
List authentication users |
auth_get_user |
Get user details by UID |
get_security_rules |
Read Firestore and Storage security rules |
messaging_send |
Send Cloud Messaging messages |
crashlytics_* |
Access crash analytics data |
Tool names and availability may vary; use --only to limit which categories activate (e.g., firestore,auth).
Setup by tool
Cursor
First authenticate the Firebase CLI:
npx -y firebase-tools@latest login --reauth
Create .cursor/mcp.json in your project root:
{
"mcpServers": {
"firebase": {
"command": "npx",
"args": ["-y", "firebase-tools@latest", "experimental:mcp"]
}
}
}
Add --dir to specify the directory containing firebase.json if needed. Add --only firestore,auth to limit activated tools.
Claude Code
Add to your Claude configuration file:
{
"mcpServers": {
"firebase": {
"command": "npx",
"args": ["-y", "firebase-tools@latest", "experimental:mcp"]
}
}
}
VS Code / GitHub Copilot
Run firebase login first, then add the same configuration to your MCP client.
When to use this
- Firestore development: Agents query and explore Firestore data and collections during implementation
- Auth user management: Inspect users and troubleshoot auth flows
- Security rules: Read current rules before proposing changes
- Project setup: List projects and apps when scaffolding or documenting
- Crash analysis: Pull crash data into debugging conversations
Security considerations
- This is experimental; behavior may change without notice
- Firebase CLI uses your Google account; the agent inherits your project access
- Use only with development or staging projects, not production
- Limit tools with
--onlywhen possible to reduce the surface area - Never commit Firebase config with sensitive data; use environment variables for CI
Compatibility
Related MCP servers
AWS MCP Server
Open source MCP servers from AWS Labs that give AI coding agents access to AWS documentation, best practices, and contextual guidance for building on AWS.
Docker MCP Server
Docker MCP Gateway orchestrates MCP servers in isolated containers, providing secure discovery and execution of Model Context Protocol servers across AI coding tools.
Figma MCP Server
Official Figma MCP server that brings design context, variables, components, and Code Connect data into AI coding sessions for design-to-code workflows.