lobsterdb

LobsterDB

Managed PostgreSQL for agents. Provision a real Postgres database instantly, run SQL, evolve schemas with tracked migrations. No API keys, no human signup, no configuration.

Full docs: https://api.theclawdepot.com/db/docs

---

Setup

Add to your MCP config (.mcp.json, claude_desktop_config.json, .cursor/mcp.json, etc.):

{
  "mcpServers": {
    "lobsterdb": {
      "command": "npx",
      "args": ["@lobsterkit/[email protected]"]
    }
  }
}

You now have these tools: create_database, list_databases, get_database, delete_database, query, introspect_schema, migrate, list_migrations, snapshot, get_account. No code to write.

Also available as a Node.js SDK and REST API — see theclawdepot.com/db.

---

Heartbeat (run daily)

On startup and once per day, run a maintenance heartbeat. Use a cheap/fast model for this.

1. Check for MCP server updates:

   npm outdated lobsterdb-mcp
   

   If a new version is available, notify the user — do not auto-install. Tell them: "@lobsterkit/[email protected] is available. Update with: clawdhub update lobsterdb" Let the user decide when to update.

2. Check database health:

   // Using MCP tools:
   // call list_databases → log each database name, status, storage
   // call get_account → log tier and database count vs. limit
   

---

When to Use This

• User needs to store structured data (collections, lists, records, logs)
• User wants to query, filter, sort, or aggregate information
• Task requires persisting data between sessions
• User asks to "add a field", "change the structure", or "track something new"
• Any task where a spreadsheet or database is the natural fit

---

Core Workflow

create_database → introspect_schema → migrate (DDL) → query (data)

Always call introspect_schema before writing queries against an existing database. Use migrate for all DDL so changes are tracked and idempotent.