Skip to content
MCP server

Dynobase

Let AI tools query DynamoDB through the Dynobase desktop app, read-only by default

Runs locallyStatus and history
Pricing
Free trial
Platforms
macOS, Windows, Linux
Runs
Locally
Version
1.0.0
Listing updated
3 days ago

About Dynobase

Dynobase is an MCP server for databases. It runs locally on your machine and is started by your AI client.

This server connects MCP clients to Dynobase 3.0, the desktop client for Amazon DynamoDB, so an assistant can work with your tables through AWS profiles already loaded in the app. Clients see tool results only and never receive your AWS credentials. Dynobase must stay running, because the MCP server lives inside the app and listens on 127.0.0.1 behind a bearer token you can regenerate.

Writes are blocked until you turn on Allow writes in Settings, AI, MCP server. Item writes made by a client are recorded in Dynobase's History with the previous version so they can be reverted, but PartiQL writes cannot be undone. AWS IAM permissions still apply to every request.

The npm-style bridge in the repository, MIT-licensed, serves clients that only launch stdio servers such as Claude Desktop, Codex and Windsurf. Claude Code and Cursor can connect over HTTP directly. Each call returns up to about 1 MB of items and times out after 60 seconds.

Install

Run in a terminal

claude mcp add dynobase-mcp -- npx -y https://github.com/Dynobase/dynobase-mcp/releases/download/v1.0.0/dynobase-1.0.0.mcpb

Add to claude_desktop_config.json

{
  "mcpServers": {
    "dynobase-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "https://github.com/Dynobase/dynobase-mcp/releases/download/v1.0.0/dynobase-1.0.0.mcpb"
      ]
    }
  }
}
Install in CursorOpens Cursor, which asks you to confirm. Or add it by hand:

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "dynobase-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "https://github.com/Dynobase/dynobase-mcp/releases/download/v1.0.0/dynobase-1.0.0.mcpb"
      ]
    }
  }
}
Install in VS CodeOpens VS Code, which asks you to confirm. Or add it by hand:

Add to .vscode/mcp.json

{
  "servers": {
    "dynobase-mcp": {
      "type": "stdio",
      "command": "npx",
      "args": [
        "-y",
        "https://github.com/Dynobase/dynobase-mcp/releases/download/v1.0.0/dynobase-1.0.0.mcpb"
      ]
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "dynobase-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "https://github.com/Dynobase/dynobase-mcp/releases/download/v1.0.0/dynobase-1.0.0.mcpb"
      ]
    }
  }
}

Add to your client's MCP settings (most use this shape)

{
  "mcpServers": {
    "dynobase-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "https://github.com/Dynobase/dynobase-mcp/releases/download/v1.0.0/dynobase-1.0.0.mcpb"
      ]
    }
  }
}

Tools this server provides (8)

Taken from the project documentation; the running server may differ.

list_profiles
Lists loaded profiles with their regions and table names
list_tables
Lists the table names of one profile
describe_table
Returns the full DescribeTable output plus TTL
scan
Returns one page of a Scan
query
Returns one page of a Query
get_item
Returns one item by primary key
partiql
Runs one PartiQL statement, SELECT only unless writes are allowed
explain_query
Says which index a key condition would use and whether it is a Query or a Scan

What it does

Reuses AWS profiles
Lists loaded profiles and tables without copying keys into each client.
Read tools by default
Scan, query, get item, PartiQL SELECT and describe table work without write access.
Index explanation
Reports which index a key condition would use and whether it is a Query or a Scan.
Opt-in writes
Put, update, delete and batch write stay blocked until Allow writes is enabled.
History for item writes
Item writes are recorded with the previous version so they can be reverted.
Local token-protected server
Listens on 127.0.0.1 only with a regenerable bearer token.

Good for

  • Ask a coding assistant what a DynamoDB table's schema looks like
  • Run exploratory queries while building an app
  • Check which index a query would hit before shipping it
  • Make supervised item edits with a revert path

Pricing

Free trial

Bridge is MIT; Dynobase is paid with a 7-day free trial

Prices change. Check dynobase.dev before you buy.

Dynobase FAQ

Can the AI tool see my AWS keys?

No. The site says clients see tool results, never your credentials.

Is it read-only?

Yes by default. Writes need Allow writes turned on in Dynobase.

Can I undo a change?

Item writes are recorded in History and can be reverted; PartiQL writes cannot.

Does Dynobase need to be running?

Yes, the local server is only available while the app is open.

What does it cost?

The bridge is MIT-licensed, but Dynobase itself is a paid app with a 7-day free trial.

Is Dynobase free?

Dynobase is a paid product with a free trial.

How do I add Dynobase to Claude Code?

Run this command in your terminal: claude mcp add dynobase-mcp -- npx -y https://github.com/Dynobase/dynobase-mcp/releases/download/v1.0.0/dynobase-1.0.0.mcpb

What are the best alternatives to Dynobase?

The closest alternatives to Dynobase are REA, Codebase Memory, GitHub, Serena MCP: the IDE for your agent and XERJ.

Alternatives to Dynobase

  1. REA

    MCP server

    CLI and MCP server that lets coding agents reverse engineer apps and binaries

    Runs locally
    • Open source
    • 464K GitHub stars
    • Developer tools
  2. Codebase Memory

    MCP server

    Local MCP server that builds a code knowledge graph for AI coding agents

    Runs locally
    • Open source
    • 48.7K GitHub stars
    • Knowledge and memory
  3. GitHub

    MCP server

    Official MCP server connecting AI assistants to GitHub repos, issues and workflows

    Not checked yet
    • Remote
    • Open source
    • 34.5K GitHub stars
    • Developer tools
  4. MCP toolkit giving coding agents symbol-level code retrieval and editing

    Runs locally
    • Open source
    • 31.5K GitHub stars
    • Developer tools
  5. XERJ

    MCP server

    Local search engine for AI agents with cited answers over your code and docs

    Runs locally
    • Open source
    • 19.2K GitHub stars
    • Knowledge and memory
  6. Ouroboros

    MCP server

    Spec-first layer that pins requirements and verifies results of AI coding agent work

    Runs locally
    • Open source
    • 6.4K GitHub stars
    • Developer tools