ArcadeDB
Give your agent SQL, Cypher, Gremlin, and GraphQL access to one multi-model ArcadeDB instance through its built-in MCP server.
Overview
ArcadeDB is a MCP server for the Build phase that lets agents query a multi-model database via SQL, Cypher, Gremlin, and GraphQL.
What is this MCP server?
- Multi-model database: document, graph, and relational-style access in one engine
- Built-in MCP server for agent-driven querying
- Query languages: SQL, Cypher, Gremlin, and GraphQL
- Open-source GitHub repository ArcadeData/arcadedb (schema version 26.3.1)
- Catalog server version 26.3.1
- 4 query interfaces documented: SQL, Cypher, Gremlin, GraphQL
- Built-in MCP server on multi-model engine
Community signal: 931 GitHub stars.
What problem does it solve?
Juggling separate graph, document, and SQL stores plus custom agent adapters slows backend iteration when one engine could expose everything over MCP.
Who is it for?
Builders shipping data-heavy SaaS or APIs who want graph and document models together with agent-friendly MCP access.
Skip if: Static landing-page-only projects, teams that only need a hosted Postgres with no graph layer, or browser-scraping workflows.
What do I get? / Deliverables
After you run ArcadeDB and register its MCP server, agents can work across multiple query languages on one database from a standard protocol.
- Agent-accessible SQL, Cypher, Gremlin, and GraphQL operations against ArcadeDB
- Unified multi-model backend without separate MCP shims per language
Recommended MCP Servers
Journey fit
Choosing and wiring a primary data store happens while building the product backend, before you optimize launch distribution. ArcadeDB is a database with native MCP—core backend persistence and query surfaces, not marketing or error paging.
How it compares
Built-in database MCP with four query languages, not a browser automation or finance router.
Common Questions / FAQ
Who is ArcadeDB for?
Developers and agent users who need a multi-model database and want MCP-native querying without writing a custom database adapter.
When should I use ArcadeDB?
During backend build when your product needs SQL, graph (Cypher/Gremlin), or GraphQL against one datastore and you want agents to help explore and query it.
How do I add ArcadeDB to my agent?
Deploy ArcadeDB from the official repository, enable its MCP server per project docs, then add the MCP entry in Claude Code, Cursor, or your client with connection settings for your instance.