What it does
ArcadeDB is a multi-model database management system with graph, document, key/value, search, time-series, vector embedding and geospatial models. It supports ACID transactions, full-text indexing, geospatial queries, more than 70 built-in graph algorithms, materialized views, parallel query execution and hash indexes. Its built-in MCP server lets AI assistants and LLMs connect to the database.
It suits projects that need graph traversal, vector similarity and time-series data in one database, or that want to reuse existing drivers, such as PostgreSQL, MongoDB, Redis, HTTP/JSON or Python bindings. The publisher lists examples including recommendation engines, knowledge graphs, Graph RAG, fraud detection, real-time analytics, social network analytics and supply chain visibility. For Claude Code, a ready-made plugin is available.
It can be started as a Docker container, which serves the Studio web interface on port 2480 with a sample OpenBeer database. Alternatively, a release can be downloaded and started with bin/server.sh or bin/server.bat on Windows. Docker and Kubernetes deployment are supported. The language requires Java 21 or later.
Limitations and requirements
Redis and MongoDB driver access implements only a subset of operations. The AI assistant in Studio is in Beta.
Safety
Our team has not reviewed this resource. Only automated rules checked it; resources with high-risk signals are not published. These are signals, not a guarantee that installing it is safe. Try it in a sandbox project first.
- Runs code or changes an assistant's behaviour
Sources
- Official MCP Registry first seen 13 Nov 2026 · last seen 13 Nov 2026 · its page on the source
ArcadeDB
Multi-Model database with built-in MCP server for SQL, Cypher, Gremlin, and GraphQL.
v26.3.1As shown on registry.modelcontextprotocol.io when we last read it