MCP Server for ArangoDB

What is MCP Server for ArangoDB?

MCP Server for ArangoDB is a TypeScript-based Model Context Protocol server that provides database interaction capabilities through ArangoDB. It implements core database operations and works with Claude Desktop, VSCode extensions like Cline, OpenCode, and other MCP-compatible clients.

How to use MCP Server for ArangoDB?

Install globally via npm install -g arango-server or run directly with npx -y arango-server. Configure environment variables (ARANGO_URL, ARANGO_DB, ARANGO_USERNAME, ARANGO_PASSWORD, optional ARANGO_BACKUP_ROOT) and add the server configuration to your MCP client (e.g., Claude Desktop, VSCode MCP settings, OpenCode). Then prompt your AI agent to execute any of the 30+ exposed tools.

Key features of MCP Server for ArangoDB

• Execute AQL queries with guardrails and bind variables
• Perform read-only queries with write/DDL keyword rejection
• Discover database schema: collections, indexes, views, graphs
• Create, update, insert, and delete documents and collections
• Build and query ArangoSearch Views with BM25 ranking
• Traverse graphs and find shortest paths between vertices
• Backup collections to JSON files under a configurable root directory

Use cases of MCP Server for ArangoDB

• Let an AI agent explore and query an ArangoDB database using natural language
• Automate schema discovery and documentation by listing collections, indexes, and fields
• Build AI-assisted CRUD workflows for development databases
• Perform graph traversal and pathfinding queries without writing raw AQL
• Create ArangoSearch analyzers and views for full-text search tasks

FAQ from MCP Server for ArangoDB

What tools does MCP Server for ArangoDB provide?

It offers over 30 tools covering query execution (with read-only and validate variants), collection management, document CRUD, index creation, ArangoSearch view creation and search, graph operations, and database backup.

Is MCP Server for ArangoDB read-only?

No. The server includes both read-only tools (e.g., arango_read_query, arango_describe_database) and mutating tools (e.g., arango_insert, arango_create_collection, arango_drop_collection). Read-only tools reject write/DDL AQL keywords.

What environment variables are required?

ARANGO_URL, ARANGO_DB, ARANGO_USERNAME, and ARANGO_PASSWORD are required. ARANGO_BACKUP_ROOT is optional and defaults to ./backups; it defines where backup output is written.

Is this server intended for production use?

No. The README explicitly states it is designed for local development environments only and discourages connecting to production databases due to security risks.

How does the server handle backup file paths?

The arango_backup tool rejects absolute paths, path traversal, symlink escapes, and existing output files to mitigate arbitrary file write risks. Output is always written under the ARANGO_BACKUP_ROOT directory.