← Back to directory
COMPARE UP TO 4 SERVERS

Compare MCP servers

Compare scores, permissions, risks, and fit in one decision-focused table.

DimensionDirectus Content MCP ServerLet LLMs interact directly with your Directus data.MCP Toolbox for DatabasesGoogle's official database MCP toolbox — define AI-accessible database tools via config
FMRS70 / 100 · B74 / 100 · B
Reliability10 / 2013 / 20
Security and permissions15 / 2013 / 20
Maintenance12 / 2017 / 20
Documentation18 / 2016 / 20
Setup experience15 / 2015 / 20
Best for
  • Content teams already on Directus who want to read and write content via LLMs
  • Analysts querying Directus data with natural language
  • Directus projects wanting controlled, guardrailed data access for non-developers
  • Teams that want precise control over which database operations an AI can perform, rather than open arbitrary SQL execution
  • Scenarios needing a unified MCP setup across multiple database engines
Not for
  • Projects not using Directus
  • Developers needing full destructive/admin operations on dev instances (field and collection deletion is intentionally limited)
  • Clients that do not support MCP
  • Lightweight cases that just want to run a few ad-hoc SQL queries without maintaining a tools.yaml config (a simpler single-database MCP may be a better fit)
Required permissions
  • Requires the Directus instance URL plus a static token or email/password credentials
  • Actions are limited by the permissions of the associated Directus user
  • Database credentials (username/password/connection string) are supplied via env vars or config
  • A tool's actual permission is whatever SQL statement is defined in tools.yaml — designed for least privilege, but misconfiguration can still over-expose access
Risks and side effects
  • Has write access to create, update, and delete items and fields, so misconfiguration can still modify data
  • Leaked credentials grant LLM access to Directus data
  • While field/collection deletion is limited, item deletion remains available
  • If tools.yaml defines SQL statements that allow unconstrained writes or deletes, the AI could accidentally modify data
  • The prebuilt toolsets (--prebuilt) favor convenience and may expose broader query capability than a specific business actually needs — use a custom tools.yaml in production
Supported clientsClaude Desktop, CursorClaude Code, Gemini CLI, Zed, Antigravity
Tools210