Agent♥︎Age
Catalog

AgentSQL

Official

by cuipengdba · Go

Guard AI agents' PostgreSQL/MySQL access via MCP: SQL audit, auth, masking, write approval

io.github.cuipengdba/agentsql MCP Server

The io.github.cuipengdba/agentsql MCP server provides access control for Guard AI agents’ interaction with PostgreSQL and MySQL. It focuses on enforcing SQL audit, authentication, data masking, and requiring write approval for database operations.

🛠️ Key Features

  • SQL audit for agent database activity
  • Authentication controls for PostgreSQL/MySQL access
  • Data masking
  • Write approval for write operations

🚀 Use Cases

  • Regulated or monitored agent workflows needing database visibility
  • Environments requiring authenticated access to PostgreSQL and MySQL
  • Scenarios where sensitive data must be masked
  • Workflows that require approval before agents perform database writes

⚡ Developer Benefits

  • Centralized control over agent SQL access to PostgreSQL/MySQL
  • Auditable and restricted database interactions
  • Support for masking and approval-based write governance

⚠️ Limitations

  • Scope is explicitly PostgreSQL/MySQL access for Guard AI agents; other databases are not described.

Topics

ai-agentdata-maskingdatabase-securitymcpmcp-servermodel-context-protocolmysqlpostgresqlsecurity-gatewaysql-audit

Related servers

More in Security