Skip to content

pgEdge AI DBA Workbench

An AI-powered environment for monitoring, managing, and troubleshooting PostgreSQL systems.

Overview

The pgEdge AI DBA Workbench combines a Model Context Protocol (MCP) server with a web-based user interface, data collector, and alert monitoring service. The Workbench enables users to query, analyze, and manage distributed PostgreSQL clusters using natural language and intelligent automation. The system exposes pgEdge tools and data sources to both cloud-connected and locally hosted language models, ensuring full functionality in air-gapped or secure environments.

Architecture

The AI DBA Workbench consists of four components that work together to provide monitoring, alerting, and AI-powered database management.

Data Collection Layer

The Collector continuously monitors PostgreSQL servers and collects metrics into a centralized datastore. The Collector provides the following features:

  • The Collector supports multi-server monitoring with independent connection pools.
  • 34 built-in probes cover PostgreSQL system views.
  • Automated data management handles partitioning and retention policies.
  • Secure connections use encryption and SSL/TLS support.

Intelligence Layer

The MCP Server implements the Model Context Protocol and provides AI assistants with standardized access to PostgreSQL systems. The MCP Server provides the following features:

  • The server uses HTTP/HTTPS transport with JSON-RPC 2.0.
  • SQLite-based authentication supports users and tokens.
  • Role-based access control manages groups and privileges.
  • Database tools enable queries, schema introspection, and analysis.
  • The LLM proxy supports Anthropic, OpenAI, Gemini, and Ollama providers.
  • Conversation history management preserves chat context.

Alert Monitoring Layer

The Alerter evaluates collected metrics against thresholds and uses AI-powered anomaly detection to generate alerts. The Alerter provides the following features:

  • Threshold-based alerting supports configurable rules.
  • Tiered anomaly detection uses statistical analysis, embeddings, and LLM classification.
  • Blackout scheduling accommodates maintenance windows.
  • Notification delivery integrates with email, Slack, Mattermost, and webhooks.

Presentation Layer

The Client provides a web-based user interface for cluster monitoring and management. The Client provides the following features:

  • Hierarchical dashboards display estate, cluster, and server metrics.
  • Cluster topology visualization shows replication edges.
  • The AI-powered chat interface supports natural language queries.
  • The administration panel manages users, groups, and tokens.

Where to Start

Choose a section based on your role and goals.

  • The Quick Start guide helps new users set up the Workbench for the first time.
  • The User Guide covers dashboards, alerts, and AI features for day-to-day usage.
  • The Administrator's Guide explains authentication, connections, and system configuration.
  • The Developer's Guide provides architecture details and contribution guidelines for each component.

System Requirements

The Workbench requires the following software and resources.

  • PostgreSQL 14 or later is required for the datastore.
  • Go 1.24 or later is required to build server-side components from source.
  • Node.js 18 or later is required to build the web client from source.
  • Network connectivity must exist between all components.
  • Database credentials must have appropriate permissions.

License

Copyright (c) 2025 - 2026, pgEdge, Inc.

This software is released under the PostgreSQL License.