DBA Coworker

AI-Powered PostgreSQL Database Administration Tool
Built for DBAs who want pgAdmin + SSMS + dbForge + DBtune — all in one tool, with AI superpowers.
AMVANA Software India Pvt. Ltd. | PouraDBA

Quick Start • Features • Architecture • Screenshots • API • Contributing

What is DBA Coworker?

DBA Coworker is a self-hosted, open-source PostgreSQL administration tool that combines the best features of pgAdmin, SSMS Activity Monitor, dbForge Studio, and DBtune into a single, AI-powered application.

Unlike traditional monitoring tools that just show you data, DBA Coworker understands your workload and actively helps you optimize, troubleshoot, and manage your PostgreSQL fleet.

Why DBA Coworker?

Traditional Tools	DBA Coworker
Show metrics, you interpret	AI analyzes your data and recommends actions
One server at a time	Fleet-wide overview of all servers
Separate tools for monitoring, tuning, backups	Everything in one place
Manual config tuning (PGTune)	Workload-aware tuning score + AI recommendations
No schema version control	Built-in schema compare with Git projects
Copy-paste SQL from docs	SQL IDE with autocomplete, formatter, visual query builder

🚀 Quick Start

Prerequisites

Node.js 18+ (nodejs.org)
PostgreSQL 14-17 server(s) to monitor
Anthropic API key (optional, for AI features) — console.anthropic.com

Installation

Option 1: Download Release (Recommended)

Download the latest release zip from GitHub Releases
Extract the zip to your preferred location (e.g., D:\dba-coworker)

Option 2: Clone from GitHub

git clone https://github.com/PouraDBA/dba-coworker.git
cd dba-coworker

Starting the Application

Windows (PowerShell)

cd D:\dba-coworker

# If you get an execution policy error, run this first (current session only):
Set-ExecutionPolicy -ExecutionPolicy Bypass -Scope Process

.\start.ps1

Windows (Command Prompt)

cd D:\dba-coworker
start.bat

Windows (Manual)

cd D:\dba-coworker\backend
npm install
node server.js

Linux / macOS

cd dba-coworker

# Install backend dependencies
cd backend
npm install

# Install and build frontend
cd ../frontend
npm install
npm run build

# Start the application
cd ../backend
node server.js

Open http://localhost:4000 in your browser.

Docker

docker-compose up -d
# Open http://localhost:4000

First-Time Setup

Open http://localhost:4000 — you'll see the welcome screen
Click "+ Register Server" — enter your PostgreSQL host, port, credentials
Click Test Connection to verify
All monitoring tabs populate with live data immediately
(Optional) Go to AI Auto-Fix tab → enter your Anthropic API key for AI features

✨ Features

DBA Coworker covers 7 pillars of database management, inspired by dbForge Studio's comprehensive approach:

                    ┌─────────────────┐
                    │  DBA Coworker   │
                    └────────┬────────┘
        ┌───────┬───────┬────┴────┬───────┬───────┬───────┐
        ▼       ▼       ▼        ▼       ▼       ▼       ▼
   ┌─────────┐┌──────┐┌───────┐┌──────┐┌──────┐┌───────┐┌──────┐
   │  SQL    ││ DB   ││ Data  ││ DB   ││  DB  ││Report ││Auto- │
   │  Dev    ││Design││ Mgmt  ││ Ops  ││Admin ││& Data ││mation│
   └─────────┘└──────┘└───────┘└──────┘└──────┘└───────┘└──────┘

🌐 Fleet Management

Feature	Description
Server Registration	pgAdmin-style — register servers with host, port, credentials, groups, colors
Fleet Overview	Scan all servers at once — online/offline status, PG version, connections, latency
Server Groups	Organize by environment (Production, Staging, Dev) with color coding
Multi-Database Switcher	Dropdown in header — switch databases without re-registering

📊 Monitoring Dashboard (22 Tabs)

Tab	Inspired By	Description
Overview	SSMS Activity Monitor	Connections, active queries, cache hit %, locks, DB size, replication — with real-time sparkline graphs
Connections	SSMS Processes	Live `pg_stat_activity` with state badges, Cancel/Kill buttons per PID
Databases	pgAdmin Dashboard	Per-database size, cache hit, commits, rollbacks, deadlocks, temp files
Tables	pgAdmin/SSMS	Size, live/dead rows, bloat %, seq/idx scans, last vacuum/analyze, one-click VACUUM/ANALYZE/REINDEX
Indexes	SSMS	Usage stats, scan counts — find unused indexes consuming disk
Objects	SSMS Object Explorer	Tables, views, materialized views, sequences, foreign tables
Functions	pgAdmin	Functions & procedures with argument signatures and return types
Replication	SSMS Always On	Slots, lag, WAL status, streaming replication monitoring
Locks	—	Blocking lock chains with blocker → blocked PID mapping
Wait Events	SSMS Resource Waits	Wait events grouped by type (Lock, IO, Client, IPC)
File I/O	SSMS Data File I/O	Block reads, hits, cache ratio, temp files, read/write times
Bloat	myDBA.dev	Dead tuple analysis with bloat % per table
Maintenance	pgAdmin	One-click VACUUM, ANALYZE, REINDEX buttons per table
Slow Queries	SSMS Expensive Queries	Top queries from `pg_stat_statements` by avg/total time
Extensions	pgAdmin	Installed extensions with versions
Roles	SSMS Security	Roles with superuser, replication, create DB, member_of
Backup/Server	SSMS Properties	Archive mode, WAL level, data directory, uptime, all DB sizes
Settings	pgAdmin	Key `postgresql.conf` parameters with restart/reload indicators

Extra on every tab: 📥 Export CSV button for all data.

📈 Real-Time Graphs (SSMS Activity Monitor-style)

Six live sparkline charts on the Overview tab, auto-refreshing every 10 seconds:

Connections — total client connections over time
Transactions/sec — TPS derived from pg_stat_database
Cache Hit % — buffer cache efficiency
Active Queries — concurrent active query count
Idle in Transaction — stale transaction detection
Lock Waits — lock contention trending

🤖 AI-Powered DBA (Claude API)

"Like having a senior DBA sitting next to you 24/7"

Feature	Tab	Description
AI Chat	💬 AI Chat	Chat about your live server — AI has full context of connections, bloat, replication, settings, extensions
Auto-Fix Engine	🤖 AI Auto-Fix	Scans 7 issue types with one-click fix buttons
Query Optimizer	🤖 AI Auto-Fix	Paste SQL → gets EXPLAIN + schemas → AI rewrites with index suggestions
SQL Generator	🤖 AI Auto-Fix	Natural language → production SQL using your actual table schemas
Incident Playbooks	🤖 AI Auto-Fix	Describe incident → step-by-step resolution with SQL
Migration Helper	🤖 AI Auto-Fix	Describe changes → ALTER TABLE with pre-checks, rollback, verification
Doc Generator	🤖 AI Auto-Fix	One-click: Runbook, Architecture Doc, or DR Plan — from live data

How AI context works:

Your PG Server → Backend collects live data → Sent as context to Claude
                  (connections, bloat, replication,     ↓
                   settings, extensions, databases)   Specific advice with
                                                      real table names,
                                                      exact SQL, actual IPs

🎯 Tuning Advisor (DBtune-style)

View	Description
📊 Tuning Score	12 automated checks scored A-F. Instant — no AI needed. Checks: cache hit ratio, connection usage, shared_buffers, effective_cache_size, work_mem, random_page_cost, io_concurrency, idle txn timeout, slow query logging, table bloat, unused indexes, checkpoint tuning.
🔬 Workload Profile	Deep analysis: OLTP/OLAP/Mixed classification, read/write %, all tunable params, index health, top queries from pg_stat_statements
🧠 AI Recommendations	Claude analyzes full workload → 🔴 CRITICAL / 🟡 IMPORTANT / 🟢 NICE-TO-HAVE with ready-to-run `ALTER SYSTEM` commands

🔀 Schema Compare (dbForge-style)

Feature	Description
Upload & Compare	Drag-drop Git project SQL files → instant diff against live database
Side-by-Side Diff	Split-panel: PROJECT (Git) vs SERVER (Live) with line-by-line highlighting
8 Object Types	Tables (with constraints, indexes), functions, views, materialized views, triggers, sequences, extensions, custom types
Status Detection	✅ Identical, 🔶 Modified, ➕ Project Only, ⚠️ Server Only
Export Schema	Extract all objects as individual SQL files for Git initialization

💾 Backup Manager (11 Backup Types)

Type	Tool
Single Database	`pg_dump` with format/compression/parallel options
Schema Only	DDL only (no data)
Data Only	COPY/INSERT only (no DDL)
Per-Object Export	Separate files for tables, functions, views, sequences, triggers
All Databases	Full script with per-DB timing and logging
Roles & Users	`pg_dumpall --roles-only`
Globals	Roles + tablespaces
Physical	`pg_basebackup` with WAL streaming
Incremental	pgBackRest setup (full/diff/incr schedule)
PITR Setup	WAL archiving + recovery guide + restore points
Restore	Commands for any format

Plus: Backup Analysis (archive mode alerts), Log Analyzer (paste log → parsed), PITR Restore Points (create before migrations).

🔗 FDW Manager

View foreign servers, user mappings, foreign tables with column-level detail
Create: Foreign servers (tds_fdw template), tables (with OPTIONS), user mappings
Import Foreign Schema: Bulk import with LIMIT TO
Script All DDL: Complete export with column OPTIONS, COLLATE, NOT NULL
Drop with CASCADE confirmation

▷ SQL IDE

Feature	Description
Autocomplete	Tables, columns (`table.`), functions, keywords. Ctrl+Space trigger.
SQL Formatter	One-click keyword uppercasing, indentation, clause separation
Visual Query Builder	Click tables → columns → auto-FK joins → WHERE/ORDER/LIMIT → generate
12 SQL Snippets	SELECT, JOIN, CTE, INSERT, UPDATE, CREATE TABLE/FUNCTION/INDEX, GRANT, EXPLAIN
EXPLAIN Visualizer	Visual plan tree with color-coded timings and buffer stats
Query History	Persistent sidebar, click to reload, saved across restarts
Export CSV	One-click on any result

❤️ Health Report

8-point automated health check: cache hit ratio, connection usage, idle-in-transaction, replication slots, deadlocks, bloated tables, long queries, uptime — scored out of 100.

🏗 Architecture

┌──────────────────────────────┐
│    React Frontend (:3000)    │
│  ┌──────────────────────┐    │
│  │ Server Tree (sidebar)│    │
│  │ 22 Monitoring Tabs   │    │
│  │ SQL IDE              │    │
│  │ Schema Compare       │    │
│  │ Backup Manager       │    │
│  │ Tuning Advisor       │    │
│  │ AI Chat / Auto-Fix   │    │
│  │ FDW Manager          │    │
│  └──────────────────────┘    │
└─────────────┬────────────────┘
              │ /api (REST)
┌─────────────▼────────────────┐
│   Express.js Backend (:4000) │
│  ┌──────────────────────┐    │     ┌────────────────┐
│  │ Connection Pool Mgr  │─────────▶│ PostgreSQL     │
│  │ Monitoring Queries   │    │     │ Servers (fleet) │
│  │ Backup Engine        │    │     └────────────────┘
│  │ Schema Extractor     │    │
│  │ Tuning Analyzer      │    │     ┌────────────────┐
│  │ FDW Manager          │─────────▶│ Claude API     │
│  │ SQL Formatter        │    │     │ (Anthropic)    │
│  │ AI Proxy             │    │     └────────────────┘
│  └──────────────────────┘    │
│  servers.json                │
│  query_history.json          │
│  api_key.json                │
└──────────────────────────────┘

Tech Stack

Component	Technology
Frontend	React 18, Vite 6, SVG charts (zero CSS/chart dependencies)
Backend	Express.js, node-postgres (`pg`)
AI Engine	Anthropic Claude API (Sonnet)
Deployment	Node.js / Docker

Zero Dependencies Philosophy

Backend: express, cors, pg — 3 packages
Frontend: react, react-dom, vite — 3 packages
No Redis, MongoDB, Elasticsearch — config in JSON files
SVG sparklines — hand-coded, no chart library
Inline styles — no Tailwind, no CSS framework

🔒 Security

Best Practices

Run inside your network — do NOT expose to the internet
Use a read-only monitoring user:

CREATE ROLE dba_coworker LOGIN PASSWORD 'secure_password';
GRANT pg_monitor TO dba_coworker;
GRANT pg_read_all_stats TO dba_coworker;

servers.json contains credentials — restrict with chmod 600
Blocked via SQL Editor: DROP DATABASE, DROP TABLESPACE, ALTER SYSTEM, COPY TO PROGRAM
All queries have 30-second timeout
Max 3 connections per server pool

📡 API Reference

Server Management

Method	Endpoint	Description
`GET`	`/api/servers`	List servers
`POST`	`/api/servers`	Register server
`PUT`	`/api/servers/:id`	Update server
`DELETE`	`/api/servers/:id`	Remove server
`POST`	`/api/servers/:id/test`	Test connection
`GET`	`/api/fleet/status`	Fleet scan

Monitoring

Method	Endpoint	Description
`GET`	`/api/servers/:id/monitor/:type`	Monitoring data (20+ types)
`GET`	`/api/servers/:id/metrics`	Time-series for graphs
`GET`	`/api/servers/:id/health-report`	Health check with score

SQL Development

Method	Endpoint	Description
`POST`	`/api/servers/:id/query`	Execute SQL
`POST`	`/api/servers/:id/explain`	EXPLAIN ANALYZE
`POST`	`/api/servers/:id/sql/format`	Format SQL
`GET`	`/api/servers/:id/sql/autocomplete`	Tables, columns, functions
`GET`	`/api/servers/:id/sql/relationships`	FK map for query builder

Tuning

Method	Endpoint	Description
`GET`	`/api/servers/:id/tuning/score`	12-point score
`GET`	`/api/servers/:id/tuning/workload`	Workload profile
`POST`	`/api/servers/:id/tuning/recommend`	AI recommendations

Schema Compare

Method	Endpoint	Description
`GET`	`/api/servers/:id/schema/extract`	Extract schema DDL
`POST`	`/api/servers/:id/schema/compare`	Compare project vs live DB
`GET`	`/api/servers/:id/schema/export-project`	Export as project files

Backup

Method	Endpoint	Description
`GET`	`/api/servers/:id/backup/analysis`	Backup analysis
`POST`	`/api/servers/:id/backup/generate-command`	Generate commands (11 types)
`POST`	`/api/servers/:id/backup/restore-point`	Create PITR point

AI

Method	Endpoint	Description
`POST`	`/api/servers/:id/ai/chat`	AI chat with server context
`POST`	`/api/servers/:id/ai/optimize`	Query optimizer
`POST`	`/api/servers/:id/ai/generate-sql`	NL → SQL
`GET`	`/api/servers/:id/ai/auto-detect`	Issue detection
`POST`	`/api/servers/:id/ai/playbook`	Incident playbook
`POST`	`/api/servers/:id/ai/migration`	Migration helper
`POST`	`/api/servers/:id/ai/documentation`	Doc generator

FDW

Method	Endpoint	Description
`GET`	`/api/servers/:id/fdw`	FDW inventory
`POST`	`/api/servers/:id/fdw/create-server`	Create foreign server
`POST`	`/api/servers/:id/fdw/create-table`	Create foreign table
`POST`	`/api/servers/:id/fdw/import-schema`	Import remote schema
`GET`	`/api/servers/:id/fdw/generate-ddl`	Export all FDW DDL

🗺️ Roadmap

✅ Completed

🗓️ Planned

🤝 Contributing

Contributions welcome! See CONTRIBUTING.md for guidelines.

# Development setup
cd backend && npm install && node --watch server.js    # API on :4000
cd frontend && npm install && npm run dev              # UI on :3000 (proxied)

Project Structure

dba-coworker/
├── backend/
│   ├── server.js              # Express API (~2,600 lines)
│   └── package.json
├── frontend/
│   ├── src/
│   │   ├── App.jsx            # React app (~2,100 lines)
│   │   └── main.jsx
│   ├── index.html
│   └── vite.config.js
├── docker-compose.yml
├── Dockerfile
├── LICENSE
└── README.md

📄 License

MIT License — see LICENSE for details.

🙏 Acknowledgments

PostgreSQL — the world's most advanced open-source database
Anthropic Claude — AI engine powering the coworker features
Inspired by: pgAdmin, SSMS, dbForge Studio, DBtune, myDBA.dev

Built with ❤️ by PouraDBA

Name		Name	Last commit message	Last commit date
Latest commit History 3 Commits
backend		backend
docs/images		docs/images
frontend		frontend
.gitignore		.gitignore
CHANGELOG.md		CHANGELOG.md
CONTRIBUTING.md		CONTRIBUTING.md
Dockerfile		Dockerfile
LICENSE		LICENSE
README.md		README.md
SECURITY.md		SECURITY.md
docker-compose.yml		docker-compose.yml
start.bat		start.bat
start.ps1		start.ps1

Folders and files

Latest commit

History

Repository files navigation

DBA Coworker

What is DBA Coworker?

Why DBA Coworker?

🚀 Quick Start

Prerequisites

Installation

Option 1: Download Release (Recommended)

Option 2: Clone from GitHub

Starting the Application

Windows (PowerShell)

Windows (Command Prompt)

Windows (Manual)

Linux / macOS

Docker

First-Time Setup

✨ Features

🌐 Fleet Management

📊 Monitoring Dashboard (22 Tabs)

📈 Real-Time Graphs (SSMS Activity Monitor-style)

🤖 AI-Powered DBA (Claude API)

🎯 Tuning Advisor (DBtune-style)

🔀 Schema Compare (dbForge-style)

💾 Backup Manager (11 Backup Types)

🔗 FDW Manager

▷ SQL IDE

❤️ Health Report

🏗 Architecture

Tech Stack

Zero Dependencies Philosophy

🔒 Security

Best Practices

📡 API Reference

Server Management

Monitoring

SQL Development

Tuning

Schema Compare

Backup

AI

FDW

🗺️ Roadmap

✅ Completed

🗓️ Planned

🤝 Contributing

Project Structure

📄 License

🙏 Acknowledgments

About

Topics

Resources

License

Contributing

Security policy

Uh oh!

Stars

Watchers

Forks

Releases

Packages 0

Uh oh!

Uh oh!

Contributors

Uh oh!

Languages

Packages