PostgreSQL extension for Visual Studio Code

The PostgreSQL extension for Visual Studio Code (ms-ossdata.vscode-pgsql) brings database development tools directly into your editor. You can connect to local, on-premises, or cloud-provided PostgreSQL servers, write queries with IntelliSense, explore database objects, visualize query plans and schemas, and manage Azure Database for PostgreSQL flexible server resources, all without leaving VS Code. The extension also integrates with AI assistants for AI-assisted query authoring, analysis, and agent-driven workflows.

Get started

If you're new to the extension, start with one of these guides:

Install the extension

1. Open Visual Studio Code.
2. Open the Extensions view (Ctrl+Shift+X, or Cmd+Shift+X on macOS).
3. Search for postgresql.
4. Select PostgreSQL by Microsoft and then select Install.

Key features

The following sections summarize each major feature area. Select the link at the end of each section for the full article.

Connection dialog

The connection dialog connects you to local, on-premises, and cloud-provided PostgreSQL servers, with integrated authentication support for Microsoft Entra ID and AWS IAM (RDS/Aurora). Enter connection parameters individually, paste a connection string, or browse your Azure subscriptions to discover Azure Database for PostgreSQL flexible servers and Azure HorizonDB (Preview) clusters.

• Three input modes: Parameters for individual values such as host, port, database, and username; Connection String for a full connection string; or Browse Azure to navigate your Azure assets and select a resource with a PostgreSQL endpoint.
• Saved and recent connections: saved profiles appear in the Connections list for quick reuse; recently used connections are always available.
• Server groups: organize connections into named groups with color coding to quickly identify connections in the Connections tree.

For details, see Connections and identity and Advanced connection options.

Object explorer

The object explorer provides a navigable tree view of your connected PostgreSQL servers. Expand servers, databases, and schemas to inspect tables, views, functions, sequences, types, extensions, roles, tablespaces, and event triggers.

• Search: find objects by name across all supported object types with the built-in search.
• Scripting: right-click any object to generate SELECT, CREATE, ALTER, DROP, or EXECUTE scripts that open in the query editor.
• Deep refresh: refresh any node to recursively update all expanded descendants, reflecting schema changes without disconnecting.

For details, see Object explorer.

Query editor and IntelliSense

The query editor provides a rich editing surface for PostgreSQL SQL. IntelliSense offers context-aware completions for table names, column names, functions, schemas, keywords, and join clauses.

• Run Query and Run Current Statement: run the full editor contents or only the statement at the cursor.
• Code snippets: insert common SQL patterns such as CREATE TABLE, INSERT, and SELECT.
• Query history: access previously run queries from the Command Palette to rerun or modify them.

For details, see Query editor and IntelliSense.

Query results

Results appear in the PostgreSQL Query Results panel below the editor. The results grid supports sorting, filtering, and column resizing.

• Export: save results to CSV, JSON, or Excel.
• Chart visualization: visualize results as line, bar, pie, or scatter charts.
• Graph view: Apache AGE graph-query results switch automatically to a node-and-edge view.
• Messages tab: review row counts, execution times, and server messages.

For details, see Query editor and IntelliSense.

Query plan visualizer

The query plan visualizer presents EXPLAIN output in multiple interactive formats so that you can identify performance bottlenecks.

• Four views: Tree View, Icicle View, Table View, and Source View.
• Metrics: inspect cost estimates, actual row counts, execution time, and buffer usage per node.
• Import plans: visualize saved plan files or editor content without a live database connection.
• Copilot analysis: select Analyze with Copilot (or Analyze with AI in Cursor) for an AI-generated explanation of the plan.

For details, see Query plan visualizer.

Schema visualizer

The schema visualizer generates an interactive entity-relationship diagram. Tables appear as draggable nodes with columns listed inside, and foreign key relationships display as connecting edges.

• Color-coded schemas: each schema receives a distinct color for quick identification.
• Legend toggle: show or hide the schema legend to maximize diagram space.

For details, see Schema visualizer.

Server dashboard

The server dashboard shows connection properties, server version, and SSL status for any connected server. For Azure Database for PostgreSQL flexible servers, it also displays Azure Monitor metrics. Investigation tabs let you drill into specific areas of server activity.

• Overview: monitor CPU, memory, storage, connections, IOPS, and other metrics grouped by category.
• Queries: view top SQL statements ranked by execution time, call count, or other metrics, with drill-down into individual query details.
• Waits: analyze wait events with a ranked table and over-time chart.
• Sessions: inspect active sessions, blocking trees, and lock activity.
• Toolbar actions: connect, disconnect, open a new query editor, launch the schema visualizer, open an AI chat, or start Azure management actions from the dashboard toolbar.

For details, see Server dashboard.

Azure server management

Manage Azure Database for PostgreSQL flexible server resources without leaving VS Code.

• Start, stop, and restart: control the server lifecycle from the Connections tree or the dashboard.
• Firewall rules: view and update network access rules.
• Parameters: browse and modify parameters.
• Clone server: duplicate an existing server configuration.

For details, see Azure server management.

Server creation

Provision new PostgreSQL servers directly from VS Code.

• Docker: create a local PostgreSQL server in a Docker container for development and testing.
• Azure Database for PostgreSQL flexible server: walk through a guided setup that configures region, compute tier, storage, and authentication.
• Azure HorizonDB (Preview): provision a cloud-native, highly scalable PostgreSQL cluster with configurable vCores and optional AI capabilities.

For details, see Create a PostgreSQL server.

Copilot integration

The extension integrates with AI assistants across chat, editor actions, and multi-step tool workflows.

• @pgsql chat participant: ask natural-language questions about your databases, generate queries, or get help with PostgreSQL concepts in the GitHub Copilot Chat panel. Visual Studio Code only. For details, see @pgsql chat participant.
• Editor actions: right-click SQL in the editor to explain a query, rewrite it for optimization, or analyze its execution plan from the AI Query Actions submenu. Available in both Visual Studio Code and Cursor.
• Agent mode: the AI agent invokes PostgreSQL tools directly, combining AI reasoning with live database operations. Available in both Visual Studio Code and Cursor. For details, see Agent mode.

For an overview of all AI capabilities, see Copilot integration.

MCP server

The extension registers a Model Context Protocol (MCP) server that exposes PostgreSQL tools, including connection management, schema exploration, query execution, and query plan visualization, to supported AI-enabled hosts. External AI tools can interact with your PostgreSQL connections programmatically through this registration.

For details, see MCP server.

Use psql in the terminal

Open a psql terminal session connected to any database in the Connections tree, or run .sql files through psql directly from the editor context menu. This gives you full access to native psql features such as \ meta-commands, COPY, and interactive scripting.

For details, see Use psql in the terminal.

Oracle to PostgreSQL migration

An AI-assisted migration workflow helps you convert Oracle database schemas and objects to PostgreSQL-compatible SQL.

• Guided migration: connect to an Oracle source, select objects to migrate, and review the converted output.
• AI-assisted conversion: A Microsoft Foundry model translates Oracle-specific syntax, data types, and procedural code to PostgreSQL equivalents.

For details, see Oracle to PostgreSQL migration.

Supported operating systems

Feedback and support

To report a bug or request a feature, file an issue on the vscode-pgsql GitHub repository.

Related content

• Quickstart: Connect and query PostgreSQL
• Connections and identity
• Query editor and IntelliSense
• Copilot integration