SQL Manager for PostgreSQL 7.0: meet the AI Assistant

This is a major release, and the headline feature is our new AI Assistant. We're starting to roll out AI capabilities across our database tools, and SQL Manager for PostgreSQL is the first to get them. It's also available in SQL Management Studio for PostgreSQL, which ships with SQL Manager.
You'll find all the AI Assistant features you'd expect: write a query from a plain-language description, explain existing SQL, fix errors, optimize, make sense of an EXPLAIN plan.
But you can also take on bigger jobs:
Here's the key part: you decide how much of the schema the model sees, down to a single table. You stay in full control of exactly what context goes to the model.
Connect as many providers as you like — all it takes is an API key (stored encrypted), and you can switch models right in the chat. Need to stay completely self-contained? It works with Ollama, so your data never leaves your machine.

Facts Only

The AI Assistant is featured in SQL Manager for PostgreSQL and SQL Management Studio for PostgreSQL.
AI features include: writing queries from plain language descriptions, explaining SQL, fixing errors, optimizing, and interpreting EXPLAIN plans.
Users control the schema context, selecting how much of the schema the model sees, down to a single table.
Connectivity allows linking to various providers using API keys.
It supports local operation via Ollama.

Executive Summary

A new AI Assistant is being rolled out across database tools, starting with SQL Manager for PostgreSQL and SQL Management Studio for PostgreSQL. This assistant offers capabilities such as writing queries from plain language, explaining existing SQL, fixing errors, optimizing code, and interpreting EXPLAIN plans. Users maintain control over context by specifying how much of the schema the model views, down to individual tables. The system supports connecting to various providers via API keys and allows switching models in the chat. For a self-contained operation, it functions with Ollama.

Full Take

The introduction of an AI layer into core database tools establishes a new paradigm for interacting with complex data infrastructure, shifting focus from procedural command entry to semantic instruction. The explicit control provided over schema context—allowing users to define the scope down to single tables—introduces a critical dimension of agency: balancing powerful generative capabilities with necessary security and data confinement. The ability to integrate multiple external models and local execution environments (Ollama) suggests a push toward distributed, user-controlled intelligence rather than centralized, proprietary solutions. The pattern observed is an attempt to democratize advanced analytical power while simultaneously asserting control over the scope of that power. The implication for agency lies in whether this level of granular control translates into genuine cognitive sovereignty or if it merely masks new layers of complexity behind familiar interface features. What structures govern the implicit choice of context, and what are the real-world costs associated with the self-imposed limits on model knowledge?

Sentinel — Human

Confidence

The text reads like a product announcement designed for clarity, exhibiting strong internal coherence typical of direct communication rather than synthetic generation.

Signals Detected

Moderate sentence length variance; rhythm is engaging rather than perfectly uniform.

Clear, direct feature description with a conversational but precise tone; no excessive hedging.

Flow is logical (feature introduction -> specific functions -> advanced control -> integration), suggesting deliberate structure rather than rote template matching.

Claims about technical capabilities (SQL features, model interaction, Ollama compatibility) are presented as product announcements, which is standard for tech communication.

Human Indicators

Use of direct, slightly informal framing ('Here's the key part:') suggests a human voice aiming to emphasize a specific point within a technical announcement.

The flow transitions smoothly between high-level features and specific implementation details without resorting to boilerplate hedging.