Overview

StackQL lets you work with cloud, SaaS and API resources using SQL statements instead of cloud‑specific SDKs or configuration files. The engine is open source, runs as a single binary or in Docker, and can be embedded in applications or exposed as a PostgreSQL‑compatible server for BI tools. It is positioned as a “universal interface” for both human operators and AI agents that need to read or change infrastructure.

The tool supports the usual SQL verbs, SELECT, INSERT, UPDATE, DELETE, SHOW and DESCRIBE, mapped to cloud‑provider actions. A SELECT query can retrieve live state, for example counting EC2 instances by status, while an INSERT statement can provision a new Google Compute disk in a single line. Because there is no state file and no per‑cloud SDK required, you can start with a simple grammar and later combine StackQL with existing IaC pipelines. The documentation mentions both declarative and procedural styles, so you can write one‑off queries or embed the engine in larger automation scripts.

StackQL is aimed at platform‑engineering and SRE teams that manage multi‑cloud estates, as well as security, FinOps and data‑engineering groups that need up‑to‑date inventory, compliance data or cost metrics. Developers building AI agents also have a dedicated Model Context Protocol (MCP) server that gives those agents the same SQL interface. The engine runs on macOS, Linux and Windows, can be invoked from CI pipelines, cloud shells, or as a service reachable by BI notebooks. Installation options include a direct binary, Docker image, package manager, or cloud‑shell scripts, and the project provides command‑line subcommands such as `exec`, `shell`, `srv`, `mcp` and `registry`.

Key features

SQL‑based CRUD operations mapped to cloud, SaaS and API resources
Live, read‑only queries that return current provider state without a state file
Support for multiple providers with separate reference sites for each
MCP server that exposes the same SQL surface to AI agents
Runs as a single binary, Docker container, or embedded PostgreSQL‑compatible server
Open‑source MIT license with community support via GitHub Discussions and Discord

Details

Pricing: Free
Category: Agents

Suggest an edit

Sign in to suggest an edit.

Pros and cons

Pros

No need to manage per‑cloud SDKs or Terraform‑style state files
Unified grammar works across major cloud providers and SaaS APIs
Can be used both interactively and as part of CI/CD pipelines
Provides a PostgreSQL‑compatible endpoint for existing BI tools

Cons

Pricing information is not published; commercial support requires direct inquiry
Documentation does not list a full catalog of supported providers in this excerpt
Advanced policy‑binding or multi‑writer conflict resolution is mentioned but not detailed.