Consort framework enforces test driven coding on live databases
Introducing Consort: A Spec-First Agent Framework for Enforced, Test-Driven Development on Live Database Branches
Software EngineeringArtificial IntelligenceDatabases
Summary
Writing code with AI agents can be tricky because the code might not always be clean or correct. The authors describe Consort, a new system that makes the AI follow strict rules and tests while working on a real copy of a database. This approach helps keep the AI’s code honest, easy to check, and easier to maintain, much like how human developers work under careful review. The system uses separate roles and controls that the AI can’t bypass to enforce quality.
What this means in practice
- •For software development teams: Run automated agents that develop database-backed applications under strict test enforcement using live database branches to ensure correctness and maintainability.
- •For enterprise database administrators: Use a controlled framework to let automated tools safely build and test changes on live database copies without risking production data integrity.
Authors
Kevin Hartman
Abstract
When an agent writes code, the development framework becomes the control system for a non-deterministic worker. Spec-first, agent-driven frameworks have gained rapid traction since 2025; the installable ones, GitHub Spec Kit, obra/superpowers, BMAD, and GSD, and our own, all capture intent through a specification or durable planning artifacts. Since they agree on capturing intent up front, what separates them is how each enforces the engineering discipline that keeps agent-written code clean, correct, and maintainable. Every framework enforces that discipline somehow; they differ in how. We characterize three modes: enforcement by persuasion (prompt discipline the model may ignore), by front-loaded structure (strong specs, then a trusted build), and through controls the agent cannot edit (a deterministic orchestrator, human-approved gates, immutable tests, and a green result that must pass against a live, branched database). We introduce Consort, a spec-first, test-driven agent framework built on the third, enforcing that discipline through controls the agent runs inside but cannot bypass, in which a deterministic orchestrator drives separate role agents through a spec-first design lane and a test-driven build lane on a live database branch. We argue that enforcing the tests and gates in code keeps agent-written code honest and verifiable, while its specialized roles, like the human roles before them, are what make it maintainable, claims we frame as a pre-registered, testable hypothesis.