Five Tasks, Two Modes, One Model: A Controlled Experiment
We gave Claude the same 5 backend tasks twice — once with a plain Express prompt, once with SysMARA specs. Then we counted the invariant violations. 50% vs 0%.
Read moreBlog
Thoughts on AI-native development, machine-readable architecture, and building systems that AI agents can safely operate.
Constraint Visibility in Practice: A Multi-Module Case Study
We built an e-commerce system with 2 modules, 3 invariants, 1 policy, and 1 flow — then traced exactly how SysMARA makes every constraint visible to AI agents.
Read more
Building a TODO API with MySQL from Scratch
Step-by-step tutorial: create a fully working CRUD API backed by MySQL using SysMARA. From empty directory to running server in under 5 minutes.
Read more
Implement Mode: When Your Scaffold Writes Real Code
SysMARA v0.6.2 changes scaffold generation from TODO stubs to working ORM-based implementations. Capability handlers, policy enforcers, and invariant validators now come with real logic out of the box.
Read more
Zero to Database in One Command: SysMARA's New Init Pipeline
SysMARA v0.6.2 integrates database selection, Docker environment, and environment files directly into the init command. One command gives you a fully configured project with a running database.
Read more
SysMARA ORM: The Database Layer AI Agents Can Actually Read
Traditional ORMs hide SQL behind abstractions designed for humans. SysMARA ORM does the opposite — every query is scoped to a capability, every operation is logged as structured JSON, and migrations know which invariants they affect.
Read more
From Specs to Code: Why SysMARA Now Scaffolds Your Implementation
SysMARA v0.5.1 introduces the scaffold generator: YAML specs in, typed TypeScript implementation files out. No more empty app/ directories.
Read more
The Path Bug That Mirrored Your Filesystem
How a double path-join bug in the capability compiler caused generated files to recreate the entire filesystem tree inside app/generated/. A postmortem on path handling.
Read more
Why Traditional Frameworks Hide Too Much for AI Agents
Convention over configuration breaks down when the consumer is not human. Decorators, middleware chains, and implicit routing create blind spots that AI agents cannot navigate.
Read more
What Machine-Readable Architecture Actually Means
Machine-readable does not mean "has an API" or "uses JSON." It means the system's architecture is a formal, queryable artifact that tools and agents can traverse.
Read more
Why Invariants Should Be First-Class Citizens
Business rules scattered across middleware, services, and database constraints are invisible to AI agents. Making invariants named, described, and bound to entities changes everything.
Read more
Safe Edit Zones: Teaching AI Agents Where They Can and Cannot Write
AI agents will modify any file you let them. Safe edit zones create explicit file ownership boundaries that prevent accidental damage without blocking productive work.
Read more
Capability-First System Design
Instead of starting with routes or controllers, start with capabilities: named operations with typed inputs/outputs, entity bindings, policy gates, and invariant checks.
Read more