Dispatches from the missing layer.

The execution layer of the software supply chain was built for same input, same output. Once models enter the loop, non-determinism is not an edge case but a property of the system, and the graph needs a different execution model.

LangChain pipelines written in 2023 don’t run in 2026. The code is fine. The abstractions underneath it aren’t. The model API changed, the retrieval pattern changed, the wrapper library rewrote its surface twice. The pipeline is a fossil of dependencies that were never stable. This is not a library quality problem — it is the absence of a layer.

Observability has three mature output formats: logs, metrics, traces. All three answer what happened. None of them answer why. That answer has always lived somewhere else — in a dashboard, in a runbook, in the head of the engineer who got paged at 3am. For thirty years the stack got away with this, because the consumer was human. When the consumer is an AI agent, the whole model breaks.

An agent writes a function, then writes a test for that function. The same brain wrote both sides. Luotain breaks the loop by making the target a black box — you describe what your software should do in markdown, an agent probes the live system from the outside, and reports what matches and what doesn't.