Deterministic chaining

Lint passed, so… proceed? That's not a decision — it's a computation. Praxec runs the computable steps itself and wakes the model only when there's a genuine choice.

Three round trips for zero decisions

A pipeline of lint → test → build → deploy has exactly one decision in it: whether to deploy. The first three steps either pass or fail. But a naive agent reads each result, reasons about a “choice” that doesn’t exist, and burns a full round trip each time — latency and tokens spent rubber-stamping outcomes.

Tag the computable steps; the kernel chains them

Every transition carries an actor — it says who performs the step: code (actor: deterministic), the model (actor: agent), or a person (actor: human). Mark the computable steps deterministic and the kernel runs them back-to-back on its own, threading each result into the next, and only surfaces to the model when it reaches a step that needs judgment.

states:
  lint:
    transitions:
      run_lint:  { target: test,  actor: deterministic }
  test:
    transitions:
      run_tests: { target: build, actor: deterministic }
  build:
    transitions:
      package:   { target: ready, actor: deterministic }
  ready:
    transitions:
      deploy:    { target: live,  actor: agent }   # the chain stops here

The model calls praxec.command once and the response arrives at ready — three real commands run, zero LLM round trips, with a chain trace attached. The model is only consulted where a decision actually lives: whether to deploy.