← Back to runbooks
Sprint 16.5 Controlled Intelligence

Runbook 16.5 — Action Method Synthesis

Convert validated locators into typed elements, primitive action methods, and high-confidence composite actions. This sprint creates the bridge between locator intelligence and safe page-object / test-skeleton generation.

What this sprint proves

  • Validated locators can be typed into meaningful UI intents
  • Primitive methods can be generated deterministically
  • Composite workflows can be synthesized only when confidence is high
  • Bounded generation is safer than full test guessing

Core output chain

  • locator-map.json
  • element-type-map.json
  • action-method-candidates.json
  • composite-action-candidates.json

What gets generated

Primitive actions

  • fillUsername(value)
  • fillPassword(value)
  • submitLogin()
  • clickPublish()

Composite actions

  • loginAs(username, password)
  • Only when intent evidence is strong
  • Only when method confidence clears threshold

Why this matters

Most automation systems either stop at locator discovery or jump too quickly into unsafe full-test generation. This sprint takes the middle path: turn validated system understanding into reusable methods without pretending the system knows more than it actually does.

Who this is for