.ai/policies/semantic-compliance.toon

semanticCompliance:
  version: 1
  principle: semantic-compliance-over-literal-compliance
  strategyIdentity: hypothesis, expectedEvidence, targetFailure, intendedOutcome
  progress: requirement-completed, blocker-removed, acceptance-verified, relevant-test-fixed, deliverable-completed, uncertainty-reduced
  depthFollows: parent, causedBy, delegatedFrom
  depthIgnores: label, agent, counterReset
  completion: mandatoryEvidence, requiredTests, resolvedRequiredFindings
  blocker: condition, evidence, boundedRecovery, exhaustedAlternatives, missingInput
  override: reason, evidence, limit, scope, exitCondition