/ 02. Test.Auth^{^}

Test.Auth^{^}

Tests that verify software against formal requirements, not just assumptions.

Critical industries demand a rigorous, requirements-driven approach to ensure safety, compliance, and reliability. Test.Auth^ authors complete, executable test suites directly from a requirements baseline, ensuring exhaustive coverage.

From requirement to test

Verify against formal requirements, not assumptions.

One suite, many tests by type — every one traced back to the requirement it verifies ↓

How it differs

// the old way

Implementation-coverage tests

Tests written from the code. Brittle, circular, expensive — and useless for certification.

// the new way

Requirement-driven verification.

Tests written from the spec. Pass means "the software satisfies the requirement." Robust under refactor. Compliance-ready.

// and

Your simulations, integrated.

Often, control software can only be appropriately tested against a model of the hardware or environment in which it operates. When a requirement calls for it, Test.Auth^ wires your existing simulations directly into the suites it builds — no glue code, no manual harness. And if the simulation doesn't yet exist, Test.Auth^ specifies exactly what interface, signals and fidelity the requirement needs — so the gap is defined, not guessed.

The output

—test.auth · run

$ test.auth run --suite phase-current-limit
  → loaded 7 cases (1,012 scenarios)
  → simulating · monte-carlo · n=1000
  ─────────────────────────────────────
  nominal             [ ✓ ] 0/0 fail
  boundary-below      [ ✓ ] 0/0 fail
  boundary-above      [ ✓ ] 0/0 fail
  fault-short         [ ✓ ] 0/0 fail
  fault-open          [ ✓ ] 0/0 fail
  recovery            [ ✓ ] 0/0 fail
  monte-carlo         [ ✓ ] 0/1000 fail
  ─────────────────────────────────────
  COVERAGE · requirement: 100%
  COVERAGE · MC/DC: 96.2%

Output format: MATLAB Test · C/C++ · Python · dSPACE TA-SDK
Coverage: Functional · Safe-state · Boundary Value · Error Handling
Workflow: Test Cases · Test Scripts · Executable Test Suites
Test types: SiL · HiL · Bench · Product
Traceability: Audit-ready matrix

Next stage →

SW.Syn^

SW.Syn^ synthesises software that passes every one of these tests, autonomously, but it requires tests that don't just provide a pass/fail result, but an objective score indicating how close the result is to being correct. Test.Auth^ does.

Open SW.Syn^

Output · Software model