Reproduce The Loop
cargo run -p veritas-cli -- --root examples/go-evolution-loop verify --lang go --target .
cargo run -p veritas-cli -- --root examples/go-evolution-loop score
cargo run -p veritas-cli -- --root examples/go-evolution-loop evolve --dry-run
Initial Signal
Mutation
Generated 14, executed 12, killed 7, survived 4, score 58%.
Fuzzing
Generated harnesses 1, targets executed 3, failures 0, persisted repros 4.
Differential Replay
Targets 3, cases 16, with replay artifacts available to turn behavior drift into assertions.
Evolution
Suites 1, candidates 14, selected 12, average fitness 76%.
Confidence
Score 55, grade Medium. The report is useful, but not done.
Top Candidate
Add the smallest assertion that kills a surviving ParseInvoiceTotal mutant. Fitness 95%.
The Assertion
func TestParseInvoiceTotal(t *testing.T) {
value, ok := ParseInvoiceTotal(" 10_000 ")
if !ok || value != 10000 {
t.Fatalf("expected normalized total, got %d ok=%v", value, ok)
}
value, ok = ParseInvoiceTotal("1000000")
if !ok || value != 1000000 {
t.Fatalf("expected boundary invoice total, got %d ok=%v", value, ok)
}
value, ok = ParseInvoiceTotal("not-a-number")
if ok || value != 0 {
t.Fatalf("expected invalid invoice total to be rejected with zero value, got %d ok=%v", value, ok)
}
value, ok = ParseInvoiceTotal("1000001")
if ok || value != 0 {
t.Fatalf("expected huge invoice total to be rejected with zero value, got %d ok=%v", value, ok)
}
}
After The Candidate Lands
cargo run -p veritas-cli -- --root examples/go-evolution-loop verify --lang go --target .
cargo run -p veritas-cli -- --root examples/go-evolution-loop score
- Mutation score rises to 91%.
- Surviving mutants drop to 0.
- Differential replay retains 16 cases.
- Evolution selected candidates drop to 0.
- Confidence score rises to 98, grade High.