Baseline (measured against the current post-split code, agent-comms CI run 34813979900): 60.19% (43 killed, 19 timeout, 41 survived, 104 errors, out of 104 covered mutants).
Trace every surviving/timeout mutant against the real call graph: a genuine gap gets a real regression test; a provably equivalent/dead mutant gets documented reasoning, not a contrived test. Target the repo's own configured thresholds.high: 80 in stryker.config.ts.
Sub-issue of #64.
Baseline (measured against the current post-split code, agent-comms CI run 34813979900): 60.19% (43 killed, 19 timeout, 41 survived, 104 errors, out of 104 covered mutants).
Trace every surviving/timeout mutant against the real call graph: a genuine gap gets a real regression test; a provably equivalent/dead mutant gets documented reasoning, not a contrived test. Target the repo's own configured
thresholds.high: 80instryker.config.ts.Sub-issue of #64.