diff --git a/packages/loopover-engine/README.md b/packages/loopover-engine/README.md
index da979fa75..0380cf6f3 100644
--- a/packages/loopover-engine/README.md
+++ b/packages/loopover-engine/README.md
@@ -35,7 +35,8 @@ flags, so it works on the whole declared `engines` range.
 
 Codecov (`codecov/patch`) only reads the root vitest suite, so modules that need a Codecov-visible mirror also
 have a root test under `test/unit/` (e.g. `test/unit/reviewer-consensus-calibration.test.ts` for
-`src/reviewer-consensus-calibration.ts` — #8349). The package-local `node:test` suite remains the package's own
+`src/reviewer-consensus-calibration.ts` — #8349, or `test/unit/signal-tracking.test.ts` for
+`src/calibration/signal-tracking.ts` — #8343). The package-local `node:test` suite remains the package's own
 gate; the root mirror is what makes the same scenarios gradeable by Codecov.
 
 ## `opportunity-ranker`
diff --git a/test/unit/signal-tracking.test.ts b/test/unit/signal-tracking.test.ts
new file mode 100644
index 000000000..d18710062
--- /dev/null
+++ b/test/unit/signal-tracking.test.ts
@@ -0,0 +1,165 @@
+// Root-level vitest coverage twin for `packages/loopover-engine/src/calibration/signal-tracking.ts` (#7982).
+//
+// The engine package already has a full, passing `node --test` suite at
+// `packages/loopover-engine/test/signal-tracking.test.ts`, but that runner is NOT part of the root vitest run
+// Codecov reads `codecov/patch` from — so `computeRulePrecision`/`computeRuleRepeatCount`/
+// `evaluateRuleRepeatAlarm` report as ~0% covered despite being genuinely tested (#8343, same blind spot as
+// #6250). This file re-exercises the same behavior through the engine barrel so vitest (and therefore Codecov)
+// sees it too, exactly like the sibling twins `test/unit/calibration-dashboard.test.ts` and
+// `test/unit/discovery-soft-claim.test.ts`. It intentionally mirrors every scenario in the package's own suite
+// and covers both arms of every branch in the three functions.
+import { describe, expect, it } from "vitest";
+import {
+  computeRulePrecision,
+  computeRuleRepeatCount,
+  evaluateRuleRepeatAlarm,
+} from "../../packages/loopover-engine/src/index";
+import type { HumanOverrideEvent, RuleFiredEvent } from "../../packages/loopover-engine/src/index";
+
+function fired(ruleId: string, targetKey: string, overrides: Partial<RuleFiredEvent> = {}): RuleFiredEvent {
+  return { ruleId, targetKey, outcome: "block", occurredAt: "2026-07-22T00:00:00.000Z", ...overrides };
+}
+
+function override(
+  ruleId: string,
+  targetKey: string,
+  verdict: HumanOverrideEvent["verdict"],
+  overrides: Partial<HumanOverrideEvent> = {},
+): HumanOverrideEvent {
+  return { ruleId, targetKey, verdict, occurredAt: "2026-07-22T00:00:00.000Z", ...overrides };
+}
+
+describe("barrel: signal-tracking primitives are re-exported from the engine entrypoint (#7982/#7983)", () => {
+  it("exposes the three primitives as functions", () => {
+    expect(typeof computeRulePrecision).toBe("function");
+    expect(typeof computeRuleRepeatCount).toBe("function");
+    expect(typeof evaluateRuleRepeatAlarm).toBe("function");
+  });
+});
+
+describe("computeRulePrecision", () => {
+  it("no overrides -> decided is 0 and precision is null (unknown stays unknown, never coerced)", () => {
+    const report = computeRulePrecision(
+      "missing_linked_issue",
+      [fired("missing_linked_issue", "a#1"), fired("missing_linked_issue", "a#2")],
+      [],
+    );
+    expect(report).toEqual({
+      ruleId: "missing_linked_issue",
+      fired: 2,
+      reversed: 0,
+      confirmed: 0,
+      decided: 0,
+      precision: null,
+    });
+  });
+
+  it("mixes confirmed and reversed verdicts into a real precision (decided > 0 arm)", () => {
+    const report = computeRulePrecision(
+      "missing_linked_issue",
+      [
+        fired("missing_linked_issue", "a#1"),
+        fired("missing_linked_issue", "a#2"),
+        fired("missing_linked_issue", "a#3"),
+      ],
+      [
+        override("missing_linked_issue", "a#1", "confirmed"),
+        override("missing_linked_issue", "a#2", "confirmed"),
+        override("missing_linked_issue", "a#3", "reversed"),
+      ],
+    );
+    expect(report.fired).toBe(3);
+    expect(report.confirmed).toBe(2);
+    expect(report.reversed).toBe(1);
+    expect(report.decided).toBe(3);
+    expect(report.precision).toBe(2 / 3);
+  });
+
+  it("100% reversed yields precision 0, not null (a real, scored bad outcome, not an unknown one)", () => {
+    const report = computeRulePrecision("bad_rule", [fired("bad_rule", "a#1")], [override("bad_rule", "a#1", "reversed")]);
+    expect(report.decided).toBe(1);
+    expect(report.precision).toBe(0);
+  });
+
+  it("ignores fired/override events for a DIFFERENT ruleId entirely (both operand branches)", () => {
+    const report = computeRulePrecision(
+      "rule_a",
+      [fired("rule_a", "a#1"), fired("rule_b", "a#2")],
+      [override("rule_a", "a#1", "confirmed"), override("rule_b", "a#2", "reversed")],
+    );
+    expect(report.fired).toBe(1);
+    expect(report.confirmed).toBe(1);
+    expect(report.reversed).toBe(0);
+  });
+
+  it("an override with no matching fired event still counts toward decided (no cross-validation between the two lists)", () => {
+    const report = computeRulePrecision("rule_a", [], [override("rule_a", "a#1", "confirmed")]);
+    expect(report.fired).toBe(0);
+    expect(report.decided).toBe(1);
+    expect(report.precision).toBe(1);
+  });
+});
+
+describe("computeRuleRepeatCount", () => {
+  it("counts only fires matching BOTH ruleId and targetKey (each && operand's false arm exercised)", () => {
+    const events = [
+      fired("rule_a", "a#1"),
+      fired("rule_a", "a#1"),
+      fired("rule_a", "a#2"), // same rule, different target -> targetKey operand false
+      fired("rule_b", "a#1"), // different rule, same target -> ruleId operand false
+    ];
+    expect(computeRuleRepeatCount("rule_a", "a#1", events)).toBe(2);
+    expect(computeRuleRepeatCount("rule_a", "a#2", events)).toBe(1);
+    expect(computeRuleRepeatCount("rule_b", "a#1", events)).toBe(1);
+    expect(computeRuleRepeatCount("rule_a", "a#3", events)).toBe(0);
+  });
+
+  it("zero fired events yields 0, not an error", () => {
+    expect(computeRuleRepeatCount("rule_a", "a#1", [])).toBe(0);
+  });
+});
+
+describe("evaluateRuleRepeatAlarm (#7983)", () => {
+  it("not triggered below the threshold", () => {
+    const verdict = evaluateRuleRepeatAlarm("rule_a", [fired("rule_a", "a#1"), fired("rule_a", "a#2")], 3);
+    expect(verdict.triggered).toBe(false);
+    expect(verdict.affectedTargets).toEqual(["a#1", "a#2"]);
+    expect(verdict.threshold).toBe(3);
+  });
+
+  it("triggers once distinct targets reach the threshold (boundary: exactly `threshold` distinct targets)", () => {
+    const verdict = evaluateRuleRepeatAlarm(
+      "rule_a",
+      [fired("rule_a", "a#1"), fired("rule_a", "a#2"), fired("rule_a", "a#3")],
+      3,
+    );
+    expect(verdict.triggered).toBe(true);
+    expect(verdict.affectedTargets).toEqual(["a#1", "a#2", "a#3"]);
+  });
+
+  it("the SAME target firing repeatedly counts once, deduplicated in first-seen order (seen.has arm)", () => {
+    const verdict = evaluateRuleRepeatAlarm(
+      "rule_a",
+      [fired("rule_a", "a#1"), fired("rule_a", "a#1"), fired("rule_a", "a#1")],
+      2,
+    );
+    expect(verdict.affectedTargets).toEqual(["a#1"]);
+    expect(verdict.triggered).toBe(false);
+  });
+
+  it("ignores fired events for a DIFFERENT ruleId entirely (ruleId mismatch arm)", () => {
+    const verdict = evaluateRuleRepeatAlarm(
+      "rule_a",
+      [fired("rule_a", "a#1"), fired("rule_b", "a#2"), fired("rule_b", "a#3")],
+      2,
+    );
+    expect(verdict.affectedTargets).toEqual(["a#1"]);
+    expect(verdict.triggered).toBe(false);
+  });
+
+  it("zero fired events never triggers", () => {
+    const verdict = evaluateRuleRepeatAlarm("rule_a", [], 1);
+    expect(verdict.triggered).toBe(false);
+    expect(verdict.affectedTargets).toEqual([]);
+  });
+});