diff --git a/src/tools/prompt-score.ts b/src/tools/prompt-score.ts
index 1cecf01..b4b3a50 100644
--- a/src/tools/prompt-score.ts
+++ b/src/tools/prompt-score.ts
@@ -40,7 +40,7 @@ interface ScoreResult {
   feedback: string[];
 }
 
-function scorePrompt(text: string): ScoreResult {
+export function scorePrompt(text: string): ScoreResult {
   const feedback: string[] = [];
   let specificity: number;
   let scope: number;
@@ -59,11 +59,18 @@ function scorePrompt(text: string): ScoreResult {
   }
 
   // Scope: bounded task
-  if (/\b(only|just|single|one|specific|this)\b/i.test(text) || text.length > 100) {
+  const hasScopeKeyword = /\b(only|just|single|one|specific|this)\b/i.test(text);
+  const hasBroadKeyword = /\b(all|every|entire|whole)\b/i.test(text);
+  if (hasScopeKeyword) {
     scope = 25;
-  } else if (/\b(all|every|entire|whole)\b/i.test(text)) {
+  } else if (hasBroadKeyword) {
     scope = 10;
     feedback.push("🎯 'All/every' is broad — can you narrow the scope?");
+  } else if (text.length > 100) {
+    // Longer prompts tend to be more detailed, but without explicit scope words
+    // they only get partial credit
+    scope = 20;
+    feedback.push("🎯 Scope seems implicit — add 'only'/'just' to be explicit for +5");
   } else {
     scope = 10;
     feedback.push("🎯 Scope unclear — how much should change?");
diff --git a/tests/tools/prompt-score.test.ts b/tests/tools/prompt-score.test.ts
new file mode 100644
index 0000000..19f3b01
--- /dev/null
+++ b/tests/tools/prompt-score.test.ts
@@ -0,0 +1,83 @@
+import { describe, it, expect } from "vitest";
+import { scorePrompt } from "../../src/tools/prompt-score.js";
+
+describe("scorePrompt", () => {
+  it("gives high score to a well-formed prompt", () => {
+    const result = scorePrompt(
+      "Rename the `processOrder` function in src/orders/handler.ts to `handleOrder` and update only the call sites in this module. The tests should still pass."
+    );
+    expect(result.total).toBeGreaterThanOrEqual(80);
+    expect(result.grade).toMatch(/^[AB]/);
+  });
+
+  it("gives low score to a vague prompt", () => {
+    const result = scorePrompt("make it better");
+    expect(result.total).toBeLessThanOrEqual(40);
+    expect(result.grade).toMatch(/^[DF]/);
+    expect(result.feedback.length).toBeGreaterThan(0);
+  });
+
+  it("rewards specificity for file paths", () => {
+    const withPath = scorePrompt("fix src/index.ts");
+    const without = scorePrompt("fix the code");
+    expect(withPath.specificity).toBeGreaterThan(without.specificity);
+  });
+
+  it("rewards specificity for backtick-quoted identifiers", () => {
+    const result = scorePrompt("rename `foo` to `bar`");
+    expect(result.specificity).toBe(25);
+  });
+
+  it("penalizes broad scope keywords", () => {
+    const broad = scorePrompt("refactor every component");
+    expect(broad.scope).toBeLessThanOrEqual(10);
+    expect(broad.feedback.some((f) => f.includes("broad"))).toBe(true);
+  });
+
+  it("gives explicit scope words full marks", () => {
+    const result = scorePrompt("update only this function");
+    expect(result.scope).toBe(25);
+  });
+
+  it("does not give full scope just for being long", () => {
+    const longVague = scorePrompt(
+      "I want you to go ahead and improve the performance of the application because it has been running really slowly lately and users are complaining about it"
+    );
+    // Should get partial credit (20) not full (25)
+    expect(longVague.scope).toBeLessThan(25);
+    expect(longVague.scope).toBeGreaterThanOrEqual(15);
+  });
+
+  it("rewards action verbs", () => {
+    const good = scorePrompt("extract the validation logic");
+    const vague = scorePrompt("handle the validation logic");
+    expect(good.actionability).toBeGreaterThan(vague.actionability);
+  });
+
+  it("rewards done conditions", () => {
+    const result = scorePrompt("add a test that should return 200");
+    expect(result.doneCondition).toBe(25);
+  });
+
+  it("gives questions partial done-condition credit", () => {
+    // Use a question that doesn't contain done-condition keywords
+    const result = scorePrompt("what is the best approach here?");
+    expect(result.doneCondition).toBe(20);
+  });
+
+  it("grade boundaries are correct", () => {
+    // A perfect prompt should hit A range
+    const perfect = scorePrompt(
+      "Fix only the `validate` function in src/lib/validators.ts — it should return false for empty strings"
+    );
+    expect(perfect.total).toBeGreaterThanOrEqual(85);
+  });
+
+  it("returns feedback array even for perfect prompts", () => {
+    const result = scorePrompt(
+      "Fix only the `validate` function in src/lib/validators.ts — it should return false for empty strings"
+    );
+    expect(Array.isArray(result.feedback)).toBe(true);
+    expect(result.feedback.length).toBeGreaterThan(0);
+  });
+});