diff --git a/.github/ISSUE_TEMPLATE/bug_report.yml b/.github/ISSUE_TEMPLATE/bug_report.yml
index 7f10d85..15ac75a 100644
--- a/.github/ISSUE_TEMPLATE/bug_report.yml
+++ b/.github/ISSUE_TEMPLATE/bug_report.yml
@@ -70,6 +70,21 @@ body:
     validations:
       required: false
 
+  - type: textarea
+    id: impact
+    attributes:
+      label: Measurable impact verification
+      description: |
+        How will we verify the fix works and measure its impact? Provide a table with concrete metrics.
+        Every metric must be quantifiable — no vague criteria like "works better" or "improved".
+      placeholder: |
+        | Metric | Before (broken) | After (target) | How to measure |
+        |--------|-----------------|----------------|----------------|
+        | Error on query X | throws TypeError | returns valid answer | `bun run cli -- query --question "X"` |
+        | Test pass rate | 38/39 (1 failure) | 39/39 | `bun run test` |
+    validations:
+      required: true
+
   - type: textarea
     id: context
     attributes:
diff --git a/.github/ISSUE_TEMPLATE/feature_request.yml b/.github/ISSUE_TEMPLATE/feature_request.yml
index eb1b7ee..66d4d77 100644
--- a/.github/ISSUE_TEMPLATE/feature_request.yml
+++ b/.github/ISSUE_TEMPLATE/feature_request.yml
@@ -41,6 +41,21 @@ body:
     validations:
       required: true
 
+  - type: textarea
+    id: impact
+    attributes:
+      label: Measurable impact verification
+      description: |
+        How will we verify the feature works and measure its impact? Provide a table with concrete metrics.
+        Every metric must be quantifiable — no vague criteria like "works better" or "improved".
+      placeholder: |
+        | Metric | Before | After (target) | How to measure |
+        |--------|--------|----------------|----------------|
+        | Total MCP tool definitions | 10 | 12 | `grep -c createTool src/mcp/tools.ts` |
+        | Feature test cases added | 0 | ≥5 | `grep -Ec '\b(it|test)\(' tests/feature.test.ts` |
+    validations:
+      required: true
+
   - type: textarea
     id: alternatives
     attributes:
diff --git a/.github/ISSUE_TEMPLATE/task.yml b/.github/ISSUE_TEMPLATE/task.yml
index 007adef..32c8909 100644
--- a/.github/ISSUE_TEMPLATE/task.yml
+++ b/.github/ISSUE_TEMPLATE/task.yml
@@ -59,6 +59,21 @@ body:
     validations:
       required: false
 
+  - type: textarea
+    id: impact
+    attributes:
+      label: Measurable impact verification
+      description: |
+        How will we verify the task is complete and measure its impact? Provide a table with concrete metrics.
+        Every metric must be quantifiable — no vague criteria like "works better" or "improved".
+      placeholder: |
+        | Metric | Before | After (target) | How to measure |
+        |--------|--------|----------------|----------------|
+        | LOC in target file | 1614 | ≤400 | `wc -l src/runtime/target.ts` |
+        | Existing test pass rate | all pass | all pass | `bun run test` |
+    validations:
+      required: true
+
   - type: dropdown
     id: agent_delegable
     attributes:
diff --git a/.github/pull_request_template.md b/.github/pull_request_template.md
index 74b11a9..e7d9ea6 100644
--- a/.github/pull_request_template.md
+++ b/.github/pull_request_template.md
@@ -18,6 +18,15 @@
 
 <!-- Key changes, one bullet per logical unit -->
 
+## Measurable impact verification
+
+<!-- Required. Provide a table showing before/after metrics that prove this PR's impact.
+     Every metric must be quantifiable — no vague criteria like "works better" or "improved". -->
+
+| Metric | Before | After (target) | How to measure |
+|--------|--------|----------------|----------------|
+| <!-- e.g. LOC in target file --> | <!-- e.g. 1614 --> | <!-- e.g. ≤400 --> | <!-- e.g. `wc -l src/runtime/target.ts` --> |
+
 ## Validation
 
 - [ ] `bun run lint` passes locally