diff --git a/.github/workflows/pr-validation.yml b/.github/workflows/pr-validation.yml
index 215d98a..e4c9881 100644
--- a/.github/workflows/pr-validation.yml
+++ b/.github/workflows/pr-validation.yml
@@ -23,30 +23,31 @@ jobs:
           PR_BODY="${{ github.event.pull_request.body }}"
 
           # Check if PR title or body contains issue reference
-          # Accepts: TRCLI-### (JIRA), GIT-### (GitHub), #123 (GitHub), issues/123
-          if echo "$PR_TITLE $PR_BODY" | grep -qE "TRCLI-[0-9]+|GIT-[0-9]+|#[0-9]+|issues/[0-9]+"; then
-            echo "issue_found=true" >> $GITHUB_OUTPUT
+          if echo "$PR_TITLE $PR_BODY" | grep -qE "(TRCLI-[0-9]+|GIT-[0-9]+|#[0-9]+|issues/[0-9]+)"; then
+            echo "issue_found=true" >> "$GITHUB_OUTPUT"
 
-            # Extract the issue key/number
             if echo "$PR_TITLE $PR_BODY" | grep -qE "TRCLI-[0-9]+"; then
               ISSUE_KEY=$(echo "$PR_TITLE $PR_BODY" | grep -oE "TRCLI-[0-9]+" | head -1)
-              echo "issue_key=$ISSUE_KEY" >> $GITHUB_OUTPUT
-              echo "issue_type=JIRA" >> $GITHUB_OUTPUT
+              echo "issue_key=$ISSUE_KEY" >> "$GITHUB_OUTPUT"
+              echo "issue_type=JIRA" >> "$GITHUB_OUTPUT"
+
             elif echo "$PR_TITLE $PR_BODY" | grep -qE "GIT-[0-9]+"; then
               ISSUE_KEY=$(echo "$PR_TITLE $PR_BODY" | grep -oE "GIT-[0-9]+" | head -1)
-              echo "issue_key=$ISSUE_KEY" >> $GITHUB_OUTPUT
-              echo "issue_type=GitHub" >> $GITHUB_OUTPUT
+              echo "issue_key=$ISSUE_KEY" >> "$GITHUB_OUTPUT"
+              echo "issue_type=GitHub" >> "$GITHUB_OUTPUT"
+
             elif echo "$PR_TITLE $PR_BODY" | grep -qE "#[0-9]+"; then
               ISSUE_KEY=$(echo "$PR_TITLE $PR_BODY" | grep -oE "#[0-9]+" | head -1)
-              echo "issue_key=$ISSUE_KEY" >> $GITHUB_OUTPUT
-              echo "issue_type=GitHub" >> $GITHUB_OUTPUT
+              echo "issue_key=$ISSUE_KEY" >> "$GITHUB_OUTPUT"
+              echo "issue_type=GitHub" >> "$GITHUB_OUTPUT"
+
             elif echo "$PR_BODY" | grep -qE "issues/[0-9]+"; then
               ISSUE_KEY=$(echo "$PR_BODY" | grep -oE "issues/[0-9]+" | head -1)
-              echo "issue_key=$ISSUE_KEY" >> $GITHUB_OUTPUT
-              echo "issue_type=GitHub" >> $GITHUB_OUTPUT
+              echo "issue_key=$ISSUE_KEY" >> "$GITHUB_OUTPUT"
+              echo "issue_type=GitHub" >> "$GITHUB_OUTPUT"
             fi
           else
-            echo "issue_found=false" >> $GITHUB_OUTPUT
+            echo "issue_found=false" >> "$GITHUB_OUTPUT"
           fi
 
       - name: Comment on PR if issue reference missing
@@ -61,18 +62,15 @@ jobs:
               repo: context.repo.repo,
               body: `## ⚠️ Missing Issue Reference
 
-              This PR does not reference an issue. Please include a reference in either:
+              This PR does not reference an issue. Please include one.
 
-              **JIRA tickets:**
-              - PR title: "feat(api): TRCLI-123 Add new endpoint"
-              - PR description: "Resolves TRCLI-123"
+              **JIRA example:**
+              - TRCLI-123
 
-              **GitHub issues:**
-              - PR title: "feat(api): GIT-123 Add new endpoint"
-              - PR description: "Resolves GIT-123" or "Fixes #123"
-              - Or link to the GitHub issue
-
-              This helps with tracking and project management. Thank you!`
+              **GitHub examples:**
+              - GIT-123
+              - Fixes #123
+              - issues/123`
             })
 
       - name: Check PR Description Completeness
@@ -80,23 +78,22 @@ jobs:
         run: |
           PR_BODY="${{ github.event.pull_request.body }}"
 
-          # Check for required sections
           if echo "$PR_BODY" | grep -q "Issue being resolved"; then
-            echo "has_issue=true" >> $GITHUB_OUTPUT
+            echo "has_issue=true" >> "$GITHUB_OUTPUT"
           else
-            echo "has_issue=false" >> $GITHUB_OUTPUT
+            echo "has_issue=false" >> "$GITHUB_OUTPUT"
           fi
 
           if echo "$PR_BODY" | grep -q "Solution description"; then
-            echo "has_solution=true" >> $GITHUB_OUTPUT
+            echo "has_solution=true" >> "$GITHUB_OUTPUT"
           else
-            echo "has_solution=false" >> $GITHUB_OUTPUT
+            echo "has_solution=false" >> "$GITHUB_OUTPUT"
           fi
 
           if echo "$PR_BODY" | grep -q "Steps to test"; then
-            echo "has_test_steps=true" >> $GITHUB_OUTPUT
+            echo "has_test_steps=true" >> "$GITHUB_OUTPUT"
           else
-            echo "has_test_steps=false" >> $GITHUB_OUTPUT
+            echo "has_test_steps=false" >> "$GITHUB_OUTPUT"
           fi
 
       - name: Generate PR Validation Summary
@@ -107,6 +104,7 @@ jobs:
             const issueFound = '${{ steps.check_issue.outputs.issue_found }}' === 'true';
             const issueKey = '${{ steps.check_issue.outputs.issue_key }}';
             const issueType = '${{ steps.check_issue.outputs.issue_type }}';
+
             const hasIssue = '${{ steps.check_description.outputs.has_issue }}' === 'true';
             const hasSolution = '${{ steps.check_description.outputs.has_solution }}' === 'true';
             const hasTestSteps = '${{ steps.check_description.outputs.has_test_steps }}' === 'true';
@@ -124,9 +122,9 @@ jobs:
             | Solution Description | ${hasSolution ? '✅ Present' : '⚠️ Missing'} |
             | Test Steps | ${hasTestSteps ? '✅ Present' : '⚠️ Missing'} |
 
-            ${issueFound && hasSolution && hasTestSteps ? '✅ All checks passed!' : '⚠️ Some optional sections are missing. Consider adding them for better review context.'}
+            ${issueFound && hasSolution && hasTestSteps
+              ? '✅ All checks passed!'
+              : '⚠️ Some optional sections are missing.'}
             `;
 
-            await core.summary
-              .addRaw(summary)
-              .write();
+            await core.summary.addRaw(summary).write();
diff --git a/tests/test_api_request_handler.py b/tests/test_api_request_handler.py
index a093d22..c22157c 100644
--- a/tests/test_api_request_handler.py
+++ b/tests/test_api_request_handler.py
@@ -827,6 +827,242 @@ def test_delete_run(self, api_request_handler_verify: ApiRequestHandler, request
         resources_added, error = api_request_handler_verify.delete_run(run_id)
         assert error == "", "There should be no error in verification."
 
+    @pytest.mark.api_handler
+    def test_update_run_with_include_all_false_standalone(self, api_request_handler: ApiRequestHandler, requests_mock):
+        """Test update_run for standalone run with include_all=false"""
+        run_id = 100
+        run_name = "Updated Test Run"
+
+        # Mock get_run response - standalone run (no plan_id), include_all=false
+        get_run_response = {
+            "id": run_id,
+            "name": "Original Run",
+            "description": "Original description",
+            "refs": "REF-1",
+            "include_all": False,
+            "plan_id": None,
+            "config_ids": [],
+        }
+
+        # Mock get_tests response - existing cases in run
+        get_tests_response = {
+            "offset": 0,
+            "limit": 250,
+            "size": 2,
+            "_links": {"next": None, "prev": None},
+            "tests": [{"id": 1, "case_id": 1, "status_id": 1}, {"id": 2, "case_id": 2, "status_id": 1}],
+        }
+
+        # Mock update_run response
+        update_run_response = {"id": run_id, "name": run_name}
+
+        requests_mock.get(create_url(f"get_run/{run_id}"), json=get_run_response)
+        requests_mock.get(create_url(f"get_tests/{run_id}"), json=get_tests_response)
+        requests_mock.post(create_url(f"update_run/{run_id}"), json=update_run_response)
+
+        # Execute update_run
+        run_data, error = api_request_handler.update_run(run_id, run_name)
+
+        # Assertions
+        assert error == "", "No error should occur"
+        assert run_data["id"] == run_id, "Run ID should match"
+
+        # Verify the payload sent to update_run
+        request_history = requests_mock.request_history
+        update_request = [r for r in request_history if "update_run" in r.url and r.method == "POST"][0]
+        payload = update_request.json()
+
+        assert payload["include_all"] == False, "include_all should be False"
+        assert "case_ids" in payload, "case_ids should be present"
+        # Should contain union of existing (1, 2) and report cases
+        assert set(payload["case_ids"]) >= {1, 2}, "Should include existing case IDs"
+
+    @pytest.mark.api_handler
+    def test_update_run_with_include_all_false_plan_with_config(
+        self, api_request_handler: ApiRequestHandler, requests_mock
+    ):
+        """Test update_run for run in plan with config and include_all=false (the bug scenario)"""
+        run_id = 200
+        run_name = "Updated Test Run in Plan"
+
+        # Mock get_run response - run in plan with config, include_all=false
+        get_run_response = {
+            "id": run_id,
+            "name": "Original Run",
+            "description": "Original description",
+            "refs": "REF-1",
+            "include_all": False,
+            "plan_id": 10,
+            "config_ids": [5, 6],  # Has configs - will use update_run_in_plan_entry
+        }
+
+        # Mock get_tests response - existing cases
+        get_tests_response = {
+            "offset": 0,
+            "limit": 250,
+            "size": 3,
+            "_links": {"next": None, "prev": None},
+            "tests": [
+                {"id": 1, "case_id": 188, "status_id": 1},
+                {"id": 2, "case_id": 180, "status_id": 1},
+                {"id": 3, "case_id": 191, "status_id": 1},
+            ],
+        }
+
+        # Mock update_run_in_plan_entry response
+        update_run_response = {"id": run_id, "name": run_name}
+
+        requests_mock.get(create_url(f"get_run/{run_id}"), json=get_run_response)
+        requests_mock.get(create_url(f"get_tests/{run_id}"), json=get_tests_response)
+        requests_mock.post(create_url(f"update_run_in_plan_entry/{run_id}"), json=update_run_response)
+
+        # Execute update_run
+        run_data, error = api_request_handler.update_run(run_id, run_name)
+
+        # Assertions
+        assert error == "", "No error should occur"
+        assert run_data["id"] == run_id, "Run ID should match"
+
+        # Verify the payload sent to update_run_in_plan_entry
+        request_history = requests_mock.request_history
+        update_request = [r for r in request_history if "update_run_in_plan_entry" in r.url][0]
+        payload = update_request.json()
+
+        # THIS IS THE CRITICAL FIX - must include include_all=False
+        assert payload["include_all"] == False, "include_all must be False (fixes the bug)"
+        assert "case_ids" in payload, "case_ids should be present"
+        # Should contain union of existing (188, 180, 191) and report cases
+        assert set(payload["case_ids"]) >= {188, 180, 191}, "Should preserve existing case IDs"
+
+    @pytest.mark.api_handler
+    def test_update_run_with_include_all_true_preserves_setting(
+        self, api_request_handler: ApiRequestHandler, requests_mock
+    ):
+        """Test update_run preserves include_all=true and doesn't send case_ids"""
+        run_id = 300
+        run_name = "Updated Run with Include All"
+
+        # Mock get_run response - include_all=true
+        get_run_response = {
+            "id": run_id,
+            "name": "Original Run",
+            "description": "Original description",
+            "refs": "REF-1",
+            "include_all": True,  # Run includes all cases
+            "plan_id": None,
+            "config_ids": [],
+        }
+
+        # Mock update_run response
+        update_run_response = {"id": run_id, "name": run_name, "include_all": True}
+
+        requests_mock.get(create_url(f"get_run/{run_id}"), json=get_run_response)
+        requests_mock.post(create_url(f"update_run/{run_id}"), json=update_run_response)
+
+        # Execute update_run
+        run_data, error = api_request_handler.update_run(run_id, run_name)
+
+        # Assertions
+        assert error == "", "No error should occur"
+        assert run_data["include_all"] == True, "include_all should be preserved"
+
+        # Verify the payload sent to update_run
+        request_history = requests_mock.request_history
+        update_request = [r for r in request_history if "update_run" in r.url and r.method == "POST"][0]
+        payload = update_request.json()
+
+        assert payload["include_all"] == True, "include_all should be True"
+        assert "case_ids" not in payload, "case_ids should NOT be present when include_all=True"
+
+    @pytest.mark.api_handler
+    def test_update_run_handles_get_tests_error(self, api_request_handler: ApiRequestHandler, requests_mock):
+        """Test update_run handles errors from get_tests gracefully"""
+        run_id = 400
+        run_name = "Test Run"
+
+        # Mock get_run response - include_all=false
+        get_run_response = {
+            "id": run_id,
+            "name": "Original Run",
+            "description": "Original description",
+            "refs": "REF-1",
+            "include_all": False,
+            "plan_id": None,
+            "config_ids": [],
+        }
+
+        # Mock get_tests to return error (403 Forbidden, for example)
+        requests_mock.get(create_url(f"get_run/{run_id}"), json=get_run_response)
+        requests_mock.get(create_url(f"get_tests/{run_id}"), status_code=403, json={"error": "Access denied"})
+
+        # Execute update_run - should fail gracefully
+        run_data, error = api_request_handler.update_run(run_id, run_name)
+
+        # Assertions
+        assert run_data is None, "run_data should be None on error"
+        assert error is not None, "Error message should be present"
+        assert "Failed to get tests in run" in error, "Error should indicate get_tests failure"
+
+    @pytest.mark.api_handler
+    def test_update_run_with_include_all_false_plan_without_config(
+        self, api_request_handler: ApiRequestHandler, requests_mock
+    ):
+        """Test update_run for run in plan without config uses update_plan_entry"""
+        run_id = 500
+        run_name = "Updated Test Run in Plan No Config"
+        plan_id = 20
+        entry_id = "abc-123"
+
+        # Mock get_run response - run in plan without config
+        get_run_response = {
+            "id": run_id,
+            "name": "Original Run",
+            "description": "Original description",
+            "refs": "REF-1",
+            "include_all": False,
+            "plan_id": plan_id,
+            "config_ids": [],  # No configs - will use update_plan_entry
+        }
+
+        # Mock get_tests response
+        get_tests_response = {
+            "offset": 0,
+            "limit": 250,
+            "size": 1,
+            "_links": {"next": None, "prev": None},
+            "tests": [{"id": 1, "case_id": 50, "status_id": 1}],
+        }
+
+        # Mock get_plan response
+        get_plan_response = {
+            "id": plan_id,
+            "entries": [{"id": entry_id, "runs": [{"id": run_id, "entry_id": entry_id}]}],
+        }
+
+        # Mock update_plan_entry response
+        update_plan_response = {"id": run_id, "name": run_name}
+
+        requests_mock.get(create_url(f"get_run/{run_id}"), json=get_run_response)
+        requests_mock.get(create_url(f"get_tests/{run_id}"), json=get_tests_response)
+        requests_mock.get(create_url(f"get_plan/{plan_id}"), json=get_plan_response)
+        requests_mock.post(create_url(f"update_plan_entry/{plan_id}/{entry_id}"), json=update_plan_response)
+
+        # Execute update_run
+        run_data, error = api_request_handler.update_run(run_id, run_name)
+
+        # Assertions
+        assert error == "", "No error should occur"
+        assert run_data["id"] == run_id, "Run ID should match"
+
+        # Verify update_plan_entry was called with correct payload
+        request_history = requests_mock.request_history
+        update_request = [r for r in request_history if f"update_plan_entry/{plan_id}/{entry_id}" in r.url][0]
+        payload = update_request.json()
+
+        assert payload["include_all"] == False, "include_all should be False"
+        assert "case_ids" in payload, "case_ids should be present"
+        assert 50 in payload["case_ids"], "Should include existing case ID"
+
     @pytest.mark.api_handler
     def test_upload_attachments_413_error(self, api_request_handler: ApiRequestHandler, requests_mock, tmp_path):
         """Test that 413 errors (file too large) are properly reported."""
diff --git a/trcli/api/api_request_handler.py b/trcli/api/api_request_handler.py
index 4d4b190..77d6b3a 100644
--- a/trcli/api/api_request_handler.py
+++ b/trcli/api/api_request_handler.py
@@ -539,11 +539,22 @@ def update_run(
         else:
             add_run_data["refs"] = existing_refs  # Keep existing refs if none provided
 
-        run_tests, error_message = self.__get_all_tests_in_run(run_id)
-        run_case_ids = [test["case_id"] for test in run_tests]
-        report_case_ids = add_run_data["case_ids"]
-        joint_case_ids = list(set(report_case_ids + run_case_ids))
-        add_run_data["case_ids"] = joint_case_ids
+        existing_include_all = run_response.response_text.get("include_all", False)
+        add_run_data["include_all"] = existing_include_all
+
+        if not existing_include_all:
+            # Only manage explicit case_ids when include_all=False
+            run_tests, error_message = self.__get_all_tests_in_run(run_id)
+            if error_message:
+                return None, f"Failed to get tests in run: {error_message}"
+            run_case_ids = [test["case_id"] for test in run_tests]
+            report_case_ids = add_run_data["case_ids"]
+            joint_case_ids = list(set(report_case_ids + run_case_ids))
+            add_run_data["case_ids"] = joint_case_ids
+        else:
+            # include_all=True: TestRail includes all suite cases automatically
+            # Do NOT send case_ids array (TestRail ignores it anyway)
+            add_run_data.pop("case_ids", None)
 
         plan_id = run_response.response_text["plan_id"]
         config_ids = run_response.response_text["config_ids"]