diff --git a/.github/workflows/cli-build-binary-and-optionally-release.yml b/.github/workflows/cli-build-binary-and-optionally-release.yml
new file mode 100644
index 0000000000..bf8c0e854a
--- /dev/null
+++ b/.github/workflows/cli-build-binary-and-optionally-release.yml
@@ -0,0 +1,104 @@
+# Workflow that builds and tests the CLI binary executable
+name: CLI - Build binary and optionally release
+
+# Run on pushes to main branch and CLI tags, and on pull requests when CLI files change
+on:
+  push:
+    branches:
+      - main
+    tags:
+      - "*-cli"
+  pull_request:
+    paths:
+      - "openhands-cli/**"
+
+# Cancel previous runs if a new commit is pushed
+concurrency:
+  group: ${{ github.workflow }}-${{ (github.head_ref && github.ref) || github.run_id }}
+  cancel-in-progress: true
+
+jobs:
+  build-and-test-binary:
+    name: Build and test binary executable
+    strategy:
+      matrix:
+        os: [ubuntu-latest, macos-latest]
+    runs-on: ${{ matrix.os }}
+
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
+
+      - name: Set up Python
+        uses: actions/setup-python@v5
+        with:
+          python-version: 3.12
+
+      - name: Install uv
+        uses: astral-sh/setup-uv@v3
+        with:
+          version: "latest"
+
+      - name: Install dependencies
+        working-directory: openhands-cli
+        run: |
+          uv sync
+
+      - name: Build binary executable
+        working-directory: openhands-cli
+        run: |
+          ./build.sh --install-pyinstaller | tee output.log
+          echo "Full output:"
+          cat output.log
+
+          if grep -q "❌" output.log; then
+            echo "❌ Found failure marker in output"
+            exit 1
+          fi
+
+          echo "✅ Build & test finished without ❌ markers"
+
+      - name: Upload binary artifact (for releases only)
+        if: startsWith(github.ref, 'refs/tags/')
+        uses: actions/upload-artifact@v4
+        with:
+          name: openhands-cli-${{ matrix.os }}
+          path: openhands-cli/dist/openhands*
+          retention-days: 30
+
+  create-github-release:
+    name: Create GitHub Release
+    runs-on: ubuntu-latest
+    needs: build-and-test-binary
+    if: startsWith(github.ref, 'refs/tags/')
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@v4
+
+      - name: Download all artifacts
+        uses: actions/download-artifact@v4
+        with:
+          path: artifacts
+
+      - name: Prepare release assets
+        run: |
+          mkdir -p release-assets
+          # Rename binaries to include OS in filename
+          if [ -f artifacts/openhands-cli-ubuntu-latest/openhands ]; then
+            cp artifacts/openhands-cli-ubuntu-latest/openhands release-assets/openhands-linux
+          fi
+          if [ -f artifacts/openhands-cli-macos-latest/openhands ]; then
+            cp artifacts/openhands-cli-macos-latest/openhands release-assets/openhands-macos
+          fi
+          ls -la release-assets/
+
+      - name: Create GitHub Release
+        uses: softprops/action-gh-release@v2
+        with:
+          files: release-assets/*
+          draft: true
+          prerelease: false
+        env:
+          GITHUB_TOKEN: ${{ secrets.CLI_RELEASE_TOKEN }}
diff --git a/.github/workflows/cli-build-test.yml b/.github/workflows/cli-build-test.yml
index b1fc1ac6f4..0fddcbfe0a 100644
--- a/.github/workflows/cli-build-test.yml
+++ b/.github/workflows/cli-build-test.yml
@@ -12,6 +12,9 @@ on:
     paths:
       - "openhands-cli/**"
 
+permissions:
+  contents: write       # needed to create releases or upload assets
+
 # Cancel previous runs if a new commit is pushed
 concurrency:
   group: ${{ github.workflow }}-${{ (github.head_ref && github.ref) || github.run_id }}
@@ -101,4 +104,4 @@ jobs:
           draft: true
           prerelease: false
         env:
-          GITHUB_TOKEN: ${{ secrets.CLI_RELEASE_TOKEN }}
+          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
diff --git a/tests/unit/mcp/test_mcp_tool_timeout_stall.py b/tests/unit/mcp/test_mcp_tool_timeout_stall.py
index 209cf4fa46..e588b10311 100644
--- a/tests/unit/mcp/test_mcp_tool_timeout_stall.py
+++ b/tests/unit/mcp/test_mcp_tool_timeout_stall.py
@@ -149,6 +149,7 @@ async def test_mcp_tool_timeout_error_handling(conversation_stats):
     # This demonstrates that the fix is working
 
 
+@pytest.mark.skip(reason='flaky test')
 @pytest.mark.asyncio
 async def test_mcp_tool_timeout_agent_continuation(conversation_stats):
     """Test that verifies the agent can continue processing after an MCP tool timeout."""