diff --git a/.github/workflows/reindex_on_merge.yml b/.github/workflows/reindex_on_merge.yml
index c31e08b..bb3d290 100644
--- a/.github/workflows/reindex_on_merge.yml
+++ b/.github/workflows/reindex_on_merge.yml
@@ -14,23 +14,53 @@ jobs:
     # This condition ensures the job only runs if the pull request was actually merged.
     if: github.event.pull_request.merged == true || github.event_name == 'workflow_dispatch'
 
-    # *** KEY CHANGE ***
     # This tells GitHub to run this job on one of your self-hosted runners.
-    # You can also add labels to target specific servers, e.g., [self-hosted, linux, x64, my-app]
     runs-on: Windows
 
+    # Set a default shell for all run steps in the job.
+    defaults:
+      run:
+        shell: bash
+
     steps:
       # Step 1: Check out the repository's code
       # This downloads the latest version of your 'main' branch into the runner's working directory.
       - name: Checkout repository
         uses: actions/checkout@v4
 
-      # Step 2: Run the indexing script
-      # This executes your 'create_index.py' script using the Python environment on your server.
-      # It assumes Python and all dependencies from requirements.txt are already installed on the server.
+      # Step 2: Set up a specific Python version
+      # This ensures a consistent Python version is used for the venv.
+      - name: Set up Python 3.11
+        id: setup-python
+        uses: actions/setup-python@v4
+        with:
+          python-version: '3.11'
+
+      # Step 3: Create and activate a virtual environment
+      # This creates a '.venv' directory in your project folder if it doesn't exist,
+      # and then activates it for subsequent steps.
+      - name: Create and activate virtual environment
+        run: |
+          if [ ! -d ".venv" ]; then
+            python -m venv .venv
+          fi
+          source .venv/bin/activate
+
+      # Step 4: Install or update dependencies
+      # This uses the pip from the virtual environment to install all required libraries.
+      - name: Install dependencies
+        run: |
+          source .venv/bin/activate
+          pip install --upgrade pip
+          pip install -r requirements.txt
+      
+      # Step 5: Run the indexing script within the virtual environment
+      # It executes your 'create_index.py' script using the Python from the venv.
       # The GITHUB_TOKEN is still passed securely to the script.
       - name: Run indexing script
-        run: python create_index.py
+        run: |
+          source .venv/bin/activate
+          python create_index.py
         env:
           GITHUB_REPOSITORY: ${{ github.repository }}
           GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
diff --git a/create_index.py b/create_index.py
index d1292c7..d22ee01 100644
--- a/create_index.py
+++ b/create_index.py
@@ -19,7 +19,7 @@ from tools.github_tool import GitHubTool
 EMBEDDING_MODEL_PATH = """C:\Models\embeddings\Qwen3-Embedding-0.6B"""
 
 # Path to store the local vector database
-CHROMA_DB_PATH = "C:\Models\embeddings\embedding_result\chroma_db"
+CHROMA_DB_PATH = """C:\Models\embeddings\embedding_result\chroma_db"""
 # Name of the collection within the database
 CHROMA_COLLECTION_NAME = "github_repo"
 # Files with these extensions will be indexed. Add any other text-based files you need.