diff --git a/tools/contribution_analysis.py b/tools/contribution_analysis.py
new file mode 100644
index 0000000..36f96ca
--- /dev/null
+++ b/tools/contribution_analysis.py
@@ -0,0 +1,42 @@
+import requests
+import csv
+
+def analysis_issue():
+    total_data = []
+    for page_num in range(20):
+        print(f"Pulling page {page_num}")
+        response = requests.get('https://api.github.com/repos/openmlsys/openmlsys-zh/issues', params={"state": "all", "per_page": 30, "page": page_num})
+        print(f"Result status: {response}")
+        data = response.json()
+        if len(data) == 0:
+            break
+        total_data += data    
+
+    all_issues = []
+    node_ids = set([])
+    for item in total_data:
+        if isinstance(item, dict):
+            if item["node_id"] not in node_ids:
+                all_issues.append([item["user"]["login"], item["author_association"], 1, [item["url"].split("/")[-1]]])
+                node_ids.add(item["node_id"])
+    print(f"All issues and pr count {len(all_issues)}")
+    total_count = {}
+    for issue_item in all_issues:
+        if total_count.get(issue_item[0], None) is None:
+            total_count[issue_item[0]] = issue_item
+        else:
+            total_count[issue_item[0]][-2] += 1
+            total_count[issue_item[0]][-1] += issue_item[-1]
+    
+    keys = sorted(total_count, key=lambda x: total_count[x][-2])
+    
+    final_res = []
+    for key in keys:
+        total_count[key][-1] = ",".join(total_count[key][-1])
+        final_res.append(total_count[key])
+        
+    res_file = open("contribution_stats.csv", "w")
+    csv_writer = csv.writer(res_file)
+    csv_writer.writerow(["github id", "role", "issue and pr count", "issue or pr ids"])
+    csv_writer.writerows(final_res)
+analysis_issue()
\ No newline at end of file