diff --git a/scripts/aggregate_comments_by_claim.py b/scripts/aggregate_comments_by_claim.py
new file mode 100644
index 0000000..cf7baa2
--- /dev/null
+++ b/scripts/aggregate_comments_by_claim.py
@@ -0,0 +1,66 @@
+import asyncio
+import json
+
+import aiohttp
+
+from src.database.queries import obtain_connection
+
+
+async def main():
+    conn = obtain_connection('your_path_here')
+    with conn:
+        curs = conn.cursor()
+        res = curs.execute("""
+            SELECT DISTINCT LbryClaimId claim_id FROM COMMENT; 
+        """).fetchall()
+        rows = [tuple(r)[0] for r in res]
+        claims = {cid: {"comments": []} for cid in rows}
+
+        comments = curs.execute("""
+            SELECT channel_name, comment, comment_id, claim_id, timestamp 
+            FROM COMMENTS_ON_CLAIMS
+        """).fetchall()
+        comments = [dict(r) for r in comments]
+        while len(comments) > 0:
+            c = comments.pop()
+            cid = c.pop('claim_id')
+            claims[cid]['comments'].append(c)
+
+    lbrynet = 'http://localhost:5279'
+
+    async with aiohttp.ClientSession() as client:
+        i = 0
+        for cid, data in claims.items():
+            body = {
+                'method': 'claim_search',
+                'params': {
+                    'claim_id': cid,
+                    'no_totals': True
+                }
+            }
+
+            async with client.post(lbrynet, json=body) as resp:
+                res = (await resp.json())
+
+            try:
+                res = res['result']['items']
+                print(f'{i} - ok')
+                if res:
+                    data.update({
+                        'name': res[0]['name'],
+                        'permanent_url': res[0]['permanent_url']
+                    })
+            except KeyError:
+                print(f'{i}: broke')
+                # await asyncio.sleep(1)
+            i += 1
+    return claims
+
+if __name__ == '__main__':
+    loop = asyncio.get_event_loop()
+    claims = loop.run_until_complete(main())
+    # print(claims)
+    with open('comments_on_claims.json', 'w') as fp:
+        json.dump(claims, fp, indent=2)
+
+