From 822afc6a09bc17975409b64aee5689152df91c87 Mon Sep 17 00:00:00 2001
From: bb <bb@browserbase.com>
Date: Wed, 20 May 2026 23:25:24 +0000
Subject: [PATCH] [DASH-2089] [templates] smart-fetch-scraper: comment about
 Fetch markdown/json formats

---
 python/smart-fetch-scraper/main.py      | 8 ++++++++
 typescript/smart-fetch-scraper/index.ts | 8 ++++++++
 2 files changed, 16 insertions(+)

diff --git a/python/smart-fetch-scraper/main.py b/python/smart-fetch-scraper/main.py
index 302a5d45..bd8a6864 100644
--- a/python/smart-fetch-scraper/main.py
+++ b/python/smart-fetch-scraper/main.py
@@ -3,6 +3,11 @@
 # Tries the Browserbase Fetch API first (fast, no browser session needed).
 # If the page is JS-rendered or the content is insufficient, falls back to
 # a full Stagehand browser session with AI-powered extraction.
+#
+# Note: the Fetch API can return content as `raw` HTML, clean `markdown`,
+# or schema-extracted `json` — pass `format="markdown"` or `format="json"`
+# (with a JSON schema) to `bb.fetch_api.create` to skip writing your own
+# HTML parser. Docs: https://docs.browserbase.com/platform/fetch/overview
 
 import asyncio
 import json
@@ -116,6 +121,9 @@ async def try_fetch_api(url: str) -> dict | None:
     print("[Fetch API] Attempting lightweight fetch...")
 
     try:
+        # Tip: pass `format="markdown"` or `format="json"` (with a `schema`)
+        # here to have Browserbase return cleaner content or structured data
+        # directly — see https://docs.browserbase.com/platform/fetch/overview
         # Use asyncio.to_thread for synchronous SDK calls
         data = await asyncio.to_thread(bb.fetch_api.create, url=url, allow_redirects=True)
 
diff --git a/typescript/smart-fetch-scraper/index.ts b/typescript/smart-fetch-scraper/index.ts
index a31baaaa..f20b799e 100644
--- a/typescript/smart-fetch-scraper/index.ts
+++ b/typescript/smart-fetch-scraper/index.ts
@@ -3,6 +3,11 @@
 // Tries the Browserbase Fetch API first (fast, no browser session needed).
 // If the page is JS-rendered or the content is insufficient, falls back to
 // a full Stagehand browser session with AI-powered extraction.
+//
+// Note: the Fetch API can return content as `raw` HTML, clean `markdown`,
+// or schema-extracted `json` — pass `format: "markdown"` or `format: "json"`
+// (with a JSON schema) to `bb.fetchAPI.create` to skip writing your own
+// HTML parser. Docs: https://docs.browserbase.com/platform/fetch/overview
 
 import "dotenv/config";
 import Browserbase from "@browserbasehq/sdk";
@@ -91,6 +96,9 @@ async function tryFetchApi(url: string): Promise<{ content: string; statusCode:
   console.log("[Fetch API] Attempting lightweight fetch...");
 
   try {
+    // Tip: pass `format: "markdown"` or `format: "json"` (with a `schema`)
+    // here to have Browserbase return cleaner content or structured data
+    // directly — see https://docs.browserbase.com/platform/fetch/overview
     const data = await bb.fetchAPI.create({ url, allowRedirects: true });
 
     console.log(