Open-source Python library for AI browser agents, plus Browser Use Cloud: stealth browsers and a hosted agent API.
Python ai-agentsai-toolsbrowser-agentbrowser-automationbrowser-usecdpcloud-browsercomputer-useheadless-browserllmmcpplaywrightpythonremote-browserstealth-browserweb-automationweb-scraping
Update to markdownify
Magnus Müllercf09b81af3e9
diff --git a/browser_use/tools/service.py b/browser_use/tools/service.pyindex fbde50749..79c0b6739 100644--- a/browser_use/tools/service.py+++ b/browser_use/tools/service.py@@ -1067,18 +1067,20 @@ You will be given a query and the markdown of a webpage that has been filtered tpage_html = html_serializer.serialize(enhanced_dom_tree)original_html_length = len(page_html)-- # Use html2text for clean markdown conversion- import html2text-- h = html2text.HTML2Text()- h.ignore_links = not extract_links- h.ignore_images = True- h.ignore_emphasis = False- h.body_width = 0 # Don't wrap lines- h.unicode_snob = True- h.skip_internal_links = True- content = h.handle(page_html)+ # save html to file+ # with open('zzzpage_html.html', 'w') as f:+ # f.write(page_html)++ # Use markdownify for clean markdown conversion+ from markdownify import markdownify as md++ content = md(+ page_html,+ heading_style='ATX', # Use # style headings+ strip=['script', 'style'], # Remove these tags+ bullets='-', # Use - for unordered lists+ code_language='', # Don't add language to code blocks+ )initial_markdown_length = len(content)@@ -1089,6 +1091,9 @@ You will be given a query and the markdown of a webpage that has been filtered tcontent, chars_filtered = self._preprocess_markdown_content(content)final_filtered_length = len(content)+ # save content to file+ # with open('zzzcontent.md', 'w') as f:+ # f.write(content)# Content statisticsstats = {diff --git a/pyproject.toml b/pyproject.tomlindex be7ef1d95..449160e3f 100644--- a/pyproject.toml+++ b/pyproject.toml@@ -40,8 +40,8 @@ dependencies = ["reportlab>=4.0.0","cdp-use>=1.4.0","pyotp>=2.9.0",- "html2text>=2025.4.15","pillow>=11.2.1",+ "markdownify>=1.2.0",]# google-api-core: only used for Google LLM APIs# pyperclip: only used for examples that use copy/paste