<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/">
  <channel>
    <title>Windsurf on Jashid Sany</title>
    <link>https://jashidsany.com/tags/windsurf/</link>
    <description>Recent content in Windsurf on Jashid Sany</description>
    <generator>Hugo</generator>
    <language>en-us</language>
    <lastBuildDate>Sun, 15 Mar 2026 00:00:00 +0000</lastBuildDate>
    <atom:link href="https://jashidsany.com/tags/windsurf/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Windsurf Cascade: Indirect Prompt Injection and Credential Exfiltration via Github Gists</title>
      <link>https://jashidsany.com/security-research/ai-security/windsurf-indirect-prompt-injection-blog/</link>
      <pubDate>Sun, 15 Mar 2026 00:00:00 +0000</pubDate>
      <guid>https://jashidsany.com/security-research/ai-security/windsurf-indirect-prompt-injection-blog/</guid>
      <description>How a hidden HTML comment in a GitHub Gist caused Windsurf&amp;rsquo;s Cascade agent to read SSH keys and AWS credentials, then exfiltrate them to an attacker-controlled endpoint with zero user interaction.</description>
    </item>
    <item>
      <title>Windsurf Cascade: Overly Permissive IDE Agent Bypasses Auto Execution Controls</title>
      <link>https://jashidsany.com/security-research/ai-security/windsurf-overly-permissive-agent-blog/</link>
      <pubDate>Sun, 15 Mar 2026 00:00:00 +0000</pubDate>
      <guid>https://jashidsany.com/security-research/ai-security/windsurf-overly-permissive-agent-blog/</guid>
      <description>Windsurf&amp;rsquo;s Cascade agent reads and writes files outside the workspace with zero confirmation, even with Auto Execution set to Disabled. The review prompt on writes is cosmetic: files exist on disk before the user can reject.</description>
    </item>
  </channel>
</rss>
