<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>Trestle Articles</title>
    <link>https://trestlescan.com/articles/</link>
    <description>Guides on finding, fixing, and preventing leaked secrets: responding when a key leaks, how secret scanning works, and how Trestle can help.</description>
    <language>en</language>
    <lastBuildDate>Tue, 30 Jun 2026 01:00:00 GMT</lastBuildDate>
    <atom:link href="https://trestlescan.com/rss.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>AST vs regex secret scanning: what each approach catches and misses</title>
      <link>https://trestlescan.com/articles/ast-vs-regex-secret-scanning/</link>
      <guid isPermaLink="true">https://trestlescan.com/articles/ast-vs-regex-secret-scanning/</guid>
      <description>Regex matches known key formats and entropy scoring catches high-randomness strings, but both miss generic passwords and raise false positives on ordinary code. Parser-based scanning reads code structure and catches what patterns miss, while regex stays the better choice for fixed-format keys.</description>
      <category>Secret scanning</category>
      <pubDate>Tue, 30 Jun 2026 01:00:00 GMT</pubDate>
    </item>
    <item>
      <title>The B1-B4 trust boundary model for AI coding agents</title>
      <link>https://trestlescan.com/articles/b1-b4-trust-boundaries/</link>
      <guid isPermaLink="true">https://trestlescan.com/articles/b1-b4-trust-boundaries/</guid>
      <description>A pipeline threat model maps AI coding agent security onto four boundaries, from the developer&apos;s prompt to production: what each boundary is, the threats and control at each, and where hardcoded secrets fit.</description>
      <category>AI agent security</category>
      <pubDate>Mon, 29 Jun 2026 01:00:00 GMT</pubDate>
    </item>
    <item>
      <title>Deleting a secret and committing again isn&apos;t enough</title>
      <link>https://trestlescan.com/articles/deleting-secret-is-not-enough/</link>
      <guid isPermaLink="true">https://trestlescan.com/articles/deleting-secret-is-not-enough/</guid>
      <description>If you deleted a committed secret and pushed again, the secret still works. The old value stays in your git history, in forks, and in clones. The only fix is to replace the key.</description>
      <category>Incident response</category>
      <pubDate>Fri, 26 Jun 2026 01:00:00 GMT</pubDate>
    </item>
    <item>
      <title>How fast are leaked API keys found?</title>
      <link>https://trestlescan.com/articles/how-fast-are-leaked-keys-found/</link>
      <guid isPermaLink="true">https://trestlescan.com/articles/how-fast-are-leaked-keys-found/</guid>
      <description>On public GitHub, a leaked API key is found and used within minutes, sometimes within one minute. You can&apos;t outrun it, so replace the key first, not the commit.</description>
      <category>Threat research</category>
      <pubDate>Thu, 25 Jun 2026 01:00:00 GMT</pubDate>
    </item>
    <item>
      <title>Leaked secret emergency checklist</title>
      <link>https://trestlescan.com/articles/leaked-secret-checklist/</link>
      <guid isPermaLink="true">https://trestlescan.com/articles/leaked-secret-checklist/</guid>
      <description>A do-it-now checklist for a leaked API key, token, or password: contain it, check for misuse, and prevent a repeat. Replace the key first. Optionally, clean it out of your git history.</description>
      <category>Incident response</category>
      <pubDate>Wed, 24 Jun 2026 01:00:00 GMT</pubDate>
    </item>
    <item>
      <title>API key leak prevention: how to keep secrets out of your code</title>
      <link>https://trestlescan.com/articles/api-key-leak-prevention/</link>
      <guid isPermaLink="true">https://trestlescan.com/articles/api-key-leak-prevention/</guid>
      <description>A practical, vendor-neutral guide to API key leak prevention: why credentials leak, what it costs, and the layered defenses that stop secrets from escaping your codebase.</description>
      <category>Guide</category>
      <pubDate>Mon, 22 Jun 2026 01:00:00 GMT</pubDate>
    </item>
    <item>
      <title>Four secret scanners compared on four codebases</title>
      <link>https://trestlescan.com/articles/secret-scanners-compared/</link>
      <guid isPermaLink="true">https://trestlescan.com/articles/secret-scanners-compared/</guid>
      <description>Trestle, Gitleaks, TruffleHog, and detect-secrets run with default settings over four public codebases. Every scanner catches an obvious API key. The differences show up with the secrets that don&apos;t look like secrets: hashed passwords, weak passwords, credit card numbers, and values that get sent to the browser.</description>
      <category>Tool comparison</category>
      <pubDate>Wed, 17 Jun 2026 01:00:00 GMT</pubDate>
    </item>
  </channel>
</rss>
