<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0">
  <channel>
    <title>krackedtools.dev</title>
    <link>https://krackedtools.dev/</link>
    <description>Writing by Hendrik Krack on AI code review, developer tools, and climate-aware engineering.</description>
    <language>en-us</language>
    <lastBuildDate>Fri, 25 Sep 2026 20:00:00 +0000</lastBuildDate>
    <item>
      <title>The Merge: Your tests pass. Can your agent finish the job?</title>
      <link>https://krackedtools.dev/posts/the-merge-cua-can-your-agent-finish-the-job.html</link>
      <guid isPermaLink="true">https://krackedtools.dev/posts/the-merge-cua-can-your-agent-finish-the-job.html</guid>
      <pubDate>Fri, 25 Sep 2026 20:00:00 +0000</pubDate>
      <description>Cua AI co-founder Francesco Bonacci explains why developers need to check whether AI agents finish the job, even when the tools they use work correctly.</description>
    </item>
    <item>
      <title>Claude Opus 5.5 for code review: More catches, different misses</title>
      <link>https://krackedtools.dev/posts/claude-opus-5-5-for-code-review.html</link>
      <guid isPermaLink="true">https://krackedtools.dev/posts/claude-opus-5-5-for-code-review.html</guid>
      <pubDate>Mon, 21 Sep 2026 00:00:00 +0000</pubDate>
      <description>We tested Claude Opus 5.5 against CodeRabbit's production reviewer to see which bugs it catches, which it misses, and what higher reasoning effort changes.</description>
    </item>
    <item>
      <title>A software factory needs a review gate it can trust</title>
      <link>https://krackedtools.dev/posts/software-factory-review-gate.html</link>
      <guid isPermaLink="true">https://krackedtools.dev/posts/software-factory-review-gate.html</guid>
      <pubDate>Thu, 10 Sep 2026 00:00:00 +0000</pubDate>
      <description>Why agent-driven delivery still needs a visible, risk-based gate before code ships.</description>
    </item>
    <item>
      <title>What would the last software engineer still need to do?</title>
      <link>https://krackedtools.dev/posts/the-last-software-engineer-knows-what-to-build.html</link>
      <guid isPermaLink="true">https://krackedtools.dev/posts/the-last-software-engineer-knows-what-to-build.html</guid>
      <pubDate>Thu, 03 Sep 2026 00:00:00 +0000</pubDate>
      <description>Kent C. Dodds explains why knowing what to build, understanding the system, and owning outcomes matter as AI agents take on implementation.</description>
    </item>
    <item>
      <title>The Merge: Why Building AI Chat Is Harder Than It Looks</title>
      <link>https://krackedtools.dev/posts/the-merge-why-building-ai-chat-is-harder-than-it-looks.html</link>
      <guid isPermaLink="true">https://krackedtools.dev/posts/the-merge-why-building-ai-chat-is-harder-than-it-looks.html</guid>
      <pubDate>Tue, 25 Aug 2026 00:00:00 +0000</pubDate>
      <description>A conversation with Assistant UI founder Simon Farshid about the hidden work behind AI chat interfaces, product judgment, and his coding-agent loop with CodeRabbit.</description>
    </item>
    <item>
      <title>CodeRabbit CLI 0.7.1: improved performance and reliability</title>
      <link>https://krackedtools.dev/posts/coderabbit-cli-0-7-1-performance-and-reliability.html</link>
      <guid isPermaLink="true">https://krackedtools.dev/posts/coderabbit-cli-0-7-1-performance-and-reliability.html</guid>
      <pubDate>Wed, 19 Aug 2026 00:00:00 +0000</pubDate>
      <description>CodeRabbit CLI 0.7.1 brings faster local reviews, configuration validation, safer agent authentication, and CodeRabbit Skills for coding-agent workflows.</description>
    </item>
    <item>
      <title>Opus 5 for code review: Cleaner actionable comments, noisier overall</title>
      <link>https://krackedtools.dev/posts/opus-5-for-code-review.html</link>
      <guid isPermaLink="true">https://krackedtools.dev/posts/opus-5-for-code-review.html</guid>
      <pubDate>Thu, 06 Aug 2026 00:00:00 +0000</pubDate>
      <description>Opus 5 produced CodeRabbit's cleanest actionable-comment stream, but caught fewer known issues and generated roughly four times the baseline's nitpicks.</description>
    </item>
    <item>
      <title>Building an effective code-review harness</title>
      <link>https://krackedtools.dev/posts/building-an-effective-code-review-harness.html</link>
      <guid isPermaLink="true">https://krackedtools.dev/posts/building-an-effective-code-review-harness.html</guid>
      <pubDate>Wed, 08 Jul 2026 00:00:00 +0000</pubDate>
      <description>What goes inside the system that wraps the model: the twelve harness primitives, a code-review agent as the worked example, and which SDK to reach for when you build one.</description>
    </item>
    <item>
      <title>Loop Engineering: When is it worth the hassle?</title>
      <link>https://krackedtools.dev/posts/loop-engineering-when-is-it-worth-the-hassle.html</link>
      <guid isPermaLink="true">https://krackedtools.dev/posts/loop-engineering-when-is-it-worth-the-hassle.html</guid>
      <pubDate>Wed, 24 Jun 2026 00:00:00 +0000</pubDate>
      <description>Why everyone seemingly can't stop talking about loops, what the new primitives are, when it makes sense to build one, and how you'd go about it.</description>
    </item>
    <item>
      <title>What AI code reviewers catch — and what they still miss</title>
      <link>https://krackedtools.dev/posts/what-ai-code-reviewers-catch-and-what-they-still-miss.html</link>
      <guid isPermaLink="true">https://krackedtools.dev/posts/what-ai-code-reviewers-catch-and-what-they-still-miss.html</guid>
      <pubDate>Mon, 01 Jun 2026 00:00:00 +0000</pubDate>
      <description>A frank account of where AI code review earns its keep, where it quietly fails, and how to close the gap.</description>
    </item>
    <item>
      <title>Why building your own benchmark will get you ahead</title>
      <link>https://krackedtools.dev/posts/why-building-your-own-benchmark-will-get-you-ahead.html</link>
      <guid isPermaLink="true">https://krackedtools.dev/posts/why-building-your-own-benchmark-will-get-you-ahead.html</guid>
      <pubDate>Thu, 23 Apr 2026 00:00:00 +0000</pubDate>
      <description>Public benchmarks won't tell you whether your AI feature is good enough to ship. A practical three-stage guide to building your own evals.</description>
    </item>
    <item>
      <title>A field guide to pre-merge checks</title>
      <link>https://krackedtools.dev/posts/a-field-guide-to-pre-merge-checks.html</link>
      <guid isPermaLink="true">https://krackedtools.dev/posts/a-field-guide-to-pre-merge-checks.html</guid>
      <pubDate>Sun, 01 Feb 2026 00:00:00 +0000</pubDate>
      <description>What pre-merge checks are, how to stack them without slowing teams down, and the one check everyone skips.</description>
    </item>
    <item>
      <title>Carbon-aware CI: greener pipelines, same merge speed</title>
      <link>https://krackedtools.dev/posts/carbon-aware-ci-greener-pipelines-same-merge-speed.html</link>
      <guid isPermaLink="true">https://krackedtools.dev/posts/carbon-aware-ci-greener-pipelines-same-merge-speed.html</guid>
      <pubDate>Mon, 15 Dec 2025 00:00:00 +0000</pubDate>
      <description>How to shift CI pipelines to lower-carbon windows without making developers wait longer to merge.</description>
    </item>
  </channel>
</rss>
