<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:dc="http://purl.org/dc/elements/1.1/">
  <channel>
    <title>Forem: Kay Ade</title>
    <description>The latest articles on Forem by Kay Ade (@kirumachi).</description>
    <link>https://forem.com/kirumachi</link>
    <image>
      <url>https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F3751087%2F79aef2b3-cdc2-4060-a526-aebef3509891.png</url>
      <title>Forem: Kay Ade</title>
      <link>https://forem.com/kirumachi</link>
    </image>
    <atom:link rel="self" type="application/rss+xml" href="https://forem.com/feed/kirumachi"/>
    <language>en</language>
    <item>
      <title>JSON is not a User Interface (Obviously): Why I built a TUI for Local Security</title>
      <dc:creator>Kay Ade</dc:creator>
      <pubDate>Tue, 03 Feb 2026 17:16:02 +0000</pubDate>
      <link>https://forem.com/kirumachi/json-is-not-a-user-interface-obviously-why-i-built-a-tui-for-local-security-4mj</link>
      <guid>https://forem.com/kirumachi/json-is-not-a-user-interface-obviously-why-i-built-a-tui-for-local-security-4mj</guid>
      <description>&lt;p&gt;Security tools (SAST/SCA) can be great at finding bugs but terrible at talking to humans.&lt;/p&gt;

&lt;blockquote&gt;
&lt;p&gt;Scenario: You run a scan.&lt;br&gt;
Result: A 4MB JSON file or 500 lines of unreadable terminal logs.&lt;br&gt;
Action: You ignore it until CI breaks.&lt;/p&gt;
&lt;/blockquote&gt;

&lt;p&gt;The Solution (Kekkai): I built a terminal-based "Inbox" for security findings. It unifies Trivy (CVEs), Semgrep (Code), and Gitleaks (Secrets) into a single TUI.&lt;/p&gt;

&lt;p&gt;Key Features in v2.2:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;Unified TUI: Navigate findings with j/k.&lt;/li&gt;
&lt;li&gt;Code Context: Press Enter to see the actual vulnerable code snippet + syntax highlighting right in the terminal. No context switching.&lt;/li&gt;
&lt;li&gt;Local AI: Pipe the finding to Ollama (running locally) to ask "Is this a false positive?"&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;I didn't want to configure another CI pipeline manually. So I built a self-replicating init command.&lt;/p&gt;

&lt;p&gt;Run this in your repo: &lt;code&gt;kekkai init --ci&lt;/code&gt;&lt;/p&gt;

&lt;p&gt;It detects your git root and auto-generates a GitHub Actions workflow that blocks the build on "High" severity findings.&lt;/p&gt;

&lt;p&gt;Try it locally: pipx install kekkai-cli -&amp;gt; kekkai scan -&amp;gt; kekkai triage&lt;/p&gt;

&lt;p&gt;&lt;a href="https://github.com/kademoslabs/kekkai" rel="noopener noreferrer"&gt;https://github.com/kademoslabs/kekkai&lt;/a&gt;&lt;/p&gt;

</description>
      <category>cli</category>
      <category>security</category>
      <category>showdev</category>
      <category>tooling</category>
    </item>
  </channel>
</rss>
