<?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>Mistral on Dev Digest · Daily Tech Curation</title>
    <link>https://jerryni.github.io/dev-digest/en/tags/mistral/</link>
    <description>Recent content in Mistral on Dev Digest · Daily Tech Curation</description>
    <generator>Hugo</generator>
    <language>en</language>
    <lastBuildDate>Thu, 30 Apr 2026 07:00:00 +0900</lastBuildDate>
    <atom:link href="https://jerryni.github.io/dev-digest/en/tags/mistral/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>April 30 · Today&#39;s 10 Dev Picks</title>
      <link>https://jerryni.github.io/dev-digest/en/posts/2026-04-30/</link>
      <pubDate>Thu, 30 Apr 2026 07:00:00 +0900</pubDate>
      <guid>https://jerryni.github.io/dev-digest/en/posts/2026-04-30/</guid>
      <description>A telling Thursday on HN. Zed 1.0 lands with the day&amp;rsquo;s highest score (1319 pts, HN #2) — but the #1 slot belongs to a GitHub issue inside anthropics/claude-code: a config file called HERMES.md, an internal A/B experiment, double-charged a customer ~$200 and Anthropic refused the refund (836 pts). HN #13 &amp;lsquo;We need a federation of forges&amp;rsquo; (488 pts) keeps yesterday&amp;rsquo;s Ghostty-leaving-GitHub thread burning. AI: Mistral Medium 3.5 ships remote agents (HN #24). Security double feature: Copy Fail / CVE-2026-31431 (HN #3) and Ramp&amp;rsquo;s Sheets AI exfiltrating financials via prompt injection (HN #9). Japan&amp;rsquo;s Zenn delivers two strong Claude Code field reports — LayerX&amp;rsquo;s &amp;lsquo;ccgate&amp;rsquo; auto-approves 97% of permission prompts; Finatext warns against handing Google accounts to AI agents. China&amp;rsquo;s V2EX is consumed by Xiaomi&amp;rsquo;s MiMo Token Plan.</description>
    </item>
  </channel>
</rss>
