<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Python on coles.codes</title><link>https://coles.codes/tags/python/</link><description>Recent content in Python on coles.codes</description><generator>Hugo</generator><language>en-us</language><lastBuildDate>Sun, 07 Jun 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://coles.codes/tags/python/index.xml" rel="self" type="application/rss+xml"/><item><title>Building lgtmaybe: a PR reviewer for any model</title><link>https://coles.codes/posts/building-lgtmaybe/</link><pubDate>Sun, 07 Jun 2026 00:00:00 +0000</pubDate><guid>https://coles.codes/posts/building-lgtmaybe/</guid><description>&lt;p>I built a PR reviewer called &lt;a href="https://github.com/MattJColes/lgtmaybe">lgtmaybe&lt;/a>, which is the joke I wanted in the name before I&amp;rsquo;d even started writing the code. In terms of how it works, you point it at a pull request, pick a model with one flag, and it posts inline comments plus a summary. A clean PR gets a 👍 LGTM, and everything else gets a maybe.&lt;/p>
&lt;p>The thing I&amp;rsquo;m happiest with is that you can run it on OpenAI, Anthropic, OpenRouter, Bedrock, Vertex, or a local Ollama box.&lt;/p></description></item></channel></rss>