<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Github-Actions on Onur Celep</title><link>https://onurcelep.github.io/tags/github-actions/</link><description>Recent content in Github-Actions on Onur Celep</description><generator>Hugo</generator><language>en-US</language><lastBuildDate>Sat, 11 Jul 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://onurcelep.github.io/tags/github-actions/index.xml" rel="self" type="application/rss+xml"/><item><title>The green check that did nothing: a postmortem of AI agents in CI</title><link>https://onurcelep.github.io/posts/ai-agents-in-ci-silent-failures/</link><pubDate>Sat, 11 Jul 2026 00:00:00 +0000</pubDate><guid>https://onurcelep.github.io/posts/ai-agents-in-ci-silent-failures/</guid><description>&lt;p&gt;I run AI coding agents in CI across my repositories. The loop looks like
this: I open a GitHub issue describing a change and tag the agent; it reads
the repo, makes the change on a branch, and hands me a pull request. A
second agent reviews every PR automatically before I look at it. A human
merges everything. The setup is stamped from a single template repo
(&lt;a href="https://github.com/onurcelep/ai-factory" class="external-link" target="_blank" rel="noopener"&gt;ai-factory&lt;/a&gt;) so all my projects
run the same configuration — the full tour of that setup and how it came
to be is &lt;a href="https://onurcelep.github.io/posts/how-i-work-with-ai-agents/" &gt;its own post&lt;/a&gt;; this
one is about the day it broke.&lt;/p&gt;</description></item></channel></rss>