<?xml version="1.0" encoding="UTF-8"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Suniworks journal</title><link>https://suniworks.com/en/blog</link><description>Software, security and practical engineering for businesses.</description><language>en</language><atom:link href="https://suniworks.com/en/blog/rss.xml" rel="self" type="application/rss+xml"/><item><title>Personal ChatGPT at work: keeping control of company data</title><link>https://suniworks.com/en/blog/personal-chatgpt-at-work</link><guid isPermaLink="true">https://suniworks.com/en/blog/personal-chatgpt-at-work</guid><description>Define permitted AI use, choose an appropriate service and prepare practical controls before employees share company information.</description><pubDate>Sat, 17 Oct 2026 00:00:00 GMT</pubDate></item><item><title>GhostAction: review the workflow, not just the application</title><link>https://suniworks.com/en/blog/ghostaction-how-github-actions-secrets-are-stolen</link><guid isPermaLink="true">https://suniworks.com/en/blog/ghostaction-how-github-actions-secrets-are-stolen</guid><description>Deployment workflows can reach powerful credentials. Define review responsibilities, limit access and prepare a response to a compromised pipeline.</description><pubDate>Fri, 16 Oct 2026 00:00:00 GMT</pubDate></item><item><title>AI-generated code: who takes responsibility?</title><link>https://suniworks.com/en/blog/ai-generated-code-the-human-accountability-rule</link><guid isPermaLink="true">https://suniworks.com/en/blog/ai-generated-code-the-human-accountability-rule</guid><description>A useful lesson from the Linux kernel: someone must review, understand and take responsibility for the code a team delivers.</description><pubDate>Fri, 16 Oct 2026 00:00:00 GMT</pubDate></item><item><title>AI and reverse engineering: what REA means for your software</title><link>https://suniworks.com/en/blog/ai-reverse-engineering-mcp-rea</link><guid isPermaLink="true">https://suniworks.com/en/blog/ai-reverse-engineering-mcp-rea</guid><description>Agents can assist application analysis. Separate tools, hypotheses and verified evidence before making architecture or migration decisions.</description><pubDate>Thu, 15 Oct 2026 00:00:00 GMT</pubDate></item></channel></rss>