<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>Agentic Horizon · News briefs</title><link>https://agentichorizon.blog/</link><description>Original briefs and practical context on AI agents.</description><language>en</language><item><title>Microsoft puts prospecting agents into sales workflows</title><link>https://agentichorizon.blog/news/microsoft-sales-development-agent/</link><guid isPermaLink="true">https://agentichorizon.blog/news/microsoft-sales-development-agent/</guid><pubDate>Thu, 08 Oct 2026 00:00:00 GMT</pubDate><category>Workplace</category><description>Sales Development Agent is in public preview, with autonomous outreach and follow-up. Copilot credit consumption starts on 16 November.</description></item><item><title>GitHub releases a benchmark for code-review agents</title><link>https://agentichorizon.blog/news/github-reviewbench/</link><guid isPermaLink="true">https://agentichorizon.blog/news/github-reviewbench/</guid><pubDate>Thu, 08 Oct 2026 00:00:00 GMT</pubDate><category>Coding</category><description>ReviewBench tests AI reviewers on 219 pull requests across 19 languages, measuring useful findings as well as missed issues.</description></item><item><title>OpenAI&apos;s model guide focuses on completed agent tasks</title><link>https://agentichorizon.blog/news/openai-agent-model-guide/</link><guid isPermaLink="true">https://agentichorizon.blog/news/openai-agent-model-guide/</guid><pubDate>Thu, 08 Oct 2026 00:00:00 GMT</pubDate><category>Tools</category><description>The GPT-6 guide recommends measuring task success, latency and cost, with explicit boundaries for long-running work.</description></item></channel></rss>