<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>Mirrors resources</title><description>Guides on controlling what production AI agents are told, from Mirrors.</description><link>https://mirrors.dev</link><language>en-us</language><item><title>What is an agent correction? Scoped fixes for AI agents</title><link>https://mirrors.dev/resources/glossary/agent-correction</link><guid isPermaLink="true">https://mirrors.dev/resources/glossary/agent-correction</guid><description>An agent correction is a short instruction an AI agent gets only when its trigger matches. How it works, and how it differs from prompt lines, skills and guardrails.</description><pubDate>Fri, 30 Oct 2026 00:00:00 GMT</pubDate></item><item><title>Why your AI agent keeps making the same mistake in production</title><link>https://mirrors.dev/resources/ai-agent-keeps-making-same-mistake</link><guid isPermaLink="true">https://mirrors.dev/resources/ai-agent-keeps-making-same-mistake</guid><description>AI agents repeat mistakes when a fix lands in the wrong place, nobody groups the failures, or nobody measures the change. How to find the pattern and make a fix stick.</description><pubDate>Fri, 30 Oct 2026 00:00:00 GMT</pubDate></item><item><title>AI agent observability vs evals vs corrections</title><link>https://mirrors.dev/resources/vs/observability-vs-evals-vs-corrections</link><guid isPermaLink="true">https://mirrors.dev/resources/vs/observability-vs-evals-vs-corrections</guid><description>Observability shows what an agent did, evals test a change, and corrections change what it gets in one situation. How they fit with Langfuse, LangSmith and Braintrust.</description><pubDate>Fri, 30 Oct 2026 00:00:00 GMT</pubDate></item><item><title>Self-improving AI agents: what works in production</title><link>https://mirrors.dev/resources/self-improving-agents</link><guid isPermaLink="true">https://mirrors.dev/resources/self-improving-agents</guid><description>What self-improving agents do, where automatic prompt rewriting falls short, and why a review step between a lesson and live behavior is worth keeping.</description><pubDate>Fri, 30 Oct 2026 00:00:00 GMT</pubDate></item><item><title>System prompt too long? What to do instead of adding more rules</title><link>https://mirrors.dev/resources/system-prompt-too-long</link><guid isPermaLink="true">https://mirrors.dev/resources/system-prompt-too-long</guid><description>How to tell when a system prompt has grown too far, what to move out of it, and how to add an instruction only in the situations where it applies.</description><pubDate>Fri, 30 Oct 2026 00:00:00 GMT</pubDate></item></channel></rss>