<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/">
  <channel>
    <title>AI工具 on Panshi | Infrastructure · Quant Trading · AI Workflows</title>
    <link>https://blog.panshi.io/en/tags/ai%E5%B7%A5%E5%85%B7/</link>
    <description>Recent content in AI工具 on Panshi | Infrastructure · Quant Trading · AI Workflows</description>
    <image>
      <title>Panshi | Infrastructure · Quant Trading · AI Workflows</title>
      <url>https://blog.panshi.io/og-default.png</url>
      <link>https://blog.panshi.io/og-default.png</link>
    </image>
    <generator>Hugo</generator>
    <language>en-US</language>
    <lastBuildDate>Wed, 15 Apr 2026 00:00:00 +0000</lastBuildDate>
    <atom:link href="https://blog.panshi.io/en/tags/ai%E5%B7%A5%E5%85%B7/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Advanced Surge Routing Rules: Route AI Traffic through Residential IPs, Other Traffic through VPS Nodes</title>
      <link>https://blog.panshi.io/en/posts/surge-split-rules-ai-traffic/</link>
      <pubDate>Wed, 15 Apr 2026 00:00:00 +0000</pubDate>
      <guid>https://blog.panshi.io/en/posts/surge-split-rules-ai-traffic/</guid>
      <description>AI services like Claude and ChatGPT are sensitive to IP quality, and datacenter IPs easily trigger restrictions. This article explains how to configure fine-grained routing rules in Surge to route AI requests exclusively through residential IP proxies while keeping regular traffic on VPS nodes.</description>
    </item>
    <item>
      <title>OpenClaw &#43; GitHub Copilot usage management and efficiency optimization: 8 practical tips</title>
      <link>https://blog.panshi.io/en/posts/openclaw-copilot-usage-tips/</link>
      <pubDate>Wed, 15 Apr 2026 00:00:00 +0000</pubDate>
      <guid>https://blog.panshi.io/en/posts/openclaw-copilot-usage-tips/</guid>
      <description>Copilot has a monthly request limit, and OpenClaw multi-agent concurrency will quickly consume your quota. This article summarizes 8 practical tips: tracking Copilot usage, optimizing OpenClaw agent configuration, avoiding quota waste, so AI coding tools last longer and work more smoothly.</description>
    </item>
    <item>
      <title>Practical residential IP egress configuration in mihomo: keeping AI tools reliably available</title>
      <link>https://blog.panshi.io/en/posts/mihomo-residential-ip-guide/</link>
      <pubDate>Wed, 15 Apr 2026 00:00:00 +0000</pubDate>
      <guid>https://blog.panshi.io/en/posts/mihomo-residential-ip-guide/</guid>
      <description>Complete tutorial: configuring residential IP proxies as exit nodes in mihomo (Clash Meta) to bypass IP restrictions on AI tools like ChatGPT and Claude, with IPRoyal SOCKS5 integration walkthrough</description>
    </item>
    <item>
      <title>Using Gemini CLI to automatically inspect server clusters: Python &#43; SSH &#43; Telegram push hands-on guide</title>
      <link>https://blog.panshi.io/en/posts/gemini-cli-server-audit/</link>
      <pubDate>Wed, 15 Apr 2026 00:00:00 +0000</pubDate>
      <guid>https://blog.panshi.io/en/posts/gemini-cli-server-audit/</guid>
      <description>Manual SSH into servers every day to check status is too tedious. This article uses Gemini CLI as the analysis engine, with Python scripts automatically SSH collecting metrics from each server, combined with Telegram Bot pushing daily reports, to achieve zero-manual cluster health monitoring.</description>
    </item>
    <item>
      <title>Why do AI tools need residential IPs? Principles and practice</title>
      <link>https://blog.panshi.io/en/posts/ai-tools-residential-proxy/</link>
      <pubDate>Wed, 15 Apr 2026 00:00:00 +0000</pubDate>
      <guid>https://blog.panshi.io/en/posts/ai-tools-residential-proxy/</guid>
      <description>Do you often encounter various restrictions when accessing Claude or ChatGPT via VPS nodes? The root cause is that datacenter IPs are identified as proxy traffic. This article explains the principles of IP quality classification and why residential IPs can bypass these restrictions.</description>
    </item>
    <item>
      <title>How to integrate voice input into your daily workflow: Three core scenarios where I use Typeless</title>
      <link>https://blog.panshi.io/en/posts/typeless-vs-keyboard-workflow/</link>
      <pubDate>Wed, 08 Apr 2026 00:00:00 +0000</pubDate>
      <guid>https://blog.panshi.io/en/posts/typeless-vs-keyboard-workflow/</guid>
      <description>Voice input is more than just &amp;#39;speaking instead of typing&amp;#39;—combined with AI-powered formatting, it becomes a systematic speed multiplier for writing, development, and communication. Here are three real-world scenarios.</description>
    </item>
    <item>
      <title>Typeless Core Features Demystified: How AI Makes Voice Input Actually Useful</title>
      <link>https://blog.panshi.io/en/posts/typeless-privacy-ai-features/</link>
      <pubDate>Wed, 08 Apr 2026 00:00:00 +0000</pubDate>
      <guid>https://blog.panshi.io/en/posts/typeless-privacy-ai-features/</guid>
      <description>Typeless Is More Than Speech-to-Text—Here&amp;#39;s What Its AI Layer Really Does: Eliminating verbal tics, intelligent formatting, privacy safeguards, personalized learning—examining the real-world impact of each feature.</description>
    </item>
    <item>
      <title>Typeless Real Test: AI voice input is 4x faster than keyboard typing, seamless Chinese-English mixing.</title>
      <link>https://blog.panshi.io/en/posts/typeless-ai-voice-input/</link>
      <pubDate>Wed, 08 Apr 2026 00:00:00 +0000</pubDate>
      <guid>https://blog.panshi.io/en/posts/typeless-ai-voice-input/</guid>
      <description>After three weeks using Typeless AI voice input, speed jumped from 45 WPM to 200&#43;, with automatic Chinese-English switching and filler removal—sharing real experience.</description>
    </item>
  </channel>
</rss>
