<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:sy="http://purl.org/rss/1.0/modules/syndication/" xmlns:media="http://search.yahoo.com/mrss/"><channel><title>MCP on Shaowen Chen's Website</title><link>https://www.chenshaowen.com/en/tags/mcp/</link><description>Recent content in MCP on Shaowen Chen's Website</description><generator>Hugo -- gohugo.io</generator><language>en</language><copyright>&amp;copy;2016 - {year}, All Rights Reserved.</copyright><lastBuildDate>Fri, 03 Jul 2026 00:00:00 +0000</lastBuildDate><sy:updatePeriod>weekly</sy:updatePeriod><atom:link href="https://www.chenshaowen.com/en/tags/mcp/atom.xml" rel="self" type="application/rss+xml"/><item><title>Headroom: Making AI Coding Assistants Save More Tokens</title><link>https://www.chenshaowen.com/en/blog/headroom-reduce-ai-agent-tokens.html</link><pubDate>Fri, 03 Jul 2026 00:00:00 +0000</pubDate><atom:modified>Fri, 03 Jul 2026 00:00:00 +0000</atom:modified><guid>https://www.chenshaowen.com/en/blog/headroom-reduce-ai-agent-tokens.html</guid><description>1. What Headroom Is Headroom is a context compression layer for AI agents. Before the LLM receives a request, it performs content-aware compression on tool output, logs, RAG chunks, file contents, and conversation history, cutting token usage to 5%–40% of the original while keeping answer quality as unchanged as possible.</description><dc:creator>微信公众号</dc:creator><category>AI</category><category>MCP</category><category>Cursor</category><category>Headroom</category><category>R&amp;D</category></item><item><title>CodeGraph: Helping AI Coding Assistants Understand the Codebase</title><link>https://www.chenshaowen.com/en/blog/codegraph-help-ai-understand-codebase.html</link><pubDate>Thu, 18 Jun 2026 00:00:00 +0000</pubDate><atom:modified>Thu, 18 Jun 2026 00:00:00 +0000</atom:modified><guid>https://www.chenshaowen.com/en/blog/codegraph-help-ai-understand-codebase.html</guid><description>1. What CodeGraph Is CodeGraph is a local-first code intelligence tool. It parses a codebase with tree-sitter, stores symbols, relationships, and files in a local SQLite database, and exposes them as a queryable knowledge graph through MCP, a CLI, and a TypeScript API.
Put simply, it turns &amp;ldquo;grep, glob, and Read all over the code&amp;rdquo; into an index built ahead of time, so an AI assistant can answer structural questions with just a few queries.</description><dc:creator>微信公众号</dc:creator><category>AI</category><category>MCP</category><category>Cursor</category><category>CodeGraph</category><category>R&amp;D</category></item><item><title>Extending OpenClaw with Skills and MCP</title><link>https://www.chenshaowen.com/en/blog/using-skills-and-mcp-to-extend-openclaw.html</link><pubDate>Sun, 01 Mar 2026 00:00:00 +0000</pubDate><atom:modified>Sun, 01 Mar 2026 00:00:00 +0000</atom:modified><guid>https://www.chenshaowen.com/en/blog/using-skills-and-mcp-to-extend-openclaw.html</guid><description>1. What Skills Are 1 2 3 openclaw skills list Skills (7/51 ready) OpenClaw ships with a number of built-in skills, which you can view with openclaw skills list. The concrete implementations of these Skills can be found at https://github.com/openclaw/openclaw/tree/main/skills. A Skills file mainly describes the Skill&amp;rsquo;s metadata, when it</description><dc:creator>微信公众号</dc:creator><category>AI</category><category>AI Agent</category><category>Skills</category><category>MCP</category><category>OpenClaw</category><category>Operations</category><category>Learning</category></item><item><title>You Can Also Implement AI Agents This Way - The MCP Edition</title><link>https://www.chenshaowen.com/en/blog/using-mcp-to-power-ai-agent.html</link><pubDate>Thu, 29 Jan 2026 00:00:00 +0000</pubDate><atom:modified>Thu, 29 Jan 2026 00:00:00 +0000</atom:modified><guid>https://www.chenshaowen.com/en/blog/using-mcp-to-power-ai-agent.html</guid><description>1. What Is MCP MCP is short for Model Context Protocol, a protocol for passing context between models and applications.
The greatest value of MCP is that, as the model wars are winding down and AI applications are about to explode, it gives developers a unified standard for invoking capabilities beyond the model itself.</description><dc:creator>微信公众号</dc:creator><category>AI</category><category>AI Agent</category><category>MCP</category><category>LLM</category><category>R&amp;D</category></item></channel></rss>