{"id":2493,"date":"2026-08-07T12:39:08","date_gmt":"2026-08-07T07:09:08","guid":{"rendered":"https:\/\/sharanamshah.com\/mymusings\/?page_id=2493"},"modified":"2026-08-07T20:22:30","modified_gmt":"2026-08-07T14:52:30","slug":"the-48-hour-agent-turn-code-into-docs-with-python-mcp-and-a2a","status":"publish","type":"page","link":"https:\/\/sharanamshah.com\/mymusings\/the-weekend-developer-series-on-kindle\/the-48-hour-agent-turn-code-into-docs-with-python-mcp-and-a2a\/","title":{"rendered":"The 48-Hour Agent: Turn Code into Docs with Python, MCP, and A2A"},"content":{"rendered":"\n
\n
\"\"<\/figure>\n\n\n\n
\"\"<\/figure>\n<\/figure>\n\n\n\n

A hands-on Python guide to building two independent AI agents \u2014 one using the Model Context Protocol (MCP), one using Agent2Agent (A2A) \u2014 that work together to keep your documentation honest, automatically.<\/em><\/p>\n\n\n\n

Your Documentation Is Wrong Right Now \u2014 And It’s Nobody’s Fault<\/h2>\n\n\n\n

Somewhere in a repository you maintain, a README describes an endpoint that got renamed three sprints ago, a config flag that no longer exists, or a setup step nobody has needed since the project switched build tools. Nobody lied when they wrote it. The code simply moved on, the way code always does, and the sentence describing it stayed exactly where it was written \u2014 quietly getting less true with every merged pull request.<\/p>\n\n\n\n

You’ve probably tried the obvious fix: point a script at the repository, open every file, and ask a language model to summarize what it finds. It works \u2014 once. Run it again next week and it regenerates the same documentation from scratch, discarding anything a human hand-tuned, blind to the one thing that actually matters: what changed since the last time anyone looked.<\/p>\n\n\n\n

The 48-Hour Agent<\/strong> teaches you a better shape for this problem \u2014 one that doesn’t involve a single script trying to do everything. It shows you how to build two small, independent AI agent processes<\/strong>: one that reads your code and notices exactly what changed, and a second, completely separate one whose only job is to turn that change into a real, drafted documentation update. Neither one needs to be trusted with the other’s job \u2014 and that separation, not a cleverer prompt, is what makes the system trustworthy enough to run again and again, after every real commit.<\/p>\n\n\n\n

This is a Python AI agent development book<\/strong> built around two of the most important emerging protocols in applied AI engineering: the Model Context Protocol (MCP)<\/strong>, which standardizes how an agent reaches tools, and Agent2Agent (A2A)<\/strong>, which standardizes how one independent agent hands off real work to another over HTTP. If you’ve heard these terms and want to actually build<\/em> with them instead of reading a specification, this book is written for you.<\/p>\n\n\n\n

\n\n\n\n

What You’ll Build: CodeScribe<\/h2>\n\n\n\n

This isn’t a wrapped API call, and it isn’t a toy script. Over one weekend, you build CodeScribe<\/strong> \u2014 a real, two-process documentation agent designed to be genuinely useful after the book ends.<\/p>\n\n\n\n

Node 1: The Workspace Scanner (package codescribe<\/code>)<\/h3>\n\n\n\n

Runs entirely on your own machine. It reads your source files, your existing documentation, and your local git diff through two official MCP servers \u2014 and it never drafts a single sentence itself.<\/p>\n\n\n\n

Node 2: The Technical Writer (package scribewriter<\/code>)<\/h3>\n\n\n\n

Runs as its own independent process, reachable only over HTTP through A2A. It publishes a signed Agent Card, accepts a Task carrying your code and your diff, and hands back a drafted Markdown update \u2014 with no filesystem access of its own, ever.<\/p>\n\n\n\n

By the end of the book, you run codescribe update<\/code> against a real repository of your own and watch a Markdown file update itself, drafted by a second, independent AI agent process, over a genuine protocol handshake.<\/p>\n\n\n\n

\n\n\n\n

Why “48 Hours”?<\/h2>\n\n\n\n

This book carries the name The 48-Hour Agent<\/strong> for a reason that has nothing to do with marketing. Forty-eight hours is the real, honest time budget every chapter is written against, and it splits cleanly down the middle of the two protocols you build:<\/p>\n\n\n\n