<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Speak Well in Heaven on ash's Writing Archive</title><link>https://rwen.me/en/tags/speak-well-in-heaven/</link><description>Recent content in Speak Well in Heaven on ash's Writing Archive</description><generator>Hugo -- gohugo.io</generator><language>en</language><copyright>lantao studio</copyright><lastBuildDate>Thu, 01 Oct 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://rwen.me/en/tags/speak-well-in-heaven/index.xml" rel="self" type="application/rss+xml"/><item><title>I Asked AI to Make an Indie Game Just for Me</title><link>https://rwen.me/en/p/speak-well-in-heaven/</link><pubDate>Thu, 01 Oct 2026 00:00:00 +0000</pubDate><guid>https://rwen.me/en/p/speak-well-in-heaven/</guid><description>&lt;img src="https://rwen.me/p/speak-well-in-heaven/images/5.png" alt="Featured image of post I Asked AI to Make an Indie Game Just for Me" />&lt;p>My feed has been full of posts claiming, “I one-shotted a game with a single prompt.” The claim is that the AI made the whole game in a single session. Take those posts with a grain of salt, of course. Still, the games AI can turn out on its own are getting noticeably better.&lt;/p>
&lt;p>Most of these games, though, are riffs on something that already exists—&lt;em>Stardew Valley&lt;/em>, say. That&amp;rsquo;s a neat technical feat, but it isn&amp;rsquo;t much of a creative one.&lt;/p>
&lt;p>A couple of days ago, I wondered what would happen if I asked AI to dream up a &lt;strong>completely new game&lt;/strong>, built around my tastes, and make the whole thing from scratch.&lt;/p>
&lt;p>I took my Douban gaming profile and turned it into a table: more than 300 rows of games, ratings, and short reviews. Then I put the table in an empty repo and gave the AI its assignment: “&lt;strong>Make an indie game I&amp;rsquo;d give five stars.&lt;/strong>”&lt;/p>
&lt;p>The actual prompt was much longer—a rambling voice note turned into text, which I&amp;rsquo;ll spare you. The important part was this: based on my ratings and reviews, the AI had to come up with its own definition of a five-star game, then keep working until it met that standard. I left the criteria entirely up to it. My only other rule was: “Feel free to borrow and combine mechanics from different games, but don&amp;rsquo;t copy any one game.”&lt;/p>
&lt;p>So the AI wasn&amp;rsquo;t just building a game to my specifications. It was &lt;strong>coming up with the game itself&lt;/strong>.&lt;/p>
&lt;p>Its first reply stopped me in my tracks:&lt;/p>
&lt;blockquote>
&lt;p>My game: &lt;em>Speak Well in Heaven&lt;/em>&lt;/p>
&lt;p>You are the Kitchen God who lives in a family&amp;rsquo;s stove. Every Little New Year&amp;rsquo;s Eve—the 23rd day of the last lunar month—you go up to heaven and tell the Jade Emperor what the family has done, good and bad, over the past year.&lt;/p>
&lt;p>On your last night before the trip, you can revisit the year and scrub freely through time. But you can only see where there&amp;rsquo;s firelight, and every lamp you light changes how the people around it behave.&lt;/p>
&lt;p>(The observer effect is the central mechanic: to see what really happened, you need to light a lamp. Light it, and anyone up to no good stops what they&amp;rsquo;re doing.)&lt;/p>
&lt;ul>
&lt;li>One house, one stove, one family, and six Little New Year&amp;rsquo;s Eves spanning 1912 to 2026. Fire plays a different role in each era: wartime blackouts, scarce lamp oil in a famine, your territory shrinking as electric lights take over, until all you have left is a lighter and a doorbell camera…&lt;/li>
&lt;li>At the end of each night, you fill out a report and work out the truth for yourself. The game won&amp;rsquo;t hand you the answer.&lt;/li>
&lt;li>Once you know what happened, you choose: tell heaven the truth, or sweeten your lips with Kitchen God candy and put in a good word for the family.&lt;/li>
&lt;li>The look: a shadow-puppet stage, with the house shown in cut-paper cross-section. You see only what the firelight touches.&lt;/li>
&lt;li>Four to five hours long. No randomness, no grinding, and a timeline you can jump around at will.&lt;/li>
&lt;/ul>
&lt;/blockquote>
&lt;p>&lt;img src="https://rwen.me/p/speak-well-in-heaven/images/1.png"
width="1958"
height="1236"
loading="lazy"
alt="Game title screen"
class="gallery-image"
data-flex-grow="158"
data-flex-basis="380px"
>&lt;/p>
&lt;p>I was sold. A Chinese-folklore take on &lt;em>The Case of the Golden Idol&lt;/em>, with a century&amp;rsquo;s worth of cause and effect? Offering sticky candy to the Kitchen God is a real folk custom, and it gives the player a delicious little dilemma: what do you tell heaven? The shadow-puppet style is smart, too. It can look striking without demanding elaborate character animation, and keeping the story in one house saves a mountain of background art.&lt;/p>
&lt;p>It was a good idea &lt;em>and&lt;/em> a plausible one to make.&lt;/p>
&lt;p>To be clear, the AI came up with the game on its own. Most of my “involvement” was starting a fresh chat whenever my five-hour quota reset and saying, “Keep going.” I stepped in only twice during development:&lt;/p>
&lt;ol>
&lt;li>Claude&amp;rsquo;s vector art looked rough, so I suggested it write an art brief for Codex, which could generate images. I helped the two AI sessions pass the 2D assets back and forth.&lt;/li>
&lt;li>Neither AI could listen to audio directly, so I checked a few of Claude&amp;rsquo;s programmatically generated effects myself—matches striking, knocks on a door—and directed it to Freesound for free music and sound effects that didn&amp;rsquo;t require attribution.&lt;/li>
&lt;/ol>
&lt;p>Twenty-six hours of work and roughly $645 worth of tokens later, I had a game made just for me. My subscriptions kept my actual cost under $30. I spent maybe an hour or two playing it—I wasn&amp;rsquo;t watching the clock—and, to my surprise, never hit a progress-blocking bug.&lt;/p>
&lt;p>&lt;img src="https://rwen.me/p/speak-well-in-heaven/images/2.png"
width="1989"
height="1297"
loading="lazy"
alt="Main game screen"
class="gallery-image"
data-flex-grow="153"
data-flex-basis="368px"
>&lt;/p>
&lt;p>&lt;img src="https://rwen.me/p/speak-well-in-heaven/images/3.png"
width="1991"
height="1306"
loading="lazy"
alt="Answer submission screen"
class="gallery-image"
data-flex-grow="152"
data-flex-basis="365px"
>&lt;/p>
&lt;p>&lt;img src="https://rwen.me/p/speak-well-in-heaven/images/4.png"
width="1984"
height="1303"
loading="lazy"
alt="Some visual details are off, such as the position of this hand"
class="gallery-image"
data-flex-grow="152"
data-flex-basis="365px"
>&lt;/p>
&lt;p>My verdict: &lt;strong>a terrific idea, unevenly executed. In the pre-AI days, I could imagine a high school student making something like this in two months.&lt;/strong>&lt;/p>
&lt;p>I could spot its influences everywhere: &lt;em>The Case of the Golden Idol&lt;/em>, &lt;em>Ghost Trick&lt;/em>, &lt;em>Unheard&lt;/em>, &lt;em>This War of Mine&lt;/em>, &lt;em>Pentiment&lt;/em>. I could also see exactly where its ambition outran its ability. A lot of its mistakes felt familiar from Chinese indie games I&amp;rsquo;ve tried. In that sense, it plays like the work of a beginner—and beginners have room to get a lot better.&lt;/p>
&lt;p>When I finished, I left a long voice note with feedback, mostly pointing out obvious visual mistakes. They were easy enough to fix; I&amp;rsquo;m not sure why they survived the first pass. The version below has those fixes, so it&amp;rsquo;s a bit better than the one the AI made without supervision. Plenty of problems remain, especially poor guidance and too much writing that sounds unmistakably AI-generated. If you play, expect to do some figuring out on your own:&lt;/p>
&lt;p>&lt;a class="link" href="https://speak-heaven.rwen.me" target="_blank" rel="noopener"
>Play &lt;em>Speak Well in Heaven&lt;/em> online (Chinese only)&lt;/a>&lt;/p>
&lt;p>It works on a phone in landscape mode, but I&amp;rsquo;d recommend playing on a computer. I haven&amp;rsquo;t tested the mobile version from start to finish, so you may run into issues. And to repeat: apart from the feedback I mentioned, I had no hand in making this game. It&amp;rsquo;s rough. If you try it, tell me what you think—good or bad—and I&amp;rsquo;ll pass your comments along to its actual author, Claude Opus 5.5.&lt;/p>
&lt;p>Then again, maybe the smarter move is to skip the game made for me and go tell your own AI what &lt;em>you&lt;/em> want :)&lt;/p>
&lt;p>&lt;img src="https://rwen.me/p/speak-well-in-heaven/images/5.png"
width="1731"
height="975"
loading="lazy"
alt="Heavenly Registrar of Fates"
class="gallery-image"
data-flex-grow="177"
data-flex-basis="426px"
>&lt;/p>
&lt;p>One last thought. For now, the AI economy still runs on an old playbook: people who know how to use AI make things and sell them to people who want those things but don&amp;rsquo;t know how to make them. AI simply lets the sellers work faster. But what happens when AI becomes capable and independent enough to meet each person&amp;rsquo;s needs directly, from idea to finished product?&lt;/p>
&lt;blockquote>
&lt;p>Written in Chinese, translated by GPT 6.1 Sol&lt;/p>
&lt;/blockquote></description></item></channel></rss>