Developer Uses Evener Agent Harness to Autonomous Vibe-Code a Swift-Based C Compiler Capable of Building SQLite

Developer Uses Evener Agent Harness to Autonomous Vibe-Code a Swift-Based C Compiler Capable of Building SQLite Using the Evener agentic harness paired with GLM 5.2, an autonomous multi-subagent process generated an ARM64 macOS C compiler in Swift over a 21-hour run, successfully compiling the SQLite amalgamation and passing a basic smoke test. The author instructed Evener to build a standards-compliant ARM64 C compiler for macOS in Swift with full autonomy and recursive subagents to manage complexity. Over roughly 21 hours, the model autonomously resolved malloc issues, implemented variadic functions, compared its assembly output against Clang, and debugged SQLite compilation errors. The harness's web fetch tool was broken during the run, confirming the model did not search the web for solutions. While able to compile SQLite and execute basic INSERT and SELECT operations, the compiler still lacks many features and is not yet fully standards-compliant. Autonomous Prompting and Architecture The project was launched from a mobile phone to test tool use and the `/loop` feature in Evener. The author set a goal directing the system to autonomously implement a standards-compliant ARM64 C compiler for macOS in modern Swift using recursive subagents for task decomposition, targeting the ability to build SQLite. Progression and Debugging Throughout a 21-hour window, the agent tackled various implementation hurdles. It was observed debugging malloc behaviors, implementing variadic functions, analyzing differences between its own generated assembly and output from Clang, writing test scripts, and troubleshooting segmentation faults while attempting to compile the 274k-line SQLite amalgamation file. Results and Next Steps The autonomous run halted after successfully compiling SQLite and executing basic INSERT and SELECT commands. Because the author had inadvertently disabled the web fetch tool, the system operated without external web lookups. While many features are currently missing and full standards compliance remains incomplete, a follow-up goal has been initiated to run compliance test suites and address missing capabilities.

作者使用Evener与GLM 5.2在21小时内生成可编译SQLite的C编译器 作者利用智能体运行框架Evener搭配GLM 5.2进行自主编程实验，要求其使用现代Swift语言编写适用于macOS ARM64的C编译器。在没有人工干预的情况下，智能体历时约21小时成功构建出能编译SQLite并完成基本数据操作的编译器。 作者向智能体下达完全自主任务，要求通过递归子智能体构建能够编译并运行SQLite测试的Swift版本C编译器。 智能体在约21小时内自主排查了内存分配错误，实现了可变参数函数，并通过比对clang汇编输出进行调试。 编译器最终成功构建了27.4万行的sqlite.c代码，并完成了INSERT和SELECT基础冒烟测试。 实验期间联网检索工具因配置错误失效，智能体未借助外部搜索；目前编译器尚未完全符合规范，后续测试已开启。 任务设定与自主执行模式 作者为测试其智能体框架Evener的循环与工具调用能力，通过手机向系统下达提示词，要求其在macOS上用现代Swift实现符合标准的ARM64 C编译器。作者明确指示智能体完全自主运作、无需回传询问，并允许其使用递归子智能体对任务进行彻底分解以应对上下文复杂性，最终目标是能够编译SQLite并使其通过测试。 21小时的自动化调试与进展 在作者仅偶尔查看进度的情况下，智能体持续自主工作。它先后深入调试了malloc内存问题，实现了可变参数函数，并编写测试脚本。在尝试编译27.4万行的sqlite.c文件遭遇段错误及执行INSERT失败后，智能体会导出自身生成的汇编代码与clang的汇编输出进行差异比对，逐步解决问题。 达成成果与后续完善计划 经历约21小时后，目标循环因完成既定任务而停止：编译器成功编译了SQLite，并顺利执行了INSERT与SELECT操作。作者检查发现，由于工具损坏，智能体期间完全无法使用网页检索。虽然当前版本缺少大量特性且尚未完全符合语言标准，但作者已启动新的循环任务，以推进合规性测试集并补齐缺失功能。

A while back, Claude Opus [built a pretty ambitious compiler.](https://github.com/anthropics/claudes-c-compiler/issues/1) I didn't set out to do anything nearly as ambitious as Anthropic's. They were targeting multiple CPU architectures and they wanted to be able to compile a bootable Linux kernel. (I'd actually forgotten about that project until I started posting some screenshots of the work below to social media and someone reminded me.) Last night, as I was getting ready for bed I was reaching for some project to `/loop` support and tool use in Evener, our \~new agentic harness. I popped open the mobile UI on my phone and typed "Your job is to implement a standards-compliant ARM64 C compiler for macOS in Swift." Evener asked me a couple of questions. I clarified my intent a little bit: "I'm great with radical task decomposition. You should use recursive subagents to manage context and complexity.Should structure the project in whatever sane way you want. You should work fully autonomously And do not need to ask me questions." And then I set a goal: "Implement a standards-compliant C compiler in modern Swift. It should be able to build SQLIte and have SQLite pass all tests. You may decompose the project in any way you see fit. You should use subagents, including recursive subagents, to execute effectively." I don't know what I thought was going to happen. I watched long enough to make sure that it was going to start working. At about four in the morning I rolled over and looked at my phone and it was deep into debugging a malloc issue. I woke up at a little bit before seven and it was implementing variadic functions. Today I spent most of my day in meetings and so I didn't get a lot of time to watch it, but I would occasionally pull up a session and see it making little bits of progress. Occasionally it would dump out some assembler and then dump out `clang`'s version of the assembler to hunt for differences. I saw it trying to build the 274k `sqlite.c`amalgamation and segfaulting. I saw it writing little test scripts. I saw it get to the point of trying to do an `INSERT` and exploding. When I came up for air at about 8 30 p.m. I popped open Evener around my phone, and was incredibly disappointed. My `/goal` loop had just...stopped. It's not supposed to do that. But that's exactly the failure that I was testing for. I started reading the recent parts of the log, to figure out *which* obvious problem had caused the failure. I was *not* expecting to see this: It stopped because it had compiled SQLite and was able to do an `INSERT` and a `SELECT`. It did exactly what I asked. It took Evener + GLM 5.2 about 21 hours to build a C compiler capable of compiling SQLite and passing a basic smoke test. This is the checkpoint: [https://github.com/obra/toy-c-compiler/commit/456ddfa8912b5ba47b968773ecddcdca741f0df2](https://github.com/obra/toy-c-compiler/commit/456ddfa8912b5ba47b968773ecddcdca741f0df2) (I was initially worried that it had cheated by doing the obvious web searches, but discovered late in the day that I had accidentally broken the harness's web fetch tool. And so it wasn't able to cheat that way.) Checking now, it looks like it didn't even *try*, which makes me happy. There are a ton of missing features. It is not yet standards compliant, but I didn't properly set the goal to force that. I've just kicked off another `/goal` to get it to run through compliance suites and fill in the missing features. I'll probably let it run for another day or two just to see what happens.