{"slug": "from-prompt-to-wechat-dev-build-in-one-day-tests-and-evidence-boundaries", "title": "From Prompt to WeChat Dev Build in One Day: Tests and Evidence Boundaries", "summary": "A developer used AI as the primary development interface to build Weiguang Xinyu, a WeChat mini game, in one day. The project resulted in a working ten-level Canvas2D puzzle with an original soundtrack, passing 13/13 local tests, and a successfully uploaded development build. The developer emphasized that each stage ended with observable evidence, such as remote source readback, rather than relying on AI-generated code alone.", "body_md": "I recently used AI as the primary development interface for a small WeChat mini game called **Weiguang Xinyu**.\n\nThe one-day result was a working ten-level Canvas2D puzzle, an original background track, a passing local test suite, and a successfully uploaded development build. The public release followed later.\n\nThis post focuses on the technical workflow and the evidence boundary between those two stages.\n\nThe mechanic is compact: rotate path segments to connect a light source to a destination.\n\nThe build includes:\n\nThe first version did not depend on a cloud environment, external server, personal-data API, advertising, or payments. That kept the test surface controlled enough for a one-day experiment.\n\nI used voice as the primary interface and treated AI as a coordinated implementation system rather than a one-shot code generator.\n\n``` php\nrequirement\n  -> implementation\n  -> local test\n  -> visual inspection\n  -> packaging check\n  -> development upload\n  -> remote source readback\n```\n\nEach loop ended with an observable artifact. A plausible answer in a chat window did not count as completion.\n\nThe game uses Canvas2D for rendering and separates the core puzzle state from platform-specific behavior.\n\n```\ngame state + level data\n        |\n        +-- input and rotation logic\n        +-- path connectivity checks\n        +-- progress persistence\n        +-- renderer\n        +-- audio controller\n                |\n                +-- browser adapter\n                +-- WeChat adapter\n```\n\nThis separation mattered because local browser previews and the WeChat runtime do not expose identical audio APIs or lifecycle behavior.\n\nThe music starts only after the first user interaction, loops when enabled, pauses in the background or during interruptions, and stores the user's preference. Audio failure does not block gameplay.\n\nThe final local suite reported **13/13 tests passing**. The checks covered the original puzzle behavior together with the music controller and platform adapters.\n\nThe upload alone was not treated as proof that the intended source had reached the platform. After the official development-build upload succeeded, the remote source map was read back. Thirteen remote source files matched the local source files.\n\nThat is useful evidence, but it has limits. It does not replace real-device testing, prove every binary asset is byte-identical, or mean the build is already public at that moment.\n\n| State | Evidence | \n|---|---|\n| Local prototype | The project runs locally and the core interaction works | \n| Tested build | Defined tests pass and the relevant UI is inspected | \n| Uploaded development build | The platform accepts the build and the result is read back where possible | \n| Public release | Review is approved and ordinary users can access the production version | \n\nThe process screenshot captured the tested-and-uploaded stage. The project later reached public release.\n\nAI was especially effective at keeping related changes synchronized:\n\nThe speed came from many small verified loops, not from a single giant prompt.\n\nThe hardest decisions were not syntax decisions. They were questions such as:\n\nAI reduced implementation cost. It did not remove product judgment or release responsibility.\n\nYes, AI can help one person build a small game in a day.\n\nThe more useful claim is narrower: AI can compress the path from idea to a tested, uploaded development build when the scope is controlled and every stage ends with evidence. The release can then follow through its own verified steps.\n\nThat is already a major change. Just do not call the development upload a launch before the later launch is actually complete.", "url": "https://wpnews.pro/news/from-prompt-to-wechat-dev-build-in-one-day-tests-and-evidence-boundaries", "canonical_source": "https://dev.to/leoautomotiveai/from-prompt-to-wechat-dev-build-in-one-day-tests-and-evidence-boundaries-1kf3", "published_at": "2026-09-07 10:16:57+00:00", "updated_at": "2026-09-07 10:27:48.917527+00:00", "lang": "en", "topics": ["artificial-intelligence", "developer-tools", "ai-tools"], "entities": ["WeChat", "Weiguang Xinyu", "Canvas2D"], "alternates": {"html": "https://wpnews.pro/news/from-prompt-to-wechat-dev-build-in-one-day-tests-and-evidence-boundaries", "markdown": "https://wpnews.pro/news/from-prompt-to-wechat-dev-build-in-one-day-tests-and-evidence-boundaries.md", "text": "https://wpnews.pro/news/from-prompt-to-wechat-dev-build-in-one-day-tests-and-evidence-boundaries.txt", "jsonld": "https://wpnews.pro/news/from-prompt-to-wechat-dev-build-in-one-day-tests-and-evidence-boundaries.jsonld"}}