AI Decision System
Replayable, offline-verifiable records for AI-assisted go/no-go decisions: experiment promotions, typed model answers and human-approved actions.
翻译准备中
尚未检测到更新。GitHub 发布版本后会显示在这里。
个人 GitHub 账号下公开仓库的所有者可以登录并验证所有权,自行管理介绍,无需管理员批准。
登录并生成验证文件,在 30 分钟内将其提交到此仓库的默认分支。其他分支或分叉仓库中的文件无效。然后在下方检查文件,即可无需管理员批准直接管理项目。请保留文件以维持管理权限。
如需登录,将前往登录页面。登录后请回到此处,再次点击所选的验证按钮。
在项目文件夹中运行 /nomorevibe verify,即可验证所有权并自行管理介绍。
一分钟完成验证148AI
翻译准备中
取自仓库 README
正在加载完整 README。
查看完整 README ↗Deployment repository for the current 148AI V28 online review environment.
omni_prod_runtime.zipVercel runs app.py. The bootstrap validates and lazy-loads the packaged V28 runtime, serves the customer UI at /, and passes API/assets/preview/runtime traffic to the V28 FastAPI application.
Health endpoints:
/_148ai/health/_148ai/diagnosticThis is a launch candidate, not a claim of completed public production infrastructure. The current Vercel review uses memory persistence unless configured otherwise. Persistent PostgreSQL, Redis workers, object storage, real GPU/open-weight model serving, remote sandbox, live media engines, Stripe live billing, transactional email, production domain/DNS/TLS, store signing and full operational monitoring remain separate live-infrastructure gates.
The repository CI checks:
/api/status runtime API smokeCanonical full reviewed source remains the archived 148AI V28 Launch Candidate release package; this repository is the deployment layer.