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Website and Documentation for Inngest
翻译准备中
取自仓库 README
正在加载完整 README。 以下为已保存的摘录。
The Inngest website contains marketing, documentation, and blog content.
Before being able to run the app for the first time, you need to follow the steps below:
• Git
• Node.js 22
• Optional\* - Join the team on Vercel with your GitHub account.
1. Clone this repository
2. Install pnpm with
Corepack by running
corepack enable; corepack prepare
3. Install dependencies by running pnpm install
4. Optional\* - Link local project to its Vercel project by running pnpm vercel link
5. Optional\* - Download environment variables by running pnpm env:pull查看完整 README ↗The Inngest website contains marketing, documentation, and blog content.
Before being able to run the app for the first time, you need to follow the steps below:
pnpm with
Corepack by running
corepack enable; corepack preparepnpm installpnpm vercel linkpnpm env:pull* Running the website is possible with the default environment variables available in .env.development. Fetching other environment variables is only necessary if working on features that require them.
To start the app in development mode, run the following command:
$ pnpm dev
This will start a local server that will automatically rebuild the app and refresh the page when you make changes to the code. The app will be available at http://localhost:3001.
This is how you will run the app most of the time.
To run the app in production mode, run the following commands in order:
# Build the app for production usage
$ pnpm build
# Start the app in production mode
$ pnpm start
This can be useful for testing the app in production mode locally.
Videos embedded in blog posts should be served as HLS (adaptive bitrate) rather than a single large file. HLS starts playback quickly and adapts the quality to the viewer's connection, so a slow network gets a lower-resolution stream instead of a stalled video.
Prerequisite: ffmpeg (which includes ffprobe):
$ brew install ffmpeg
Transcode a source video into an HLS ladder (1080p / 720p / 480p) with:
# Writes ./hls/<name>/{master.m3u8,stream_0,stream_1,stream_2}
$ pnpm video:hls ./durable-agent-in-8-minutes.mov
# Optionally set a custom output directory
$ pnpm video:hls ./demo.mp4 ./dist/demo
Then:
Upload the entire output folder to the CDN, preserving its structure
(e.g. cdn.inngest.com/videos/<name>/). The playlists reference segments by
relative path, so the folder layout must be kept intact.
Reference the master playlist as the video src in your MDX:
<AutoplayVideo src="https://cdn.inngest.com/videos/<name>/master.m3u8" poster="..." />
The AutoplayVideo component detects .m3u8 sources automatically: Safari/iOS
play HLS natively, and other browsers use a lazy-loaded hls.js shim. Non-HLS
sources (.mp4, .mov) continue to work unchanged. The source video and local
hls/ output are build artifacts — don't commit them.
Environment variables are managed with the Vercel CLI. Use the following commands to manage them:
# Download the latest development environment variables
$ pnpm env:pull
# Add a new environment variable
$ pnpm env:add
# Remove an environment variable
$ pnpm env:rm
Check the Vercel documentation for more information.
You should never commit environment variables to the repository. If you need to add a new
environment variable, add it with the pnpm env:add command and then download it with the
pnpm env:pull command.