yomo

Why we included this project

For teams wiring up function-calling agents, YoMo packages the whole tool lifecycle into a single serverless runtime, so you can register skills and expose them through a standard chat-completions endpoint instead of building that plumbing yourself. Its geo-distributed setup keeps inference and tool execution close to the end user, which is exactly what you want when response time decides whether an assistant feels sharp or sluggish. Deployment is deliberately light: install the CLI, run the server, and you are up, which means a small team can run a far-flung setup without a dedicated infrastructure crew. If you are prototyping agentic workflows in Go or TypeScript and want to skip boilerplate without giving up a realistic deployment path, this is worth trying.