yao
Self-hosted Yao Agents platform for agents, workspaces and task boards across devices, which announces Jev support.
Overview
Yao Agents is a self-hosted platform that puts all your agents and workspaces in one place across desktop, mobile, browser and API. Agents run on your own devices, workspaces are isolated and accumulate documents into a knowledge base, and conversations become tasks on a board that can be connected to apps or shared. It exposes SSE and WebSocket APIs.
The README announces that Jev, described as TypeSafe's System One model returning typed decisions with confidence, is now available in Yao Agents, with a linked blog post. It also announces DeepSeek Harness integration. The README does not describe how Jev is used inside the platform. The README also contains a guide for AI agents with canonical facts and installation rules, intended to reduce inaccurate summaries by automated tools.
What the decision model decides
| Type | Question | Answers |
|---|---|---|
| Other | Typed decisions available to Yao agents (details not in README)(inferred) | — |
How it works
- Install Yao Desktop or the Android app.
- Create workspaces and agents.
- Optionally use Jev for typed decisions inside agents.
- Track tasks on the board.
Strengths
- Self-hosted, multi-device agent platform.
- Jev available as a model option.
Caveats
- Jev integration details are only in an external blog post.
- Android app is beta.
Getting started
Download Yao Desktop from https://yaoagents.com/downloadAt a glance
- Type
- Project
- Status
- beta
- Runtime
- desktop, Android and browser apps
- Has code
- yes
Use cases and integrations
How curated lists describe it
✨ All your agents and workspaces in one place, on every device you own.
wh000wh000/awesome-jev-live
Frequently asked questions
Does Yao support Jev?
Yes, the README announces Jev availability.
How is Jev used?
The README does not say; see the linked blog post.
Is it self-hosted?
Yes.
Related projects
Standalone Node studio for building and testing workflows of typed Jev judgments with code-owned decisions, plus games and simulations.
ProjectbetaJavaScriptBuilds auditable decision trees and runnable decision SOPs from data and a natural-language goal via LLM features and information gain.
ProjectresearchPythonDeepSeek Harness plugin that gives the human a Jev second opinion beside the agent's multiple-choice questions.
ProjectbetaTypeScriptDeepSeek Harness plugin registering a jev_decide tool so the agent can ask Noul, Choice or Score questions and gate on probabilities.
PluginbetaTypeScriptCommand-line tool and SDK that answers typed questions about text with true/false, a label or a number, for scripts and evals.
ProjectbetaRustDeepSeek Harness jev_decide tool that brings the Jev decision paradigm to any OpenAI-compatible LLM, or to real Jev with a key.
ProjectprototypeTypeScript