What is the System One pre-decision POC?

A proof of concept connecting a System One decision model (Jev/Laya) to coding agents as a pre-decision layer, so the work route is chosen before the agent starts burning tokens.

What problem does it address?

A coding agent must decide where to start — explore, implement, test or ask for more information — and normally the big model does that by thinking. The POC tries to move that decision to a fast typed model.

What are the four routes?

inspect (understand the code first), implement (the change is clear), test (verify an existing change) and the fourth route visible in the README's truncated table is asking for more information.

Did the POC actually save money or tokens?

No, per the README's own headline: cost savings did not appear in the measured task, with +13% on Pi and +16% on OpenCode. The integration itself works and is deterministic.

Which agents does it support?

Three environments: Pi, OpenCode and DSH, wired through a DSH plugin and an MCP server, per the repository description and README.

What is in the repository?

The code, the data and the measurement reports, under MIT licence, structured as a Spanish-language write-up with the route table and results. The digest truncates the route table and later sections.

One measured task, the author's own numbers, a Spanish README truncated in the capture and 0 stars. The negative result is itself the point, so the POC should be read as an experiment, not a product.

Jev directory

CommunityJev Demos & Experiments

System One pre-decision POC

A Spanish-language POC that puts a System One decision model (Jev/Laya) in front of three coding agents — Pi, OpenCode and DSH — so the route (inspect, implement, test or ask) is decided before the agent starts. The integration works; the cost savings did not appear.

See highlights Open github.com

Stars: 0
Language: JavaScript
License: MIT
Reading time: 2 min
Published by: Community
Author: joskr2
Last verified: Oct 2026

joskr2/systemone-poc

User repository on GitHub

View on GitHub

POC: System One (Jev/Laya) como capa de decisión previa para agentes de programación — plugin DSH + servidor MCP + informes de medición de coste/tokens

Forks: 0
Last pushed: Updated Sep 2026

Repo stats from the GitHub API, cached Oct 2026.

Tagscommunity github javascript mcp agents

Works forJev for Routing Jev for Decision gating

Used inJev for Software & DevTools

Languages: JavaScript
Platforms: MCP
Capabilities: Open source

Highlights

Decides the route before the agent starts: inspect, implement, test or ask, chosen by a System One model instead of the big model's thinking tokens.
Connects the decision layer to three agent environments — Pi, OpenCode and DSH — with a DSH plugin and an MCP server.
Honest negative result: the integration is deterministic and works, but the measured task showed no cost or token savings.
Ships the code, the data and the measurement reports; MIT licensed.

Watch out

README reviewed 2026-10-17 from the round-eleven capture, which is truncated inside the route table: setup instructions, the exact measurement setup and the size of the measured task are not in the digest, and the cost figures are the author's own.

Questions about this entry

Short answers drawn from the linked project or page.

Reactions & coverage

Posts, threads, and videos about this entry from around the web.

X: Jev in front of the HeyGen MCP

X: Playwright CLI + Jev vs Playwright MCP

X: A local MCP server that cuts computer use from 23s to 0.8s. Cartwright built Cortex, a local MCP server where…

X: A side panel that drives any site's WebMCP tools

From the Jev guides

Original write-ups on what this entry is for.

MCP tool routing: Jev in front of the tool call

Oct 16, 2026

Tool menus can outgrow the context that reads them. The catalog's Jev-based routers and gates narrow the list: benchmarked tool search, per-tool scoring, fail-closed approvals, and WebMCP in the browser.

Jev Agents & MCP#mcp#tool-routing#agents

Read article

Putting Jev in the Coding Loop: MCP Servers, Gates, and Skills

Sep 14, 2026

The fastest-growing corner of the Jev ecosystem is coding agents. A tour of the MCP servers, tool-call gates, context pruners, and skill rankers built for that loop.

Jev Tools & CLIs#agents#mcp#coding-agent

Read article

Per-Tick Decisions in Games

Sep 28, 2026

A game tick is a state plus a legal action list. How Clash Royale, Minecraft, and the earlier Mario, Pokémon, StarCraft, and drone loops ask Jev one step at a time.

Jev Demos & Experiments#gaming#agents#control

Read article

From the Jev AI model community

Posts from builders shipping with Jev right now.

Follow @typesafeai

System One pre-decision POC

Highlights

Questions about this entry

What is the System One pre-decision POC?

What problem does it address?

What are the four routes?

Did the POC actually save money or tokens?

Which agents does it support?

What is in the repository?

What are the limits?

Reactions & coverage

Related terms

More Jev resources like this

From the Jev guides

From the Jev AI model community