All decision model projects

CLI that filters coding-agent subscriptions by quota and policy, then asks TypeSafe to rank the rest and pick a reasoning effort.

Event-driven multi-agent network where Jev watches live state and assigns pending tasks to idle general-purpose agents.

Hackathon agent monitor that scores chains of sandboxed agent actions with Jev and escalates through a fixed L1-L5 containment ladder.

SignalChain: data-cleaning framework where Jev Choices classify CSV columns and dataset scenes, and local scripts do all the writing.

Tournament engine that ranks texts with Swiss pairing, Elo and TrueSkill, using Jev or local models to judge each head-to-head duel.

Educational multi-agent AI hedge fund that paper-trades, with TypeSafe Jev as one supported model provider for its investor agents.

Example where Jev decides whether to act on an AlgoVault crypto-perp verdict now; code applies thresholds and logs act, hold or escalate.

MCP server that drives an Android phone over adb: it reads the screen, asks Jev the next action, acts, and reports if the goal was met.

Platform pairing an LLM with werr, a claimed sub-millisecond fractal decision engine that returns typed noul, choice and score answers.

MCP server that finds relevant code with Jev Choice and Noul checks instead of grep sweeps, and audits diffs against task intent.

Semantic ArkUI linter for coding agents: a fact extractor finds candidate state bugs and a System One model judges each one.

Neo-brutalist Chinese web app styled after Ask Jeeves that gives direct verdicts on everyday dilemmas like whether to buy or what to eat.

Patched Codex CLI where Jev reads task context mid-run and picks the next reasoning effort level and how long it should last.

Codex and Claude Code skills that narrow a repository locally and let Jev judge which candidate files the coding model needs.

Experimental Paseo plugin where Jev, local Laya or OpenAI Decisions classifies each message and routes it to a provider/model preset.

Cost-, cache- and quota-aware LLM router that classifies each turn with a Jev-class model, prices every route, and checks cheap answers.

Experimental Chrome extension that asks Jev closed questions about X posts, X profiles and HN comments and labels them only when clear.

Browser game racing a human against Jev to spot spam over 25 rounds, comparing speed and accuracy, on a Cloudflare Worker.

Turkish chat toy that always replies with one of five fixed phrases, picked by a Jev Choice, with punctuation set by a score and a yes/no.

Pi coding harness where any LLM writes and Jev returns typed reflex scores that policy code turns into allow, confirm, block, warn or steer.

Single-file Python script that answers Bilibili's 100-question hardcore membership quiz by asking Jev to pick each multiple-choice answer.

Experimental lab for semantic business gates: each gate is an independent Jev Noul that code maps to NO, REVIEW or YES by threshold.

Codebase search where an ensemble of walkers descends the file tree, with Jev ranking which files and folders look relevant at each level.

Lets Jev choose concrete browser actions while Codex or Browser Use's engine handles generation, recovery and final verification.

Experimental memory admission lab: Jev weighs a proposed memory against its source and a small policy says save, skip or defer.

Floating live-call coach that transcribes locally and asks Jev after each sentence what to do next, shown with a confidence gauge.

Next.js brick breaker whose paddle can be handed to an AI pilot where Jev chooses where the paddle goes via a Choice question.

CLI and GitHub Action that scores a code change's risk from policy rules plus Jev's semantic signals, and suggests checks and reviewers.

NiceGUI desktop chess app for evaluating Jev's move choices against Stockfish, random players or itself, with PGN export.

Claude Code PreToolUse quality gate where Jev scores drafts (such as SVG illustrations) before they are written and blocks substandard work.