last-exit

A cyberpunk border encounter powered by TypeSafe Jev. Bluff the guard. Inspect the receipts.

How it uses Jev

The batch made 596 Jev requests across 298 turns. Median local HTTP reply time was 537 ms (p95 664 ms), including two sequential model calls but excluding browser paint and typing. Estimated input-token cost was $0.057 for all 100 crossings , at $0.042/million input tokens. Public hosting adds network and database overhead; these are local evaluation timings, not a hosted speed promise.

Source: community catalog. Details and metrics reflect the published community snapshot and are not a performance endorsement. Documentation details extracted from GitHub · the README · the project website.

Keep exploring

Similar to last-exit

Browse all
cyber-breach-jev project preview
Games & simulations
Cyber-Breach: The Jev Protocol - A tactical cyberpunk arena combat game powered by TypeSafe AI Jev System One decision model
heist-one project preview
Games & simulations
Observable browser stealth game: Jev makes typed guard judgments while deterministic code owns the world.
PlayJev project preview
Games & simulations
Play ten browser games from the screen with a 0.8B open model that returns a probability over the game's legal moves in one forward pass, no text generated.
twitter-jev-guard project preview
Applications
使用 TypeSafe Jev 在 X/Twitter 时间线上识别低质量、垃圾和广告帖子,并在文字区域显示醒目的半透明水印。
jev-guard project preview
Agent tooling
Prompt-injection and dangerous-action guard for coding agents (Claude Code, Codex, pi, ACP), powered by Jev
dsh-jev-prune project preview
Agent tooling
Jev-judged context compaction for DeepSeek Harness: semantic tool-result pruning + deterministic receipt compaction
jev-layer project preview
Agent tooling
Portable System-1 decision layer for agent harnesses with host-owned routing, receipts, replay, and fail-open integrations.
jev-trade-cc project preview
Applications
Jev Can Trade Stocks — a point-in-time O'Neil momentum backtest where TypeSafe's System One model picks the entries and judges the exits
jev-cli (shaharia-lab) project preview
Benchmarks & research
Command-line tool for TypeSafe AI's Jev model. Ask yes/no, multiple-choice and rubric questions about any text and get calibrated probabilities back. Answers become exit codes for shells and CI, JSON for scripts, and MCP tools for AI agents.