Agent Decisions
The decision step inside an agent loop: which action or UI element to pick next, whether to keep or drop context, when to stop, retry or hand back to a larger model. Browser, desktop and mobile control agents are the largest group.
340 public projects · page 12 of 12
Gate that asks Jev whether a timer or event is worth resuming a sleeping agent's LLM, skipping only when confidently not.
BenchmarkprototypeTypeScriptJIT-JEV Context OS: deterministic context compiler plus a Jev System 1 gate that scores AST relevance, test proofs and tool targets.
BenchmarkresearchPythonData-analysis agent where an LLM runs Python analysis and Jev decides whether to continue, switch direction, verify or stop.
ProjectprototypePythonDrives ego-browser pages with Jev: code builds the allowed actions, Jev picks one, and code acts and re-checks.
ProjectAndroid chat copilot: Jev judges intent, needs and risk in visible chats, then ranks drafted replies to fill, never send.
HookAI decision reliability control plane that evaluates autonomous agent actions with Jev before they run.
ProjectDrives a phone with Jev choosing each action while the phone-use runtime executes it on iOS and Android.
ProjectProject premised on offloading cheap routing decisions from an LLM to a decision model while the LLM writes, tests and revises.
ProjectmacOS voice agent where Jev picks each step's operation and target control, escalating to an LLM agent when unsure.
ArticleproductionDeepSeek Harness preset where Jev answers an agent's open end-of-turn questions, steering it on only above confidence and safety thresholds.
Articleprototype