Projects tagged with “llm-agents”
Browse Jev GitHub projects tagged with llm-agents. Filter, sort and explore their source repositories.
- 112jev-browserBrowser automation where an LLM plans and Jev (Typesafe System One) decides. Library, CLI and MCP server.
Ying-Kai-Liao/jev-browser · JavaScript
13625javascriptbrowser-automationllm-agentsmcp - 208JevGPTJevGPT: Thinking Fast and Slow in Autonomous Research
sunyasheng/JevGPT · TypeScript
433browsecomp-plusdeep-researchinformation-retrievalllm-agentsneuclirresearch-agent - 644JevOnlyPure Jev that can "type" and drive towards task completion.
buluoray/JevOnly · Python
50browser-agentllm-agentsplaywrightpython - DSH plugin: register TypeSafe Jev (System One decision model) as an agent tool — jev_decide returns calibrated probabilities (noul/choice/score) for routing/triage/guardrail judgments, no text generation. 把 TypeSafe Jev 决策模型注册为 DSH agent 工具
nanami-0713/dsh-jev-decide · JavaScript
42deepseek-harnessdsh-pluginllm-agentsjavascript - Long-term memory for AI agents on Jev. Keep raw records, judge them with a fast System 1 model and answer from under 4k tokens of context.
Grivn/mnemon-memory-agent · TypeScript
40agent-memoryai-agentscontext-engineeringdeepseek-harnessdsh-plugindual-processllm-agents - High-speed dual-process browser agent combining sub-150ms Jev reflex decisions with Stagehand stealth automation and Gemini/Claude System 2 fallback.
adnanahmaddev/system-one-browser-agent · TypeScript
40browser-automationbrowser-useclaudegeminillm-agentsplaywrightstagehand - 930FootworkA verified browser agent: a cheap Jev guard (evidence-checked completions, a destructive gate) in front of any LLM browser driver, with Jev taking the mechanical steps in dual mode. Built on browser-use; every number pre-registered and measured.
Tom-R-Main/Footwork · Python
21browser-agentbrowser-automationbrowser-usedual-processllm-agents - 1223jev-agent-skillDecisões tipadas em milissegundos com o Jev (System One da TypeSafe) via OpenRouter — state + perguntas noul/choice/score entram, decisões calibradas saem. Validação pesada baseada nos conceitos do modelo, bandas auto/hitl/abstain, deteção de prompt injection e servidor MCP. Zero dependências (Node ≥20).
frederico-kluser/jev-agent-skill · Python
10agent-skillsai-agentsdecision-modelllm-agentsmcpnodejsopenrouter - Verbatim context compaction for OpenClaw: a context engine powered by TypeSafe's Jev. Drops stale tool calls and results, never summarizes.
SqaaSSL/openclaw-jev-compaction · TypeScript
10compactioncontext-enginellm-agentsopenclawopenclaw-plugintypescript - 2037jev-compactPrune-not-summarise context compaction for Claude Code, guided by Jev (TypeSafe System One)
HAR5HA-7663/jev-compact · TypeScript
00claude-codecompactioncontext-windowllm-agentstypescript - Screening LLM agent tool output for relevance and prompt injection with Jev, a hosted System One classifier. Harness, test beds, run records and analysis.
zacharytgray/jev-context-filter-harness · Python
00agent-securityagentdojocontext-managementllm-agentsllm-evaluationprompt-injectiontoken-efficiency - 2106jev-dshSystem One (TypeSafe Jev) decision tool for DeepSeek Harness — bounded choice / noul / score judgments as an agent tool + skill.
ZhengSJCode/jev-dsh · JavaScript
00agent-toolsdecision-makingdeepseek-harnessdsh-pluginllm-agents - 2160jev-game-labBloons TD 6 agent: TypeSafe Jev picks each purchase through a MelonLoader bridge, with rules in code and an optional Claude strategist
greenlittleapple/jev-game-lab · JavaScript
00bloons-td-6btd6claude-codegame-aillm-agentsmelonloadertower-defense - Pokemon Red played by Jev, a model that only outputs probabilities. Based on valentynkit/jev-plays-pokemon-red (MIT).
BrickerP/jev-plays-pokemon-red · Python
00emulatorgameboyllm-agentspokemonpyboypython - AI agent that plays Super Mario Bros with TypeSafe Jev - a typed decision model reading NES RAM instead of pixels, verified by emulator rollback search. Clears 1-1 to 1-4 and 2-1 for ~$0.02.
VBS2004/jev-plays-super-mario-bros · Python
00ai-agentsdecision-modelemulatorgame-aillm-agentsnespython - 2410jev-slot-machineJev plays a slot machine until the money runs out. The local version of jevslots.live.
ella0333/jev-slot-machine · JavaScript
00gamblingllm-agentsslot-machinejavascript - Slay the Spire 2 agent: TypeSafe Jev picks each move, a Claude Code session plans the run as strategist
greenlittleapple/jev-spire-strategist · JavaScript
00claude-codegame-aillm-agentsslay-the-spire-2javascript - 2460jev-tool-callingApplying Jev to zero-shot tool calling on the Berkeley Function Calling Leaderboard (BFCL)
will-rice/jev-tool-calling · Python
01benchmarkbfclevaluationfunction-callingllm-agentstoken-classificationtool-calling - 2482jev-un-squadronA TypeSafe Jev decision model flying U.N. Squadron (SNES, 1991) from raw WRAM via BizHawk + Lua + Python
cbroker1/jev-un-squadron · Python
00bizhawkgame-aillm-agentsluareverse-engineeringsnespython - 2509jev-world-modelA frozen decision model (JEV) as an LLM agent's world model, compared with zero-shot and LoRA-fine-tuned generative LLMs in TextWorld. Code, logs, analysis and manuscript for KIIS 2026 Fall (oral). LLM 에이전트의 월드모델: 결정모델과 범용 LLM의 비교
thddydgnl/jev-world-model · Python
00decision-modelllm-agentsloraqwen3reproducible-researchresearch-papertextworld - 2637laya-guardrailsGuardrails de input, tool call y output para agentes de IA con un modelo System One tipo Jev (laya-pt-es-typed) autoalojado en una NVIDIA L4: 5 ms por check frente a 140 ms de un LLM-as-a-judge.
javimp2003/laya-guardrails · Python
00ai-safetyfastapiguardrailslayallm-agentsllm-as-a-judgenvidia-l4 - 2938tracecore-reflexVerifiable hybrid agents that use fast Jev decisions for routine actions and strong LLM fallbacks for uncertainty, with confidence gating, telemetry, replayable traces, and auditable TraceCore compliance bundles.
justindobbs/tracecore-reflex · Python
00agent-evaluationllm-agentsmodel-routingopenroutertracecorepython
1–22 of 22