JevCode /カバレッジに戻る / その他のリスト

その他のリスト

このカテゴリの 425 件。各項目に具体的な判断と開ける根拠リンクがあります。

• 0x7067/claude-jev

  README states: The run sends your past prompts to api.typesafe.ai. Start with --sample 250

  根拠https://github.com/0x7067/claude-jev/blob/main/eval/README.md#L130

  ラベルreadme-claim

• 0xArx/jevegis

  README states: retrieved documents) and Trust & Safety moderation, both powered by TypeSafe's Jev

  根拠https://github.com/0xArx/jevegis/blob/main/src/lib/engine.ts#L1

  ラベルcall-site-verified

• 0xBakeer/arbiter

  README states: Serve typed-decision models — Laya or your own — on your GPU or your Mac, with a Jev-compatible

  根拠https://github.com/0xBakeer/arbiter/blob/main/integrations/claude-code/hooks/eval.py#L57

  ラベルcall-site-verified

• 0xNatoshi/jev-codex-router

  README states: Jev chooses a model and thinking effort together for each model call, including

  根拠https://github.com/0xNatoshi/jev-codex-router/blob/main/poc/backtest_savings.py#L157

  ラベルcall-site-verified

• 0xtrou/rubikjev

  README states: You scramble like a menace. Jev — the in-house AI solver — judges the chaos, roasts you, and cooks the solve live.

  根拠https://github.com/0xtrou/rubikjev/blob/main/src/app/api/solve/route.ts#L127

  ラベルcall-site-verified

• 2FastLabs/agent-squad

  README states: ⚡ Fast and cheap — sub-second decisions billed on input tokens only (output tokens free, see pricing), versus a full LLM call per routing decision.

  根拠https://github.com/2FastLabs/agent-squad/blob/main/README.md#L157

  ラベルreadme-claim

• 3clyp50/a0-typesafe-ai

  README states: Ask Jev focused questions and get typed answers with probability cards in Agent Zero's chat. One typesafe_query call can combine Choice, Noul, Score, and Box questions over the same text and images.

  根拠https://github.com/3clyp50/a0-typesafe-ai/blob/main/extensions/webui/get_tool_message_handler/typesafe.js#L2

  ラベルcall-site-verified

• 6Mikao9/jev-agent-design-with-topk-logits-choices

  README states: Jev is the first backend we use for live experiments; the runtime API is

  根拠https://github.com/6Mikao9/jev-agent-design-with-topk-logits-choices/blob/main/jev_agent/jev_client.py#L57

  ラベルcall-site-verified

• Aayan-DEV/aayans-yc-indexor

  README states: Put a TypeSafe key in .env.local as TYPE_SAFE_KEY. Get one at typesafe.ai.

  根拠https://github.com/Aayan-DEV/aayans-yc-indexor/blob/main/README.md#L30

  ラベルreadme-claim

• AbdelStark/bicameral

  README states: Sent to TypeSafe (POST /v1/systemone via TypeSafeClient.systemOne): redacted, truncated JSON state for the current reflex (goal, proposed tool, compact diffs, recent action summaries) and the pack's questions.

  根拠https://github.com/AbdelStark/bicameral/blob/main/packages/s1-runtime/src/backends/typesafe.ts#L1

  ラベルcall-site-verified

• abe75ch/tiltmeter

  README states: Notices your Jev decisions tilting before anything visibly breaks.

  根拠https://github.com/abe75ch/tiltmeter/blob/main/tiltmeter.py#L270

  ラベルcall-site-verified

• AboveColin/HA-Jev

  README states: back. Jev is a decision model, not a chat model. It answers a typed question with a

  根拠https://github.com/AboveColin/HA-Jev/blob/main/README.md#L12

  ラベルreadme-claim

• ad-si/cai

  README states: [TypeSafe]: https://docs.typesafe.ai/models

  根拠https://github.com/ad-si/cai/blob/main/readme.md#L50

  ラベルreadme-claim

• adarshmishra07/jcm-router

  README states: level per message, using TypeSafe's Jev classifier. Trivial questions go to Haiku, everyday work to

  根拠https://github.com/adarshmishra07/jcm-router/blob/main/scripts/eval.ts#L12

  ラベルcall-site-verified

• adhyaay-karnwal/jev-chat

  README states: A research decoder that treats Jev (a System One decision model) as if it were a language model.

  根拠https://github.com/adhyaay-karnwal/jev-chat/blob/main/src/jevchat/client.py#L7

  ラベルcall-site-verified

• affirmitv/ghosthands

  README states: Configuration: GH_PLANNER=jev, GH_JEV_MODEL (default typesafe/jev-1.13; ~typesafe/jev-latest

  根拠https://github.com/affirmitv/ghosthands/blob/main/README.md#L221

  ラベルreadme-claim

• afshinm/laya-mps

  README states: Run Jev-style typed decisions locally on your Mac with low RAM usage and fast responses.

  根拠https://github.com/afshinm/laya-mps/blob/main/README.md#L3

  ラベルreadme-claim

• Ai-trainee/awesome-jev

  README states: prism-liquidity-agent — ★69 · TypeScript · Jev 评分 18 · 工具与产品 — Solana 流动性代理：可选的 Jev 影子判定用于存款选择、有毒流与市场状态压力。

  根拠https://github.com/Ai-trainee/awesome-jev/blob/main/README.md#L390

  ラベルreadme-claim

• ai4curation/jevotron

  README states: in tabular, structured, and text files, SQLite, and DuckDB, powered by Jev.

  根拠https://github.com/ai4curation/jevotron/blob/main/README.md#L8

  ラベルreadme-claim

• AiPersonacademy/jev-resume-disqualifier

  README states: Automated Candidate Knockout Engine for High-Volume Recruiting — Powered by TypeSafe Jev System One Decision Intelligence.

  根拠https://github.com/AiPersonacademy/jev-resume-disqualifier/blob/main/engine/decision_gateway.py#L20

  ラベルcall-site-verified

• alexj11324/open-jev-approvals

  README states: When the agent is about to run a tool, a hook intercepts the call, TypeSafe

  根拠https://github.com/alexj11324/open-jev-approvals/blob/main/internal/jev/client.go#L22

  ラベルcall-site-verified

• allebee/pytest-jev

  README states: All the claims about one text go to Jev in a single request. When a prompt change breaks the reply,

  根拠https://github.com/allebee/pytest-jev/blob/main/bench/llm_judge.py#L14

  ラベルcall-site-verified

• allenporter/home-assistant-typesafe

  README states: Queries the TypeSafe System One API (jev-latest) in a single network roundtrip, obtaining calibrated probabilities across all questions without hallucination or token streaming latency.

  根拠https://github.com/allenporter/home-assistant-typesafe/blob/main/README.md#L34

  ラベルreadme-claim

• alterhq/typesafe-sdk-swift

  README states: authenticated backend. Keep the TypeSafe API key in server-side secret storage. Direct API-key

  根拠https://github.com/alterhq/typesafe-sdk-swift/blob/main/Sources/TypeSafe/Requests.swift#L4

  ラベルcall-site-verified

• altryne/jevify

  README states: Use Jev during everyday agent work, and find what it can improve in your product.

  根拠https://github.com/altryne/jevify/blob/main/scripts/jev_client.py#L32

  ラベルcall-site-verified

• andrelandgraf/safer-with-jev

  README states: Public showcases of TypeSafe Jev judgments. TypeSafe Jev inspects the body, then optionally forwards the same bytes to a caller-chosen HTTPS URL.

  根拠https://github.com/andrelandgraf/safer-with-jev/blob/main/src/index.ts#L5

  ラベルcall-site-verified

• andududu/jeview

  README states: your code → Jeview (127.0.0.1:4777) → TypeSafe (api.typesafe.ai)

  根拠https://github.com/andududu/jeview/blob/main/README.md#L9

  ラベルreadme-claim

• Anil-matcha/awesome-jev-by-typesafe

  README states: ▶ Watch: Jev: 8 Wild Things People Already Built With TypeSafe's New AI Model

  根拠https://github.com/Anil-matcha/awesome-jev-by-typesafe/blob/main/README.md#L14

  ラベルreadme-claim

• aniruddh-krovvidi/switchboard

  README states: A guardrail and model router for LLM gateways, built on Jev, TypeSafe AI's

  根拠https://github.com/aniruddh-krovvidi/switchboard/blob/main/jev.py#L3

  ラベルcall-site-verified

• anishfn/shapeshift

  README states:

  根拠https://github.com/anishfn/shapeshift/blob/main/src/lib/jev/client.ts#L2

  ラベルcall-site-verified

• AnotiaWang/awesome-jev

  README states: Open-source products and demos that put a decision model in a real loop. Most use Jev today.

  根拠https://github.com/AnotiaWang/awesome-jev/blob/main/README.md#L178

  ラベルreadme-claim

• AntonioCoppe/jev-harness

  README states: It is not affiliated with TypeSafe. You need a TypeSafe API key to call Jev.

  根拠https://github.com/AntonioCoppe/jev-harness/blob/main/recipes/agent-comm-harness/message-route.ts#L1

  ラベルcall-site-verified

• AppitStudio/awesome-jev

  README states: Discover Jev-powered apps, developer resources, and runnable decision examples.

  根拠https://github.com/AppitStudio/awesome-jev/blob/main/README.md#L3

  ラベルreadme-claim

• AppSprout-dev/Unsga3

  README states: TypeSafe / Jev Pareto scoring (optional)

  根拠https://github.com/AppSprout-dev/Unsga3/blob/main/README.md#L104

  ラベルreadme-claim

• ARCJ137442/jev-life

  README states: Alongside it sits a Jev-compatible broker that wraps any OpenAI- or Anthropic-compatible LLM into a Jev backend,

  根拠https://github.com/ARCJ137442/jev-life/blob/main/src/client/api.ts#L222

  ラベルcall-site-verified

• arczhi/jet

  README states: Judgment-first routing — Jev (TypeSafe System One) handles context

  根拠https://github.com/arczhi/jet/blob/main/README.md#L43

  ラベルreadme-claim

• atharvamhaske/typesafe-sdk-go

  README states: | WithBaseURL | TYPESAFE_BASE_URL | API base URL, defaults to https://api.typesafe.ai |

  根拠https://github.com/atharvamhaske/typesafe-sdk-go/blob/main/README.md#L115

  ラベルreadme-claim

• audreyt/paseo-spacedock

  README states: Gate judgment (Jev / TypeSafe-compatible)

  根拠https://github.com/audreyt/paseo-spacedock/blob/main/README.md#L56

  ラベルreadme-claim

• automateyournetwork/netclaw

  README states: | JEV_BASE_URL | https://api.typesafe.ai; /v1/systemone appended when no path supplied; full custom paths accepted |

  根拠https://github.com/automateyournetwork/netclaw/blob/main/mcp-servers/jev-mcp/README.md#L42

  ラベルreadme-claim

• Avinash-jetwani/jevmem

  README states: To TypeSafe AI, https://api.typesafe.ai/v1/systemone, or the URL in TYPESAFE_BASE_URL if you set it:

  根拠https://github.com/Avinash-jetwani/jevmem/blob/main/plugin/README.md#L46

  ラベルreadme-claim

• awlevin/typesafe-computer-use

  README states:

  根拠https://github.com/awlevin/typesafe-computer-use/blob/main/typesafe_computer_use/actions.py#L9

  ラベルcall-site-verified

• awslabs/agentcore-samples

  README states: Jev Evaluators (--with-jev)

  根拠https://github.com/awslabs/agentcore-samples/blob/main/01-features/06-observe-evaluate-optimize-your-agent/02-evaluate/custom-code-based-evaluation/README.md#L600

  ラベルreadme-claim

• Ayush0054/metis

  README states: metis-workflow is a small Python framework for workflows powered by TypeSafe AI's

  根拠https://github.com/Ayush0054/metis/blob/main/legacy/metis-triage/src/metis_triage/_triage.py#L154

  ラベルcall-site-verified

• Baba88611/detroit-ai-player

  README states: Select the dedicated TypeSafe backend with --model jev. Append these variables to your local 03_runner/.env, preserving existing configuration. Keep the real key local:

  根拠https://github.com/Baba88611/detroit-ai-player/blob/main/README.md#L189

  ラベルreadme-claim

• bcharleson/jev-gtm-cookbook

  README states: Open-source go-to-market recipes built on TypeSafe's Jev: code does the work, Jev makes the judgment calls, and each recipe runs locally on data you already own.*

  根拠https://github.com/bcharleson/jev-gtm-cookbook/blob/main/README.md#L5

  ラベルreadme-claim

• beamnxw/minelog

  README states: Jev (TypeSafe System One) picks one bounded action from a closed catalog, plus a safety noul and an urgency score. Fast. Typed. No prose.

  根拠https://github.com/beamnxw/minelog/blob/main/README.md#L33

  ラベルreadme-claim

• BeatAPI/awesome-jev

  README states:

  根拠https://github.com/BeatAPI/awesome-jev/blob/main/scripts/curate-star50-expansion-2026-09-23.mjs#L217

  ラベルcall-site-verified

• BenCaunt/jev-box-tracking

  README states: jev-box-tracking

  根拠https://github.com/BenCaunt/jev-box-tracking/blob/main/README.md#L1

  ラベルreadme-claim

• bestagentkits/jev-skillful

  README states: The prompt is sent to api.typesafe.ai as part of the routing request. This is the only kind of

  根拠https://github.com/bestagentkits/jev-skillful/blob/main/packages/cli/src/core/eval/replay.ts#L22

  ラベルcall-site-verified

• Bewinxed/jevgpt

  README states: TypeSafe AI's jev-latest is a decision model: it does not generate text,

  根拠https://github.com/Bewinxed/jevgpt/blob/master/src/jevgpt/sampler.py#L29

  ラベルcall-site-verified

• bhaiG-de/jev-design-test

  README states: Prerequisites: Node 20+, and a Jev (TypeSafe) API key.

  根拠https://github.com/bhaiG-de/jev-design-test/blob/master/vite.config.ts#L5

  ラベルcall-site-verified

• BillionsBobby/JevRouter

  README states:

  根拠https://github.com/BillionsBobby/JevRouter/blob/main/src/provider.ts#L103

  ラベルcall-site-verified

• bladedevoff/stuntd

  README states: With the paid Jev API in front, the same demo runs with it as the teacher instead: put

  根拠https://github.com/bladedevoff/stuntd/blob/main/examples/banking/README.md#L42

  ラベルreadme-claim

• bodepudimuneendra-netizen/laya-jev-GraphRAG

  README states: │ LAYER 2 — DECISION (Laya / Jev) ← The CPU │

  根拠https://github.com/bodepudimuneendra-netizen/laya-jev-GraphRAG/blob/main/graphrag_neo4j_laya/graphrag/models/jev.py#L7

  ラベルcall-site-verified

• bohutang/sift

  README states: Search results stay native: no Jev scoring, local filters, labels, or collapsed posts on /search. The normal feed and post pages still use your filtering settings.

  根拠https://github.com/bohutang/sift/blob/main/background.js#L4

  ラベルcall-site-verified

• brainstormity/Jev-X-Sentiment-Analysis

  README states: An on-demand crypto market intelligence and decision-support terminal powered by TypeSafe AI's System One model (Jev).

  根拠https://github.com/brainstormity/Jev-X-Sentiment-Analysis/blob/main/app/services/typesafe_service.py#L3

  ラベルcall-site-verified

• brandonbryant12/transcript-scorecard

  README states: Server-side OpenAI Decisions and TypeSafe JEV adapters for incremental transcript scoring

  根拠https://github.com/brandonbryant12/transcript-scorecard/blob/main/apps/api/src/classifier.ts#L1

  ラベルcall-site-verified

• brnyxx/jev-ra

  README states:

  根拠https://github.com/brnyxx/jev-ra/blob/main/jev_ra/config.py#L11

  ラベルcall-site-verified

• buberlo/jev-trader

  README states: jev-trader

  根拠https://github.com/buberlo/jev-trader/blob/main/README.md#L1

  ラベルreadme-claim

• bugkiwi/elons-job

  README states: TypeSafe 请求使用 https://api.typesafe.ai/v1/systemone；API 返回不符合预期、超时、401、429 或 5xx 时 Fail Open。

  根拠https://github.com/bugkiwi/elons-job/blob/main/README.md#L46

  ラベルreadme-claim

• BunsDev/typesafe-ai-playground

  README states: Small experiments. Clear decisions. This is an independent community project, not an official TypeSafe AI product, production agent harness, or claim that a typed answer is correct.

  根拠https://github.com/BunsDev/typesafe-ai-playground/blob/main/lib/serverJev.ts#L52

  ラベルcall-site-verified

• BunsDev/typesafe-ui

  README states: Reusable React components and interactive interface patterns for TypeSafe AI projects, built with shadcn/ui, Base UI, and Tailwind CSS.

  根拠https://github.com/BunsDev/typesafe-ui/blob/main/README.md#L3

  ラベルreadme-claim

• burnigtm/jev-mcp

  README states: codex mcp add jev --env TYPESAFE_API_KEY=ts_... -- node /absolute/path/to/jev-mcp/dist/index.js

  根拠https://github.com/burnigtm/jev-mcp/blob/main/src/errors.ts#L1

  ラベルcall-site-verified

• BYK/jev-mcp

  README states: Most Jev integrations expose "ask the model a question". The hard part in practice is not asking —

  根拠https://github.com/BYK/jev-mcp/blob/main/src/tools/ask.ts#L37

  ラベルcall-site-verified

• bytelabs-oss/clash-jev

  README states: play around with it. I built clash-jev, a bot that uses Jev to make near real-time decisions on live

  根拠https://github.com/bytelabs-oss/clash-jev/blob/main/clash_jev/policy.py#L183

  ラベルcall-site-verified

• caiovicentino/eikos-arena

  README states: MIT) against Jev (TypeSafe's decision API). Every 5 minutes both get the same market snapshot and the same 28

  根拠https://github.com/caiovicentino/eikos-arena/blob/main/arena/arena.py#L31

  ラベルcall-site-verified

• can1357/jegrep

  README states: src/jev.rs 0.97 · 458 lines shown

  根拠https://github.com/can1357/jegrep/blob/main/README.md#L122

  ラベルreadme-claim

• CaptainCore/captaincore

  README states: Optional integrations -- fathom_api_key (analytics) and typesafe_api_key (the TypeSafe decision API behind captaincore typesafe and scan --triage; TYPESAFE_API_KEY in the environment works too)

  根拠https://github.com/CaptainCore/captaincore/blob/master/typesafe/client.go#L188

  ラベルcall-site-verified

• carldaws/hunch

  README states: Point Hunch at a System One endpoint. Jev is available through

  根拠https://github.com/carldaws/hunch/blob/main/README.md#L27

  ラベルreadme-claim

• carlosedm10/agi-jev-containment

  README states: What is Jev in this project? Jev is TypeSafe System One (jev-1.13.0): a semantic classifier that answers typed questions (criticality L0–L5, intent, concealment, …) over short-term tape + long-term key nodes. It does not

  根拠https://github.com/carlosedm10/agi-jev-containment/blob/main/README.md#L70

  ラベルreadme-claim

• cbruyndoncx/AskJev-MCP

  README states: Ask everything you want to know about a state in one call rather than several — TypeSafe bills per

  根拠https://github.com/cbruyndoncx/AskJev-MCP/blob/main/src/index.ts#L6

  ラベルcall-site-verified

• cequence-io/openai-scala-client

  README states: Decision models give typed answers with calibrated probabilities. Two APIs serve them - TypeSafe AI's System One

  根拠https://github.com/cequence-io/openai-scala-client/blob/master/typesafe-client/src/main/scala/io/cequence/openaiscala/typesafe/JsonFormats.scala#L125

  ラベルcall-site-verified

• Chandler-Sun/chat2jev

  README states: Convert OpenAI-compatible Chat Completions requests into TypeSafe System One (Jev) State / Questions, compare generated text with structured judgments, and publish reusable question sets as proxy routes.

  根拠https://github.com/Chandler-Sun/chat2jev/blob/main/src/components/answers.tsx#L49

  ラベルcall-site-verified

• CharlesFeng0314/JEV_sees

  README states: JEV is fast, structured, and built to make decisions.

  根拠https://github.com/CharlesFeng0314/JEV_sees/blob/main/jev_sees/__init__.py#L3

  ラベルcall-site-verified

• Charlyhno-eng/jev-document-classification

  README states: To use JEV, you need a Vercel account and an AI Gateway API key.

  根拠https://github.com/Charlyhno-eng/jev-document-classification/blob/main/README.md#L25

  ラベルreadme-claim

• cheeaun/jevmoji

  README states: Type up to 40 characters. Get related emojis scored by TypeSafe’s Jev — many matches, not just one.

  根拠https://github.com/cheeaun/jevmoji/blob/main/README.md#L10

  ラベルreadme-claim

• CheshiAI/Cheshi

  README states: Previous conversation recall is off by default, including for existing TypeSafe

  根拠https://github.com/CheshiAI/Cheshi/blob/main/README.md#L86

  ラベルreadme-claim

• chez-shanpu/typesafeai-go

  README states: // A TypeSafe AI compatible server running locally without authentication.

  根拠https://github.com/chez-shanpu/typesafeai-go/blob/main/option.go#L9

  ラベルcall-site-verified

• christos-antono/NPCTalk

  README states: | Actions and spoken orders | TypeSafe | TYPESAFE_API_KEY |

  根拠https://github.com/christos-antono/NPCTalk/blob/main/README.md#L57

  ラベルreadme-claim

• chy4pro/jev-for-chrome

  README states: | OpenRouter | https://openrouter.ai/api/alpha/decisions | typesafe/jev-1.13 |

  根拠https://github.com/chy4pro/jev-for-chrome/blob/main/README.md#L112

  ラベルreadme-claim

• cobanov/awesome-jev

  README states: New: Ollaya runs open decision models locally behind a Jev-compatible API.

  根拠https://github.com/cobanov/awesome-jev/blob/main/README.md#L8

  ラベルreadme-claim

• cobusgreyling/Jev

  README states: Jev Showcase — TypeSafe System One, not chat

  根拠https://github.com/cobusgreyling/Jev/blob/main/README.md#L1

  ラベルreadme-claim

• cocktailpeanut/jevthoven

  README states: There is no offline composer, audio model, or hidden song library. Every note traces to a persisted Jev decision (or a labeled fixture/manual edit).

  根拠https://github.com/cocktailpeanut/jevthoven/blob/main/app/scripts/live-smoke.mjs#L15

  ラベルcall-site-verified

• codaaiteam/jev-mcp

  README states: Official TypeSafe: get a key at console.typesafe.ai. It's used against api.typesafe.ai by default.

  根拠https://github.com/codaaiteam/jev-mcp/blob/main/README.md#L30

  ラベルreadme-claim

• Cognipeer/agent-sdk

  README states: Against the real Jev API

  根拠https://github.com/Cognipeer/agent-sdk/blob/main/examples/jev-guardrail/README.md#L24

  ラベルreadme-claim

• colliber/duckdb-jev

  README states: the extension talking to api.typesafe.ai.

  根拠https://github.com/colliber/duckdb-jev/blob/main/docs/README.md#L4

  ラベルreadme-claim

• columnar-tech/jevaro

  README states: A Python batching proxy for Jev that returns

  根拠https://github.com/columnar-tech/jevaro/blob/main/README.md#L3

  ラベルreadme-claim

• compozy/yoshi

  README states: Jev judges new context once, above a configurable size gate. Yoshi applies the

  根拠https://github.com/compozy/yoshi/blob/main/README.md#L7

  ラベルreadme-claim

• Contrastive-LM/CLM

  README states: It performs on par with Jev across computer-use, gaming and tool-calling

  根拠https://github.com/Contrastive-LM/CLM/blob/main/src/clm/__init__.py#L6

  ラベルcall-site-verified

• CorieW/JevTest

  README states: To run the same suite using Jev, copy .env.example to .env.local, set TYPESAFE_API_KEY, then run:

  根拠https://github.com/CorieW/JevTest/blob/main/src/jev.ts#L122

  ラベルcall-site-verified

• CTNicholas/jev-workflow-builder

  README states: Create an account on typesafe.ai

  根拠https://github.com/CTNicholas/jev-workflow-builder/blob/main/README.md#L14

  ラベルreadme-claim

• danielgshea/jev-as-a-judge

  README states: This project tests a third option: Jev, TypeSafe AI's System One decision model. We compare Jev with GPT-5.6 Luna, GPT-5.6 Terra, and Claude Sonnet 4.6 on the same fixed agent runs, measuring binary accuracy, score relia

  根拠https://github.com/danielgshea/jev-as-a-judge/blob/main/README.md#L5

  ラベルreadme-claim

• DanRWilloughby/snifftest

  README states: judgment rules too, put a TypeSafe key in TYPESAFE_API_KEY and drop

  根拠https://github.com/DanRWilloughby/snifftest/blob/main/src/jev.ts#L50

  ラベルcall-site-verified

• danvega/jev-spring-boot-starter

  README states: | jev.api-key | TYPESAFE_API_KEY fallback | TypeSafe API key |

  根拠https://github.com/danvega/jev-spring-boot-starter/blob/main/README.md#L145

  ラベルreadme-claim

• Das-rebel/a3m-router

  README states: model="jev-auto" — ML decision head, calibrated probabilities, ~2ms warm

  根拠https://github.com/Das-rebel/a3m-router/blob/main/README.md#L120

  ラベルreadme-claim

• daseinlabs/open-jev

  README states: make systemone # TypeSafe-style request against /v1/systemone

  根拠https://github.com/daseinlabs/open-jev/blob/main/demo/doom/client.py#L66

  ラベルcall-site-verified

• davila7/jev-explained

  README states: Learn how TypeSafe's Jev makes typed, probabilistic decisions — by running it.

  根拠https://github.com/davila7/jev-explained/blob/main/src/lib/providers.ts#L21

  ラベルcall-site-verified

• DECRUX9812/typesafe-skill-router

  README states: A TypeSafe API key (TYPESAFE_API_KEY) — create one at

  根拠https://github.com/DECRUX9812/typesafe-skill-router/blob/main/__init__.py#L30

  ラベルcall-site-verified

• delexw/claude-code-trace

  README states: Jev and provider API keys live in the OS credential store (macOS Keychain, Windows Credential

  根拠https://github.com/delexw/claude-code-trace/blob/main/src-tauri/src/efficiency/jev.rs#L12

  ラベルcall-site-verified

• dengyie/ai-register-machine

  README states: | typesafe.ai / jev | ./register.sh typesafe [count] [threads] | Python + requests + Stytch Magic-Link | typesafe-*.json (0600 API Key) |

  根拠https://github.com/dengyie/ai-register-machine/blob/main/README.md#L36

  ラベルreadme-claim

• devagrawal09/jev-review

  README states: The reviewer keeps orchestration in code and uses Jev for bounded judgments:

  根拠https://github.com/devagrawal09/jev-review/blob/main/src/review/codebase-judgments.ts#L3

  ラベルcall-site-verified

• devagrawal09/stanley-code

  README states: For built-ins, Jev answers fixed-choice questions about each piece. Using your required TypeSafe API key,

  根拠https://github.com/devagrawal09/stanley-code/blob/main/README.md#L94

  ラベルreadme-claim

• Devin-AXIS/jev-dsh-decision

  README states: 问题类型是 choice（单选）、score（等级评分）、noul（是的概率）。多个互补 Skill 可使用多个独立 noul 问题。传入最小必要上下文，不传密钥或无关私密资料。网络固定发往 https://api.typesafe.ai/v1/systemone；默认模型 jev-latest。

  根拠https://github.com/Devin-AXIS/jev-dsh-decision/blob/main/README.md#L137

  ラベルreadme-claim

• DevMortimer/pi-warden

  README states: Jev requests cost money, so each 1.0 change was measured by what it saves. Each figure is the measured change on recorded sessions of one machine, not a promise for yours.

  根拠https://github.com/DevMortimer/pi-warden/blob/main/scripts/slop-cases.mjs#L77

  ラベルcall-site-verified

• devtooligan/jevscan-evm

  README states: Produce a heat map of likely bugs - in seconds, for pennies - built with TypeSafe's Jev.

  根拠https://github.com/devtooligan/jevscan-evm/blob/main/README.md#L10

  ラベルreadme-claim

• dfinke/Jev

  README states:

  根拠https://github.com/dfinke/Jev/blob/main/README.md#L2

  ラベルreadme-claim

• Dicklesworthstone/skillranker

  README states: The right skill for the next step, powered by Jev-compatible providers.

  根拠https://github.com/Dicklesworthstone/skillranker/blob/main/src/jev/client/tests.rs#L100

  ラベルcall-site-verified

• Dimweaker/jev-libero

  README states:

  根拠https://github.com/Dimweaker/jev-libero/blob/main/src/jev_libero/client.py#L14

  ラベルcall-site-verified

• docxology/daf-jev

  README states: Modular, composable Python client and decision toolkit for the TypeSafe Jev

  根拠https://github.com/docxology/daf-jev/blob/main/src/daf_jev/__init__.py#L23

  ラベルcall-site-verified

• doeixd/discern

  README states: error. Provider-neutral: runs on TypeSafe's Jev, or any other DecisionModel.

  根拠https://github.com/doeixd/discern/blob/main/README.md#L7

  ラベルreadme-claim

• dominicrico/jev-router

  README states: | jev (default) | https://api.typesafe.ai/v1/systemone | jev-latest | TYPESAFE_API_KEY or /jev key (required) |

  根拠https://github.com/dominicrico/jev-router/blob/main/README.md#L94

  ラベルreadme-claim

• donvito/ai-backends

  README states: | TypeSafe Jev | System One decision model: choice, score, and noul questions | Available |

  根拠https://github.com/donvito/ai-backends/blob/main/README.md#L99

  ラベルreadme-claim

• dorkitude/webctl

  README states: Smart web search CLI for agents, backed by Jev. Saves a lot of tokens.

  根拠https://github.com/dorkitude/webctl/blob/main/README.md#L3

  ラベルreadme-claim

• Dujaydis/JevSysUno

  README states: A reference integration for TypeSafe AI's Jev, the first "System One" model: typed

  根拠https://github.com/Dujaydis/JevSysUno/blob/master/README.md#L3

  ラベルreadme-claim

• ellistev/typesafe-minecraft-demo

  README states: An experimental Minecraft Java bot controlled through TypeSafe choices, with a live world view and the actual API input and output beside it.

  根拠https://github.com/ellistev/typesafe-minecraft-demo/blob/main/src/decisions.cjs#L45

  ラベルcall-site-verified

• equationalapplications/system-one-reviewer

  README states: | jev | --provider jev (default; SOR_PROVIDER env also works) | hosted TypeSafe System One, TYPESAFE_API_KEY | live-tested (all published numbers) |

  根拠https://github.com/equationalapplications/system-one-reviewer/blob/main/README.md#L236

  ラベルreadme-claim

• eriestra/browser-use-olympics

  README states: echo 'TYPESAFE_API_KEY=...' > ~/.config/typesafe/env # chmod 600

  根拠https://github.com/eriestra/browser-use-olympics/blob/main/almond-fastloop.mjs#L102

  ラベルcall-site-verified

• Eriskii/ErisLint

  README states: anything useful, then map Jev's answers to warnings or errors.

  根拠https://github.com/Eriskii/ErisLint/blob/main/src/jev.rs#L24

  ラベルcall-site-verified

• EthanAlgoX/AIStock

  README states: Try JEV decisions (optional): configure its separate TypeSafe API key in Settings → Models & runtime, then select JEV · Decisions only for a new trading strategy. JEV returns buy/sell/hold, probabilities, and confidence

  根拠https://github.com/EthanAlgoX/AIStock/blob/main/README.md#L148

  ラベルreadme-claim

• etweisberg/jev-ui

  README states:

  根拠https://github.com/etweisberg/jev-ui/blob/main/apps/demo/scripts/check-key.mjs#L5

  ラベルcall-site-verified

• EugeneBoondock/jevsql

  README states:

  根拠https://github.com/EugeneBoondock/jevsql/blob/main/src/client.mjs#L28

  ラベルcall-site-verified

• evoke-build/evoke

  README states: To decide against Jev, export TYPESAFE_API_KEY for the jev adapter, or OPENJEV_API_KEY for openjev. To

  根拠https://github.com/evoke-build/evoke/blob/main/crates/evoke-adapters/src/systemone.rs#L85

  ラベルcall-site-verified

• f-tiger/jev-decision-gate

  README states: Use Jev for Issue triage without building the whole integration and review workflow.

  根拠https://github.com/f-tiger/jev-decision-gate/blob/main/skills/jev-decision-gate/scripts/jev_decision_gate/triage.py#L15

  ラベルcall-site-verified

• f/jev-leftpad

  README states: Set TYPESAFE_API_KEY before using it. jev-leftpad uses jev-latest through

  根拠https://github.com/f/jev-leftpad/blob/main/src/index.js#L1

  ラベルcall-site-verified

• fabricioctelles/skills

  README states: references/jev-integration.md — integration patterns and code examples

  根拠https://github.com/fabricioctelles/skills/blob/main/README.md#L49

  ラベルreadme-claim

• FBddcz/embodied-jev

  README states: 🧠 模型入口放在界面里：可接 TypeSafe Jev、Claude 原生 API、OpenAI 兼容 API，以及本地 MiniCPM5-2B。

  根拠https://github.com/FBddcz/embodied-jev/blob/main/README.md#L42

  ラベルreadme-claim

• featherless-ai/simple-jev

  README states: curl https://simple-jev-demo-api.featherless.ai/v1/models

  根拠https://github.com/featherless-ai/simple-jev/blob/main/README.md#L20

  ラベルreadme-claim

• feder-cr/jev

  README states:

  根拠https://github.com/feder-cr/jev/blob/main/src/main.cpp#L166

  ラベルcall-site-verified

• FeiLiuEM/tetrajev

  README states: Four readings. One routed decision. Two frozen open-weight models, four independent readings per task, one fit-free decision layer — Jev-class decision behaviour with no fine-tuning and no distillation.

  根拠https://github.com/FeiLiuEM/tetrajev/blob/main/scripts/run_jev_api.py#L5

  ラベルcall-site-verified

• FerryCorleone/crush-monitor

  README states: 一个分析你和 Crush 或对象聊天的小工具，支持 Jev、OpenAI Decisions（测试版）和常见文本大模型。帮你读懂一点对方的情绪和想法，也看看自己的回复哪里没表达好、可以怎么调整。

  根拠https://github.com/FerryCorleone/crush-monitor/blob/main/server/analysis.ts#L7

  ラベルcall-site-verified

• flitsinc/go-llms

  README states: TypeSafe System One models (Jev): typed judgments with probabilities, driven by a JSON output schema

  根拠https://github.com/flitsinc/go-llms/blob/main/README.md#L9

  ラベルreadme-claim

• FluidInference/FluidUse

  README states: Jev-style decision model: a 322M mmBERT encoder plus decision head that answers typed

  根拠https://github.com/FluidInference/FluidUse/blob/main/Sources/FluidUse/PublishedCoreML/PublishedCoreMLTypes.swift#L80

  ラベルcall-site-verified

• Foadsf/jev-for-engineers

  README states: so twelve questions in one call cost about what one call costs. TypeSafe

  根拠https://github.com/Foadsf/jev-for-engineers/blob/main/jev.py#L11

  ラベルcall-site-verified

• FrancoisChastel/jev-code

  README states: jev-code

  根拠https://github.com/FrancoisChastel/jev-code/blob/main/README.md#L1

  ラベルreadme-claim

• freepik-company/jev-mcp

  README states: TypeSafe's System One model, through OpenRouter or TypeSafe directly.

  根拠https://github.com/freepik-company/jev-mcp/blob/main/internal/judgment/tasks.go#L35

  ラベルcall-site-verified

• frostney/clean-code-review

  README states: Point it at a public GitHub pull request, a diff or a file. Jev, TypeSafe's

  根拠https://github.com/frostney/clean-code-review/blob/main/README.md#L7

  ラベルreadme-claim

• gamesonrblx/Jevbridge

  README states: ACP and MCP adapter that bridges TypeSafe Jev with any LLM.

  根拠https://github.com/gamesonrblx/Jevbridge/blob/main/src/types.ts#L111

  ラベルcall-site-verified

• gbh1234/Adaptive-rag

  README states: INTENT_BACKEND=jev

  根拠https://github.com/gbh1234/Adaptive-rag/blob/main/README.md#L83

  ラベルreadme-claim

• genspark-ai/genoffice

  README states: Six apps, one AI panel, a file search reranked by TypeSafe Jev, an Ask AI

  根拠https://github.com/genspark-ai/genoffice/blob/main/apps/shell/src/main/file-index/decider.ts#L52

  ラベルcall-site-verified

• get-convex/convex-evals

  README states: bun run decisions dry-run --provider openrouter --model typesafe/jev-1.13 --limit-evals 2

  根拠https://github.com/get-convex/convex-evals/blob/main/README.md#L44

  ラベルreadme-claim

• GhalebDweikat/winnow

  README states: winnow replay score --judged ~/.winnow/replay/judged-typesafe.jsonl

  根拠https://github.com/GhalebDweikat/winnow/blob/main/sidecar/src/winnow/hooks.py#L282

  ラベルcall-site-verified

• GhrezaKh74/JevTicktRouter

  README states: Structured support-ticket triage built on .NET 10 and React 19, using TypeSafe Jev to make

  根拠https://github.com/GhrezaKh74/JevTicktRouter/blob/master/backend/JevTicketRouter.Infrastructure/Decisions/SelfHosted/SelfHostedOptions.cs#L42

  ラベルcall-site-verified

• gi-dellav/hsom

  README states: hsom* is a strict-allowlist harness. environment.toml defines which shell commands Jev (the TypeSafe System One classifier) may select. In one step, Jev picks: command → subcommand → flags (boolean or valued) → path slot

  根拠https://github.com/gi-dellav/hsom/blob/main/README.md#L5

  ラベルreadme-claim

• githubnext/localjev

  README states: A local, Jev-compatible POST /v1/systemone API written in TypeScript for

  根拠https://github.com/githubnext/localjev/blob/main/src/server.ts#L19

  ラベルcall-site-verified

• Godefroy/micdrop

  README states: @micdrop/typesafe - Jev, the System One model of TypeSafe, classifying each turn of the user into typed decisions in a few hundred ms

  根拠https://github.com/Godefroy/micdrop/blob/main/README.md#L27

  ラベルreadme-claim

• GodsBoy/jev-agent-skill-router

  README states:

  根拠https://github.com/GodsBoy/jev-agent-skill-router/blob/main/src/jev_router/transport.py#L12

  ラベルcall-site-verified

• gopaljigaur/decide

  README states: them through one API: TypeSafe's hosted Jev, OpenRouter, OpenAI's Decisions

  根拠https://github.com/gopaljigaur/decide/blob/main/decide/backends/typesafe.py#L20

  ラベルcall-site-verified

• Govcraft/emergent

  README states: Judgment as a Pipeline Step: jev-handler

  根拠https://github.com/Govcraft/emergent/blob/main/README.md#L161

  ラベルreadme-claim

• gtaras7/typesafe-jev

  README states: A local screening workbench for a folder of CVs, built on TypeSafe's Jev model. Point it at one CV or

  根拠https://github.com/gtaras7/typesafe-jev/blob/main/cv-screen/src/cli.ts#L97

  ラベルcall-site-verified

• GuSheng107/helpme-Jev

  README states: JEV decides. LLM explains. You decide.

  根拠https://github.com/GuSheng107/helpme-Jev/blob/main/README.md#L55

  ラベルreadme-claim

• hamgua/alpha-trading-bot-okx

  README states: Jev 快车道（TypeSafe System One）

  根拠https://github.com/hamgua/alpha-trading-bot-okx/blob/main/README.md#L361

  ラベルreadme-claim

• harlan-zw/harlan-agent-kit

  README states: api.typesafe.ai or the Vercel AI Gateway.

  根拠https://github.com/harlan-zw/harlan-agent-kit/blob/main/packages/advocaat/README.md#L9

  ラベルreadme-claim

• haseeb-heaven/jev-system-one

  README states: A polished terminal interface where OpenAI generates useful answers and TypeSafe Jev independently evaluates their relevance, reliability, and quality.

  根拠https://github.com/haseeb-heaven/jev-system-one/blob/master/src/jev_system_one/jev.py#L3

  ラベルcall-site-verified

• hatt-io/jevkeep

  README states: Before compaction, JevKeep reads the conversation and asks Jev, a relevance-scoring model accessed through the Typesafe API, which excerpts still matter to the current task.

  根拠https://github.com/hatt-io/jevkeep/blob/main/README.md#L17

  ラベルreadme-claim

• hellogumbo/should-ai-kill-us-all

  README states: We ask Jev, TypeSafe AI's System One model, whether AI should kill us all. Every ten minutes. Using the actual headlines.

  根拠https://github.com/hellogumbo/should-ai-kill-us-all/blob/main/functions/api/verdict.js#L10

  ラベルcall-site-verified

• hellozenstrategist-lab/eutrya

  README states: A standalone terminal agent with Jev in its decision loop.

  根拠https://github.com/hellozenstrategist-lab/eutrya/blob/main/extensions/eutrya-adaptive-extension/integration/native-v02.mjs#L9

  ラベルcall-site-verified

• hemanth/is-lesion

  README states: jev — TypeSafe System One (jev-1.14.0) with 1 Hugging Face visual reference exemplar per class (noul + choice + score + box)

  根拠https://github.com/hemanth/is-lesion/blob/main/README.md#L48

  ラベルreadme-claim

• hev/reranker

  README states: nDCG@10, every system permuting the same BM25 top-30 shortlist per query. Jev uses the

  根拠https://github.com/hev/reranker/blob/main/hev_rerank/rerank.py#L18

  ラベルcall-site-verified

• heyjunpenn/awesome-jev

  README states: | openai-scala-client | ★ 249 | Scala | 2026-09-23 | Scala client module for TypeSafe System One that submits Jev Noul, Choice, and Score questions and returns typed answers. |

  根拠https://github.com/heyjunpenn/awesome-jev/blob/main/README.md#L80

  ラベルreadme-claim

• HorusJiang/dsh-jev-tools

  README states: 一个 DeepSeek Harness 插件：把 Jev 接进长会话，在工具结果进入上下文之前做判定。

  根拠https://github.com/HorusJiang/dsh-jev-tools/blob/main/README.md#L18

  ラベルreadme-claim

• hotchpotch/jev-reranker

  README states: | model | Explicit value → JEV_MODEL (environment/.env) → jev-latest |

  根拠https://github.com/hotchpotch/jev-reranker/blob/main/README.md#L434

  ラベルreadme-claim

• hraness/algal

  README states: or reuse a TypeSafe key. doctor --clef checks configuration without an API

  根拠https://github.com/hraness/algal/blob/main/README.md#L895

  ラベルreadme-claim

• hunkim/solar-mini4-jev

  README states: A drop-in wrapper that exposes Upstage Solar Pro4 (default; solar-mini4 optional) through the TypeSafe Jev System One API shape.

  根拠https://github.com/hunkim/solar-mini4-jev/blob/main/engine.py#L158

  ラベルcall-site-verified

• huntedman/JevLint

  README states: JevLint is powered by Jev, TypeSafe AI’s System One model. It sends selected source

  根拠https://github.com/huntedman/JevLint/blob/main/src/jev-client.ts#L52

  ラベルcall-site-verified

• HyunjunJeon/jev-judgment

  README states: TypeSafe Jev and acts on the probabilities. The

  根拠https://github.com/HyunjunJeon/jev-judgment/blob/main/README.md#L6

  ラベルreadme-claim

• iamjonatha/decisionkit-dotnet

  README states: A strongly typed, provider-independent decision engine for .NET, with TypeSafe JEV support.

  根拠https://github.com/iamjonatha/decisionkit-dotnet/blob/main/README.md#L3

  ラベルreadme-claim

• iamtoomas/JevLint

  README states: JevLint is powered by Jev, TypeSafe AI’s System One model. It sends selected source

  根拠https://github.com/iamtoomas/JevLint/blob/main/src/jev-client.ts#L52

  ラベルcall-site-verified

• IAmUnbounded/save-token-jev-clean

  README states: | TYPESAFE_API_KEY | required unless stored in the macOS Keychain (CLI and library only) | TypeSafe/Jev API key |

  根拠https://github.com/IAmUnbounded/save-token-jev-clean/blob/main/README.md#L185

  ラベルreadme-claim

• iamvatsalpatel/tiershift

  README states: The judgment comes from TypeSafe Jev , a calibrated decision model. About 180 ms. Four cents per thousand routes.

  根拠https://github.com/iamvatsalpatel/tiershift/blob/main/bench/experiments/gate-experiment.ts#L10

  ラベルcall-site-verified

• iefnaf/pi-jev

  README states:

  根拠https://github.com/iefnaf/pi-jev/blob/main/README.md#L9

  ラベルreadme-claim

• ielab/llm-rankers

  README states: : New experiments with Jev (TypeSafe AI's non-generative "System One" model) as a pointwise / pairwise / setwise / listwise reranker, including all 100 candidates in a single request. Check out the jev folder for results

  根拠https://github.com/ielab/llm-rankers/blob/main/README.md#L7

  ラベルreadme-claim

• ilyamk/jev-gmail-ai-spam-filter-and-labeling

  README states: Semantic email classification powered by Jev, with confidence-aware automation,

  根拠https://github.com/ilyamk/jev-gmail-ai-spam-filter-and-labeling/blob/main/README.md#L11

  ラベルreadme-claim

• InternLM/Intern-Decision

  README states: Intern-Decision-4B reaches 90.02% average accuracy across seven suites (Jev: 88.74%),

  根拠https://github.com/InternLM/Intern-Decision/blob/main/README.md#L97

  ラベルreadme-claim

• intikhab49/open-jev-typed-decision-engine

  README states:

  根拠https://github.com/intikhab49/open-jev-typed-decision-engine/blob/main/README.md#L3

  ラベルreadme-claim

• ipenywis/laya-ultrafast

  README states: Hosted mode still works. Set DECISION_MODEL=typesafe to use the original Jev policy unchanged.

  根拠https://github.com/ipenywis/laya-ultrafast/blob/main/laya_ultrafast/laya.py#L48

  ラベルcall-site-verified

• ItIsCuthNotCup/MetaCog

  README states: judge = SystemOneJudge.jev() # hosted; reads $TYPESAFE_API_KEY

  根拠https://github.com/ItIsCuthNotCup/MetaCog/blob/main/README.md#L48

  ラベルreadme-claim

• itsmostafa/typesafe-mcp

  README states: Give your AI agent direct access to models like TypeSafe Jev, Liquid AI d1, Cloudflare Clef, Respan Span-01, GPT-6 Luna Decisions, and Perplexity Decider Flash: quick decisions with real probabilities.

  根拠https://github.com/itsmostafa/typesafe-mcp/blob/main/README.md#L10

  ラベルreadme-claim

• itsoltech/basal-rs

  README states:

  根拠https://github.com/itsoltech/basal-rs/blob/main/contracts/typesafe/README.md#L4

  ラベルreadme-claim

• j1s4nn/duomind

  README states: Internet connection for setup and Jev API calls

  根拠https://github.com/j1s4nn/duomind/blob/main/README.md#L111

  ラベルreadme-claim

• JamesANZ/us-legal-mcp

  README states: JEV_MODEL – default jev-latest

  根拠https://github.com/JamesANZ/us-legal-mcp/blob/main/README.md#L234

  ラベルreadme-claim

• jamesward/zio-typesafe-ai

  README states: Jev / System One API — a "System One model" that answers typed, atomic

  根拠https://github.com/jamesward/zio-typesafe-ai/blob/main/src/main/scala/com/jamesward/zio_typesafe_ai/TypeSafeAI.scala#L3

  ラベルcall-site-verified

• JanOstrowka/typesafe-assist

  README states:

  根拠https://github.com/JanOstrowka/typesafe-assist/blob/main/custom_components/typesafe_conversation/api.py#L161

  ラベルcall-site-verified

• jaredpalmer/kev

  README states: Small Jev-like decision models you can train and run yourself.

  根拠https://github.com/jaredpalmer/kev/blob/main/README.md#L3

  ラベルreadme-claim

• jarrodwatts/jev-trader

  README states: With no PRIVATE_KEY it dry-runs: real book, real decisions, simulated fills. Set MODEL=jev and TYPESAFE_AI_API_KEY to use Jev; the default mock is a momentum heuristic stand-in.

  根拠https://github.com/jarrodwatts/jev-trader/blob/main/src/model.ts#L2

  ラベルcall-site-verified

• jccrump/openbot

  README states: Google, Mistral, or a local model through Ollama or LM Studio. An optional Jev

  根拠https://github.com/jccrump/openbot/blob/main/README.md#L7

  ラベルreadme-claim

• jcpsimmons/jev-macos-loop

  README states: Read https://raw.githubusercontent.com/jcpsimmons/jev-macos-loop/master/AGENTS.md

  根拠https://github.com/jcpsimmons/jev-macos-loop/blob/master/src/providers.mjs#L17

  ラベルcall-site-verified

• JedimEmO/typesafe-client

  README states: endorsed by, or supported by TypeSafe AI. For the API itself, see the

  根拠https://github.com/JedimEmO/typesafe-client/blob/main/crates/typesafe-client/src/answer.rs#L405

  ラベルcall-site-verified

• jerryfane/omp-jev-compaction

  README states: TypeSafe's Jev decision model, over either the TypeSafe API or OpenRouter.

  根拠https://github.com/jerryfane/omp-jev-compaction/blob/master/src/asker.ts#L21

  ラベルcall-site-verified

• jev-chat/jev-chat-windows

  README states: 判断 · Jev —— 判断意图、紧张度，并给三条候选排序。来源选 OpenRouter（默认，key 在

  根拠https://github.com/jev-chat/jev-chat-windows/blob/main/README.md#L47

  ラベルreadme-claim

• jhd3197/Prompture

  README states: driver = get_decision_driver_for_model("typesafe/jev-latest")

  根拠https://github.com/jhd3197/Prompture/blob/main/prompture/drivers/kev_decision_driver.py#L4

  ラベルcall-site-verified

• jkudish/jev-mcp

  README states: Jev is TypeSafe's System One model: it returns typed answers with calibrated probability distributions, not generated text. A verify call is a Choice over supports / contradicts / says_nothing, so you see the whole distr

  根拠https://github.com/jkudish/jev-mcp/blob/main/README.md#L733

  ラベルreadme-claim

• jmanhype/jev-dspy-lab

  README states: and other callers that use TypeSafe decisions inside DSPy-style pipelines.

  根拠https://github.com/jmanhype/jev-dspy-lab/blob/main/scripts/verify_integration.py#L41

  ラベルcall-site-verified

• joelhooks/pi-fast-jev-compaction

  README states: Sends a compact representation of tool calls and surrounding conversation to TypeSafe's /v1/systemone endpoint.

  根拠https://github.com/joelhooks/pi-fast-jev-compaction/blob/main/src/core/request.ts#L8

  ラベルcall-site-verified

• jomatsu/pi-jev-auto-mode

  README states: decision models, with Jev (TypeSafe System One) as the default. OpenAI Decisions,

  根拠https://github.com/jomatsu/pi-jev-auto-mode/blob/main/src/jev/engine.ts#L95

  ラベルcall-site-verified

• jomatsu/zod-jev

  README states: A single parseAsync call sends every rule of that schema to Jev in a single request, then turns

  根拠https://github.com/jomatsu/zod-jev/blob/main/scripts/check-package.mjs#L35

  ラベルcall-site-verified

• jonathanavis96/jev-kit

  README states: What is Jev, and why a kit

  根拠https://github.com/jonathanavis96/jev-kit/blob/main/README.md#L39

  ラベルreadme-claim

• jozefRudy/job_search

  README states: Optional; omit either key to use the default (https://api.typesafe.ai, jev-latest).

  根拠https://github.com/jozefRudy/job_search/blob/main/README.md#L52

  ラベルreadme-claim

• jucelioalencar/linearmemory

  README states: Contracts: OpenAI Decisions and TypeSafe System One. Settings are available at GET/PUT /api/settings/search-filter; existing search snapshots are returned by GET /api/search-results/:eventId without rerunning retrieval.

  根拠https://github.com/jucelioalencar/linearmemory/blob/main/README.md#L175

  ラベルreadme-claim

• jujumilk3/jev-calibration-audit

  README states: An independent, API-only audit of the calibration of TypeSafe AI's Jev, a

  根拠https://github.com/jujumilk3/jev-calibration-audit/blob/main/src/jev_audit/client.py#L12

  ラベルcall-site-verified

• justinhe16/trade-jev

  README states: Run: a backtest that calls Jev and stores its answers and trades in runs/ /. It's the only step that costs money.

  根拠https://github.com/justinhe16/trade-jev/blob/main/src/trade_jev/live.py#L993

  ラベルcall-site-verified

• jxucoder/mimicry

  README states: and TypeSafe; requests and responses are saved locally under ignored

  根拠https://github.com/jxucoder/mimicry/blob/master/README.md#L37

  ラベルreadme-claim

• kavehmz/typesafe-playground

  README states: Small, interactive experiments with TypeSafe Jev: from understanding a support message to making decisions for a car in a 3D world.

  根拠https://github.com/kavehmz/typesafe-playground/blob/main/demo01/server.mjs#L67

  ラベルcall-site-verified

• keltokhy/jgrep

  README states: Each line becomes one yes/no question to Jev, TypeSafe's decision

  根拠https://github.com/keltokhy/jgrep/blob/main/README.md#L16

  ラベルreadme-claim

• kesku/jev-freeform

  README states: Can TypeSafe Jev Choice generate text through successive typed decisions? Well I tried it

  根拠https://github.com/kesku/jev-freeform/blob/main/main.mjs#L172

  ラベルcall-site-verified

• kevinbadi/jev-voice

  README states: Everything runs locally except one ~250 ms call to Jev (TypeSafe's System One

  根拠https://github.com/kevinbadi/jev-voice/blob/main/jev_voice/brain.py#L174

  ラベルcall-site-verified

• Kevthetech143/super-jev

  README states: One judge sits between your agent and your data: the agent asks in its own words, Jev finds the file, checks the claim, permits the action, and remembers what you approved. Your agent stays responsible for the answer.

  根拠https://github.com/Kevthetech143/super-jev/blob/main/README.md#L9

  ラベルreadme-claim

• kieranklaassen/ruby_llm-typesafe

  README states: Both register a provider named :typesafe, so don't load both. When you upgrade, remove ruby_llm-typesafe from your Gemfile and use RubyLLM's built-in provider.

  根拠https://github.com/kieranklaassen/ruby_llm-typesafe/blob/main/README.md#L11

  ラベルreadme-claim

• KiritoKing/midscene-jev-runner

  README states: The default is TypeSafe System One: POST https://api.typesafe.ai/v1/systemone

  根拠https://github.com/KiritoKing/midscene-jev-runner/blob/main/README.md#L66

  ラベルreadme-claim

• klauswg/jev-guard

  README states: Real-time risk triage gateway for crypto exchange deposits & withdrawals — Jev (TypeSafe System One) as the triage layer, deterministic code as the judge.

  根拠https://github.com/klauswg/jev-guard/blob/master/src/main/java/com/jevguard/client/JevClient.java#L13

  ラベルcall-site-verified

• klauswg/jev-suite

  README states: Four decision-quality tools built on Jev (TypeSafe System One), one shared kernel — Jev answers structured questions; deterministic code keeps the final say.

  根拠https://github.com/klauswg/jev-suite/blob/master/jev-fidelity/src/main/java/com/jevsuite/fidelity/eval/EvalRunner.java#L52

  ラベルcall-site-verified

• knowlet/jevlens

  README states: Manifest V3 Chrome extension for annotating articles, X/Twitter posts, and Threads posts with typed Jev / System One decisions.

  根拠https://github.com/knowlet/jevlens/blob/master/extension/lib/jev.js#L4

  ラベルcall-site-verified

• kouhxp/gutsy

  README states: network calls, no per-call cost. The API follows the Jev-style Decisions format, so existing clients

  根拠https://github.com/kouhxp/gutsy/blob/main/gutsy-inference/gutsy_inference/server.py#L21

  ラベルcall-site-verified

• kraayenjon/awesome-jev

  README states:

  根拠https://github.com/kraayenjon/awesome-jev/blob/main/README.md#L2

  ラベルreadme-claim

• kshetrajna12/reflex

  README states: the "System One" model TypeSafe released in September 2026, built on top of a normal

  根拠https://github.com/kshetrajna12/reflex/blob/main/scripts/bench_latency.py#L6

  ラベルcall-site-verified

• kyegomez/open-jev

  README states: System One Models and Jev

  根拠https://github.com/kyegomez/open-jev/blob/main/README.md#L161

  ラベルreadme-claim

• kyu1204/jgrep

  README states: jgrep runs on Jev, through TypeSafe or

  根拠https://github.com/kyu1204/jgrep/blob/main/README.md#L41

  ラベルreadme-claim

• kyu1204/oh-my-harness

  README states: --preset always wins and never calls Jev. No key and no LLM provider? omh init "description" falls back to safe and says so.

  根拠https://github.com/kyu1204/oh-my-harness/blob/main/src/catalog/blocks/semantic-rule-guard.ts#L68

  ラベルcall-site-verified

• lakeday-org/perch

  README states: export PERCH_BASE_URL=https://api.typesafe.ai/v1/systemone

  根拠https://github.com/lakeday-org/perch/blob/main/.benchmarks/reviewbench/run.mjs#L15

  ラベルcall-site-verified

• lhotwll217/jev-cli

  README states: JSON in, typed decisions out. A thin CLI over the TypeSafe System One API

  根拠https://github.com/lhotwll217/jev-cli/blob/main/src/client.ts#L25

  ラベルcall-site-verified

• Li-Evan/awesome-jev

  README states: http4k TypeSafe connector - TypeSafe connector for the http4k Kotlin toolkit with a typed System One client and a fake, exposing Jev Noul, Choice and Score questions as http4k actions.

  根拠https://github.com/Li-Evan/awesome-jev/blob/main/README.md#L468

  ラベルreadme-claim

• libingzheren/Jev-Mem

  README states: Jev-Mem separates the frequent decisions of memory management from the deeper

  根拠https://github.com/libingzheren/Jev-Mem/blob/main/memory/jev_client.py#L23

  ラベルcall-site-verified

• liu-x27/XavierJev

  README states: TypeSafe: nothing here calls the Jev API, and nothing was trained on its output.

  根拠https://github.com/liu-x27/XavierJev/blob/main/eval/decision-models/run.ts#L177

  ラベルcall-site-verified

• liuyanghejerry/Clausura

  README states: model: jev-latest # Default. Decision model id.

  根拠https://github.com/liuyanghejerry/Clausura/blob/main/README.md#L330

  ラベルreadme-claim

• luisleo526/doc2mark

  README states: TypeSafe's Jev can be asked about the cases they leave open (is this

  根拠https://github.com/luisleo526/doc2mark/blob/main/README.md#L23

  ラベルreadme-claim

• lykycy123/RoboJEV

  README states: Two-stage JEV control of a Franka Panda in MuJoCo. Measured simulator state → task intent → Cartesian motion and gripper commands.

  根拠https://github.com/lykycy123/RoboJEV/blob/main/src/jev_vla_sim/config.py#L27

  ラベルcall-site-verified

• lyramakesmusic/jevbot

  README states: Completeness judge: A separate noul question asks "is the reply complete?" — jev stops when it thinks it's done

  根拠https://github.com/lyramakesmusic/jevbot/blob/master/README.md#L11

  ラベルreadme-claim

• maayanlevy/mysql-ailike

  README states: Requires a TypeSafe API key. Note: evaluated values and prompts are

  根拠https://github.com/maayanlevy/mysql-ailike/blob/main/src/typesafe.h#L16

  ラベルcall-site-verified

• madisonrickert/jev-permission-gate

  README states: A shell-aware blocklist sits in front of Jev. A call it flags can still be denied by Jev but is never allowed, and a call that carries a secret never leaves your machine.

  根拠https://github.com/madisonrickert/jev-permission-gate/blob/main/hooks/register.ts#L25

  ラベルcall-site-verified

• MahmoudAdelbghany/jev-browser

  README states: one Jev call (parallel questions, ~300 ms, no LLM tokens)

  根拠https://github.com/MahmoudAdelbghany/jev-browser/blob/main/src/jev.mjs#L1

  ラベルcall-site-verified

• mahynotch/newsscore

  README states: News sentiment scoring for stocks, powered by TypeSafe Jev by default. Mainstream financial news APIs go in, a pluggable

  根拠https://github.com/mahynotch/newsscore/blob/main/src/newsscore/scoring/jev.py#L416

  ラベルcall-site-verified

• maker-KK/todo-jev

  README states: A task-routing experiment powered by TypeSafe Jev, structured skill profiles, and basic environment checks.

  根拠https://github.com/maker-KK/todo-jev/blob/main/app/classifier.py#L14

  ラベルcall-site-verified

• malDuffin/typesafe-3d-chess

  README states: D chess powered by TypeSafe AI (Jev). AI vs AI by default, or play either side. Multiple difficulty levels.

  根拠https://github.com/malDuffin/typesafe-3d-chess/blob/main/README.md#L2

  ラベルreadme-claim

• mandu5/jevcompat

  README states: About a hundred servers say they are "Jev-compatible". Nothing said what that means. Now something does.

  根拠https://github.com/mandu5/jevcompat/blob/main/README.md#L3

  ラベルreadme-claim

• manifoldor/xtags

  README states: 页面文本会发送到所选 API 服务（默认 TypeSafe） 做判断——这是它工作的必要部分。发送的是帖子正文和作者 handle；折叠长推文会发送 X 页面数据中已有的全文，即使你尚未点击展开。

  根拠https://github.com/manifoldor/xtags/blob/main/extension/service.js#L3

  ラベルcall-site-verified

• marcreichel/laya-php

  README states: curl -fsSL https://api.typesafe.ai/openapi.json -o openapi.json

  根拠https://github.com/marcreichel/laya-php/blob/main/src/Laya.php#L197

  ラベルcall-site-verified

• MarissaFamularo/citation-verifier

  README states: supports the sentence, and Jev's confidence in its own call. A probability

  根拠https://github.com/MarissaFamularo/citation-verifier/blob/main/README.md#L34

  ラベルreadme-claim

• markjaquith/typesafe-ai-playground

  README states: The experiments live in a Rust CLI called typesafe-ai, built with

  根拠https://github.com/markjaquith/typesafe-ai-playground/blob/main/src/typesafe.rs#L57

  ラベルcall-site-verified

• MartinPuli/f1

  README states: The engine and experiment scripts remain in the repository. The published interface only plays recorded races; it has no creation or admin screen. Replays use saved decisions and frames without calling TypeSafe.

  根拠https://github.com/MartinPuli/f1/blob/main/scripts/refine-prompts.mjs#L55

  ラベルcall-site-verified

• mattn/go-jev

  README states: text. Comes with jev-cli, a command-line tool built on it for UNIX pipelines.

  根拠https://github.com/mattn/go-jev/blob/main/jev.go#L19

  ラベルcall-site-verified

• mejiasd3v/pi-jev-router

  README states:

  根拠https://github.com/mejiasd3v/pi-jev-router/blob/main/index.ts#L407

  ラベルcall-site-verified

• memos-im/xianyu-slang-helper

  README states: 追加 --ai-review 可用 TypeSafe System One 判别模型（Jev；自托管 DiffusionGemma 同 API）剔除上游的"模板化噪音"变体：每个候选变体问一个 Noul（是/否）问题，取概率 ≥ 阈值判为噪音，不生成文本、无解析、返回校准概率。未配置 key 时自动降级跳过，不阻断同步。

  根拠https://github.com/memos-im/xianyu-slang-helper/blob/main/README.md#L147

  ラベルreadme-claim

• Micha0827/snapjudge

  README states: The HTTP API follows the TypeSafe System One format

  根拠https://github.com/Micha0827/snapjudge/blob/main/README.md#L7

  ラベルreadme-claim

• misbahsy/doc-router

  README states: Jev (TypeSafe System One) — the page judge. Given the

  根拠https://github.com/misbahsy/doc-router/blob/main/README.md#L36

  ラベルreadme-claim

• mithalouni/system-one-open

  README states: Base URL: https://mithalouni--jev-serve-e2b-full-server-web.modal.run (scales to zero; first request after idle takes ~60 s).

  根拠https://github.com/mithalouni/system-one-open/blob/main/README.md#L31

  ラベルreadme-claim

• mizchi/jev-lint

  README states: export TYPESAFE_API_KEY=... # https://typesafe.ai

  根拠https://github.com/mizchi/jev-lint/blob/main/src/clef.ts#L54

  ラベルcall-site-verified

• mizchi/jev-test-filter

  README states: jev-test-filter reads a git diff, asks a model how much that change can

  根拠https://github.com/mizchi/jev-test-filter/blob/main/src/jev.ts#L26

  ラベルcall-site-verified

• mkotlikov/jev-grug

  README states: The demo now connects to the real Jev API when TYPESAFE_API_KEY is configured. Without a key it falls back to a deterministic mock, so contributors can still run the interface immediately.

  根拠https://github.com/mkotlikov/jev-grug/blob/main/app/api/chat/route.ts#L30

  ラベルcall-site-verified

• mmastrac/djev

  README states: POST /v1/systemone implements the Jev decision API. The body holds

  根拠https://github.com/mmastrac/djev/blob/main/structured_server.py#L2014

  ラベルcall-site-verified

• mmastrac/djev-spark

  README states: it implements Jev's POST /v1/systemone API and passes ordinary chat through

  根拠https://github.com/mmastrac/djev-spark/blob/main/scripts/long-context-probe.py#L35

  ラベルcall-site-verified

• mohsen1/llm-debugger-vscode-extension

  README states: Each step's "what next?" is answered by Jev, a

  根拠https://github.com/mohsen1/llm-debugger-vscode-extension/blob/main/README.md#L7

  ラベルreadme-claim

• morganlinton/Albatross

  README states: Jev Mode puts TypeSafe AI's decision model before the main LLM. Supported questions

  根拠https://github.com/morganlinton/Albatross/blob/main/src/jev.rs#L13

  ラベルcall-site-verified

• moritzkremb/jev-voice-browser

  README states: the server asks Jev (TypeSafe's System One model, jev-1.13.0) one request with a dozen typed

  根拠https://github.com/moritzkremb/jev-voice-browser/blob/main/src/jev.js#L7

  ラベルcall-site-verified

• mossyfield/ST-jeved

  README states: Jeved is a SillyTavern extension. It sends chat messages to Jev, a small decision model, with

  根拠https://github.com/mossyfield/ST-jeved/blob/main/src/classifier.js#L22

  ラベルcall-site-verified

• mrbanana55/Decision_Blackjack

  README states: Jev (Remote Cloud AI via TypeSafe System One)

  根拠https://github.com/mrbanana55/Decision_Blackjack/blob/main/README.md#L8

  ラベルreadme-claim

• mrmps/classifier-dev

  README states: Both tiers answer from TypeSafe's Jev, a decision

  根拠https://github.com/mrmps/classifier-dev/blob/main/README.md#L128

  ラベルreadme-claim

• n42blockchain/N42-rs

  README states: 用户在页面确认费用并提交，页面展示 requestId。运营方建立 job.json，包含下方字段，运行 TYPESAFE_API_KEY=... N42_DECISION_SIGNING_KEY=... cargo run -p n42-decision-relay -- evaluate job.json ./archive > result.json。TYPESAFE_API_URL 默认官方 https://api.typesa

  根拠https://github.com/n42blockchain/N42-rs/blob/main/ddn/docs/decision/README.md#L44

  ラベルreadme-claim

• Nainish-Rai/jev-frontend-qa

  README states: Bounded Jev decisions: typed operation/target choices over observed controls; exact fixture values and correctness checks remain in code.

  根拠https://github.com/Nainish-Rai/jev-frontend-qa/blob/main/src/jev_frontend_qa/core/model_client.py#L3

  ラベルcall-site-verified

• Nancy-Chauhan/hearth-jev-rental-search

  README states: A TypeSafe API key. A search costs a fraction of a cent

  根拠https://github.com/Nancy-Chauhan/hearth-jev-rental-search/blob/main/README.md#L18

  ラベルreadme-claim

• NandhaKishorM/laya

  README states: wire protocol as TypeSafe's hosted Jev API. Laya's answer payload is already

  根拠https://github.com/NandhaKishorM/laya/blob/main/laya-dotnet/tools/bench_python.py#L191

  ラベルcall-site-verified

• Nasrallah-AL/jev-cli

  README states: Fast, typed AI judgments from the command line, powered by TypeSafe's Jev model.

  根拠https://github.com/Nasrallah-AL/jev-cli/blob/main/README.md#L8

  ラベルreadme-claim

• nassim-arifette/jevgrep

  README states: authorized repository, asks Jev to score all eligible fragments, then returns the

  根拠https://github.com/nassim-arifette/jevgrep/blob/main/experiments/jev-contract/offline-worker.mjs#L4

  ラベルcall-site-verified

• ndolinschi/harnessjudge

  README states: After an agent step (paste a trace) → ok / retry / escalate / stop — for agent builders, powered by TypeSafe Jev.

  根拠https://github.com/ndolinschi/harnessjudge/blob/main/src/app/api/decide/route.ts#L7

  ラベルcall-site-verified

• ndolinschi/pulselane

  README states: Clinic triage decisions powered by TypeSafe Jev (System One).

  根拠https://github.com/ndolinschi/pulselane/blob/main/src/app/api/decide/route.ts#L9

  ラベルcall-site-verified

• ndolinschi/toolgate

  README states: Agent tool/MCP call gate — paste a planned tool call + context → allow / ask_human / deny with risk scores via TypeSafe Jev.

  根拠https://github.com/ndolinschi/toolgate/blob/main/src/app/api/decide/route.ts#L9

  ラベルcall-site-verified

• newfull5/malkuth

  README states: suites and about 129,000 questions. Every model answered the same items. Jev was measured through its API.

  根拠https://github.com/newfull5/malkuth/blob/main/tools/laya_bench.py#L27

  ラベルcall-site-verified

• NiazMorshed2007/jev-review

  README states: Your API key stays on your machine. Jev Review has no hosted backend, database, telemetry service, or author-operated proxy. The only remote request is sent directly to the configured Jev API.

  根拠https://github.com/NiazMorshed2007/jev-review/blob/main/src/jev/client.ts#L4

  ラベルcall-site-verified

• nican2018/shade-arena-jev-monitor

  README states: Strong per-action gate. Scoring each tool call before it runs, Jev ranked sabotage above almost all benign actions. Gemini Flash's scores piled up at 0 and 100.

  根拠https://github.com/nican2018/shade-arena-jev-monitor/blob/main/llms/typesafe_llm.py#L20

  ラベルcall-site-verified

• nicobailon/pi-powerline-footer

  README states: When the agent is working, Enter sends your message as steering and Alt+Enter queues it as a follow-up. Set autoFollowUp under powerline to let TypeSafe's Jev make that call for plain Enter:

  根拠https://github.com/nicobailon/pi-powerline-footer/blob/main/README.md#L329

  ラベルreadme-claim

• nicobailon/surf-cli

  README states: --allow-semantic - Opt in to bounded model decisions for a semantic workflow (model from SURF_SEMANTIC_MODEL, default jev-1.13.0; extract steps always use clairic with Cloudflare credentials)

  根拠https://github.com/nicobailon/surf-cli/blob/main/native/semantic-cli.cjs#L632

  ラベルcall-site-verified

• nidhi-singh02/agent-router

  README states: shared accounts), asks TypeSafe to rank what is left and choose an effort, starts the chosen

  根拠https://github.com/nidhi-singh02/agent-router/blob/main/packages/router/src/commands/effort.ts#L264

  ラベルcall-site-verified

• Nisaka520/JevIntent

  README states: POST https://api.typesafe.ai/v1/systemone：

  根拠https://github.com/Nisaka520/JevIntent/blob/main/README.md#L232

  ラベルreadme-claim

• NobleSpartan6/otto

  README states: In Hybrid mode, GPT-6 Astra provides a plan and generated text when needed. Jev chooses concrete actions between planning calls.

  根拠https://github.com/NobleSpartan6/otto/blob/main/core/typesafe.ts#L4

  ラベルcall-site-verified

• noetion/dsh-jev

  README states: dsh-jev is a DeepSeek Harness bundle that registers the jev_ask tool. A DSH agent on a profile that loads it can send typed noul, choice, and score questions to TypeSafe Jev.

  根拠https://github.com/noetion/dsh-jev/blob/main/src/client.ts#L10

  ラベルcall-site-verified

• NorbertBodziony/guard-jev

  README states: Text moderation demo: one TypeSafe systemOne call screens 7 Noul hazards + 1 severity Score in parallel. Verdict computed in code via policy thresholds.

  根拠https://github.com/NorbertBodziony/guard-jev/blob/main/app/api/moderate/route.ts#L2

  ラベルcall-site-verified

• notque/vexjoy-agent

  README states: /d — Jev-Powered Router

  根拠https://github.com/notque/vexjoy-agent/blob/main/README.md#L44

  ラベルreadme-claim

• ntedvs/commentcop

  README states: Use --dry-run to preview what it finds without calling Jev. Add exclusions with --ignore or a .commentcopignore file. --help has the rest.

  根拠https://github.com/ntedvs/commentcop/blob/main/src/judge.ts#L1

  ラベルcall-site-verified

• Nyarlathoteppppp/pi-heed

  README states: messages, 9,090 real tool calls) through pi-heed in enforce mode with Jev, without executing anything

  根拠https://github.com/Nyarlathoteppppp/pi-heed/blob/main/src/judge.ts#L53

  ラベルcall-site-verified

• obie/ruby_decision_model

  README states: Requests go to https://api.typesafe.ai/v1/systemone. The default model is

  根拠https://github.com/obie/ruby_decision_model/blob/main/README.md#L85

  ラベルreadme-claim

• oficcejo/aiagents-stock

  README states: Key 申请：https://console.typesafe.ai ｜ 完整说明：TypeSafe Jev决策引擎接入说明

  根拠https://github.com/oficcejo/aiagents-stock/blob/main/README.md#L170

  ラベルreadme-claim

• ojusave/beat-jev

  README states: A penalty shootout against TypeSafe Jev. Take turns shooting and keeping goal. One Render Workflows task owns the match, with scores saved in Render Postgres.

  根拠https://github.com/ojusave/beat-jev/blob/main/python/app/keeper.py#L2

  ラベルcall-site-verified

• Olli0103/openclaw-typesafe-ai

  README states: Jev returns decisions over predefined answer spaces. It does not generate conversation, prose or code. This package therefore registers a tool and no model provider, model picker entry, hook or background service.

  根拠https://github.com/Olli0103/openclaw-typesafe-ai/blob/main/src/client.ts#L4

  ラベルcall-site-verified

• olsenbrands/fable-foreman

  README states: Jev, from TypeSafe AI, is a decision model rather than a chatbot. It answers narrow yes/no, pick-one and score questions in under a second, for about 0.02 cents a call. Fable Foreman uses it to point the lead's attention

  根拠https://github.com/olsenbrands/fable-foreman/blob/main/README.md#L11

  ラベルreadme-claim

• OmniJev/OneJev

  README states:

  根拠https://github.com/OmniJev/OneJev/blob/main/qev/cli.py#L143

  ラベルcall-site-verified

• ondata/vd-jev

  README states: Ask three questions about the message column, one jev-column each:

  根拠https://github.com/ondata/vd-jev/blob/main/README.md#L55

  ラベルreadme-claim

• onlyoneaman/jev-eval

  README states: Jev is a model that does not write text. You give it some text and a question with fixed answers, and it gives back a probability for each answer. One call looks like this:

  根拠https://github.com/onlyoneaman/jev-eval/blob/main/jev_eval/backends.py#L16

  ラベルcall-site-verified

• opencx-labs/zevals

  README states: Jev is reachable through several gateways, and each names the same yes/no question differently. Three clients are bundled, all of them plain fetch with no extra dependencies:

  根拠https://github.com/opencx-labs/zevals/blob/main/README.md#L273

  ラベルreadme-claim

• openlayer-ai/jevals

  README states: export AI_GATEWAY_API_KEY=... # Jev through Vercel AI Gateway. TYPESAFE_API_KEY and OPENROUTER_API_KEY also work.

  根拠https://github.com/openlayer-ai/jevals/blob/main/src/jevals/backends/_base.py#L65

  ラベルcall-site-verified

• OskarLebuda/precog

  README states: Asks TypeSafe Jev which link a visitor is about to click, then warms

  根拠https://github.com/OskarLebuda/precog/blob/main/README.md#L19

  ラベルreadme-claim

• oso95/x-scanner

  README states: Behavioral labels on every post you scroll past on X, judged by Jev,

  根拠https://github.com/oso95/x-scanner/blob/main/README.md#L3

  ラベルreadme-claim

• PanAchy/jevvy

  README states: Custom endpoints use Jevvy's default approval policy unless you replace it. Its calibration evidence applies only when the route is verified to serve the calibrated Jev model.

  根拠https://github.com/PanAchy/jevvy/blob/main/packages/core/src/system-one.ts#L12

  ラベルcall-site-verified

• pavy23/web-doom

  README states: Tactical policy (TypeSafe System One, Jev). At each step with an enemy

  根拠https://github.com/pavy23/web-doom/blob/main/README.md#L288

  ラベルreadme-claim

• pengchujin/ad-radar

  README states: 智能识别：用你自己的 Jev API key，识别广告（硬广 / 软广 / 引流）和你选择的话题（AI、科技、军事、政治、性别对立、色情低俗、明星八卦、炒股理财、体育、游戏，也可以自己添加），标在内容旁边或折叠起来；还能估计内容由 AI 生成的比例。

  根拠https://github.com/pengchujin/ad-radar/blob/main/README.md#L6

  ラベルreadme-claim

• PerryLink/jevcore

  README states: TypeSafe Jev for DeepSeek Harness

  根拠https://github.com/PerryLink/jevcore/blob/main/README.md#L8

  ラベルreadme-claim

• peterfriese/system-one-foundation-models

  README states: Jev Cloud: TypeSafe AI hosted service on https://api.typesafe.ai.

  根拠https://github.com/peterfriese/system-one-foundation-models/blob/main/README.md#L364

  ラベルreadme-claim

• philippdubach/pi-jev-router

  README states: pi-jev-router is a pi extension. It classifies each task with Jev (TypeSafe

  根拠https://github.com/philippdubach/pi-jev-router/blob/main/src/classifier.ts#L13

  ラベルcall-site-verified

• picaye/jev-compaction

  README states: Jev judges to be stale, and it shows Jev the whole conversation while deciding.

  根拠https://github.com/picaye/jev-compaction/blob/main/README.md#L19

  ラベルreadme-claim

• PistachioAIHQ/jev-synergy-screening

  README states: export TYPESAFE_API_KEY="$(tr -d ' ' < ~/.config/typesafe/api_key)"

  根拠https://github.com/PistachioAIHQ/jev-synergy-screening/blob/main/src/config.py#L43

  ラベルcall-site-verified

• pithings/advocaat

  README states: Requests use https://ai-gateway.vercel.sh/v4/ai and model typesafe-ai/jev.

  根拠https://github.com/pithings/advocaat/blob/main/src/api.ts#L127

  ラベルcall-site-verified

• PixelML/av

  README states: Configure a TypeSafe System One key to make Jev refinement automatic for av ask:

  根拠https://github.com/PixelML/av/blob/main/README.md#L35

  ラベルreadme-claim

• pnthn-ai/polar_llama

  README states: TypeSafe System One: A native Rust inference layer for the TypeSafe API — typed, calibrated yes/no, choice and score questions answered per row, or one Pydantic contract applied to every line of a document in a single re

  根拠https://github.com/pnthn-ai/polar_llama/blob/main/README.md#L22

  ラベルreadme-claim

• PromptEngineer48/laya-vs-jev-arena

  README states: docker run -d --name arena --restart unless-stopped -p 80:8732 --cpus 1 --memory 512m -e TYPESAFE_API_KEY=your_key_here ghcr.io/promptengineer48/laya-vs-jev-arena:latest

  根拠https://github.com/PromptEngineer48/laya-vs-jev-arena/blob/main/server.py#L29

  ラベルcall-site-verified

• psyb0t/aigate

  README states: | decidealot _(optional, DECIDEALOT=1 and/or DECIDEALOT_CUDA=1)_ | Typed decisions with local Laya, Von, CLM, and hosted TypeSafe Jev over the TypeSafe System One API. POST /v1/systemone takes a model selector, a state,

  根拠https://github.com/psyb0t/aigate/blob/main/README.md#L179

  ラベルreadme-claim

• PyModel/jev-judge-mcp

  README states: uvx --from 'jev-judge-mcp[typesafe]' jev-judge-mcp setup

  根拠https://github.com/PyModel/jev-judge-mcp/blob/main/evals/confinement/broker.py#L16

  ラベルcall-site-verified

• qddegtya/qualm

  README states: Jev is the model. This is what qualm adds on top of calling its HTTP API yourself.

  根拠https://github.com/qddegtya/qualm/blob/main/README.md#L25

  ラベルreadme-claim

• qkal/Canny

  README states: Jev

  根拠https://github.com/qkal/Canny/blob/main/README.md#L155

  ラベルreadme-claim

• QuentinDanblon/pi-fast-jev-compaction

  README states: anything. It asks Jev, over the whole conversation, whether each old tool call and each

  根拠https://github.com/QuentinDanblon/pi-fast-jev-compaction/blob/main/index.ts#L206

  ラベルcall-site-verified

• Query-farm/vgi-typesafe

  README states:

  根拠https://github.com/Query-farm/vgi-typesafe/blob/main/vgi_typesafe/mock_server.py#L39

  ラベルcall-site-verified

• r-ms/mini-jev

  README states: What a Jev-style interface looks like on a frozen model: classify against a schema that arrives with the request by reading next-token logits, not by generating JSON. Measured on Qwen3-4B.

  根拠https://github.com/r-ms/mini-jev/blob/main/README.md#L3

  ラベルreadme-claim

• RafalWilinski/vibecheck

  README states: A Chrome and Firefox extension that runs your draft post through TypeSafe's Jev before you hit Post. It renders a scorecard under the composer with virality, funniness, informativeness, clarity, insult level, ragebait, c

  根拠https://github.com/RafalWilinski/vibecheck/blob/main/README.md#L3

  ラベルreadme-claim

• rajivkuriakose/typesafe-jev-examples

  README states: TypeSafe's direct API is behind an early-access waitlist. OpenRouter serves the

  根拠https://github.com/rajivkuriakose/typesafe-jev-examples/blob/main/src/jevx/client.py#L17

  ラベルcall-site-verified

• ranjan2829/AskJev

  README states: Talk to Claude in plain English. AskJev drives Brave/Chrome. TypeSafe Jev decides on-page. Guard freezes irreversible clicks.

  根拠https://github.com/ranjan2829/AskJev/blob/main/mcp/src/jev-client.ts#L16

  ラベルcall-site-verified

• RayenMlayeh/jevcrawl

  README states: | TYPESAFE_API_KEY | Jev API key (required) | none |

  根拠https://github.com/RayenMlayeh/jevcrawl/blob/main/README.md#L202

  ラベルreadme-claim

• rbalch/typesafeai-review

  README states: slices the diff, asks Jev (TypeSafe's System One model) narrow typed questions per hunk,

  根拠https://github.com/rbalch/typesafeai-review/blob/master/README.md#L4

  ラベルreadme-claim

• rchovatiya88/cyber-breach-jev

  README states: An infinite 3D arcade cyberpunk arena shooter built with Three.js and powered by Jev—the sub-100ms "System One" decision model developed by TypeSafe AI (released September 2026).

  根拠https://github.com/rchovatiya88/cyber-breach-jev/blob/main/backend/bot_pilot.py#L50

  ラベルcall-site-verified

• reachjalil/jevlogs

  README states:

  根拠https://github.com/reachjalil/jevlogs/blob/main/README.md#L2

  ラベルreadme-claim

• ReallyArtificial/stuntdouble

  README states: Find out whether a local decision model can replace Jev on your traffic before you switch.

  根拠https://github.com/ReallyArtificial/stuntdouble/blob/main/sidecars/kev-space.mjs#L42

  ラベルcall-site-verified

• realZachi/pg-jev

  README states:

  根拠https://github.com/realZachi/pg-jev/blob/master/README.md#L2

  ラベルreadme-claim

• realZachi/typesafe-adblock

  README states: BYOK: bring your own key. There is no backend. You paste your own TypeSafe API key

  根拠https://github.com/realZachi/typesafe-adblock/blob/main/src/typesafe.js#L5

  ラベルcall-site-verified

• receptron/laya

  README states: implementation (RLAgent.system_one) and as TypeSafe Jev's system_one API, and the output

  根拠https://github.com/receptron/laya/blob/main/README.md#L15

  ラベルreadme-claim

• reinhard-z/vision-jev

  README states: Move an object to ask Jev again. Drag it off the road or click its × to remove it.

  根拠https://github.com/reinhard-z/vision-jev/blob/main/README.md#L24

  ラベルreadme-claim

• RenaGao/jev-dataops

  README states: Screenshots show the repository's public demo running locally with its bundled synthetic example. The measurements demonstrate the workflow, not JEV screening quality or LLM performance.

  根拠https://github.com/RenaGao/jev-dataops/blob/main/jev_dataops/jev.py#L23

  ラベルcall-site-verified

• rohanarun/computer-use-cache

  README states: Set your own TypeSafe API key to enable JEV in either server:

  根拠https://github.com/rohanarun/computer-use-cache/blob/main/server.py#L85

  ラベルcall-site-verified

• rokbenko/quackd

  README states: quackd run arm-grip-check --robot arm-01 --by-hand --decision-llm jev --decision-mode shadow

  根拠https://github.com/rokbenko/quackd/blob/main/quackd/agent/decision/systemone.py#L100

  ラベルcall-site-verified

• RomanSlack/jev-drone

  README states: jev-drone

  根拠https://github.com/RomanSlack/jev-drone/blob/main/README.md#L1

  ラベルreadme-claim

• romeromarcelo/jev-retrieval

  README states:

  根拠https://github.com/romeromarcelo/jev-retrieval/blob/main/src/jev/client.rs#L28

  ラベルcall-site-verified

• rorshopping/jev-on-a-laptop

  README states: TypeSafe's own workflow evals (https://evals.typesafe.ai/) put Jev at 67.8% mean accuracy (61.7–76.0% per workflow) against a frontier consensus (average of GPT-6 Astra and Fable 5.1 answering every question), at $0.0004

  根拠https://github.com/rorshopping/jev-on-a-laptop/blob/main/README.md#L103

  ラベルreadme-claim

• runta-dev/jot

  README states: 

  根拠https://github.com/runta-dev/jot/blob/main/README.md#L19

  ラベルreadme-claim

• rupeshpoojary9/poorjev

  README states: The poor man's Jev. An open source, local-first "System One" decision layer for LLM apps: typed decisions with provably calibrated confidence . No API key. No waitlist.

  根拠https://github.com/rupeshpoojary9/poorjev/blob/main/crossbench/jev_client.py#L8

  ラベルcall-site-verified

• ryana/jevify

  README states: TypeSafe introduction

  根拠https://github.com/ryana/jevify/blob/main/README.md#L15

  ラベルreadme-claim

• safzanpirani/pi-jev-skill-picker

  README states: One skill_search call costs about 38,600 Jev input tokens, or $0.0016 at $42 per billion. Jev bills input only. Against gpt-6-astra that is under 1% of what a single un-stripped request wastes.

  根拠https://github.com/safzanpirani/pi-jev-skill-picker/blob/master/extensions/jev.ts#L12

  ラベルcall-site-verified

• samdotmak/jev-recall

  README states: Jev Recall gives you LLM re-ranker quality retreival at semantic search prices and speed

  根拠https://github.com/samdotmak/jev-recall/blob/main/web/lib/systems.ts#L146

  ラベルcall-site-verified

• samtay32/jev-system-architect

  README states: A system-architecture skill for TypeSafe AI's Jev / System One. Continuously asks: *Where does this system contain fuzzy semantic judgment that should become a small Jev decision primitive?*

  根拠https://github.com/samtay32/jev-system-architect/blob/main/README.md#L3

  ラベルreadme-claim

• santos-sanz/jev-audio-beeper

  README states: Private TypeScript proof of concept: identify Spanish profanity with Jev's typed probabilistic decisions and replace the matching word intervals with a beep without changing the audio duration.

  根拠https://github.com/santos-sanz/jev-audio-beeper/blob/main/index.ts#L1

  ラベルcall-site-verified

• savka777/jev-use

  README states:

  根拠https://github.com/savka777/jev-use/blob/main/Sources/JevCore/Decision.swift#L177

  ラベルcall-site-verified

• schwarzschlyle/immune

  README states: export TYPESAFE_API_KEY=... # TypeSafe Jev, which Immune asks about each call

  根拠https://github.com/schwarzschlyle/immune/blob/main/src/immune/sensing/jev.py#L15

  ラベルcall-site-verified

• scienthoon/jev-ood-calibration

  README states: Jev's probabilities are quantised to 0.01 and Choice/Score use the endpoints freely, so a correct answer sometimes receives exactly 0. A temperature cannot

  根拠https://github.com/scienthoon/jev-ood-calibration/blob/main/README.md#L39

  ラベルreadme-claim

• sdras/system-one-explainer

  README states: TypeSafe’s System One terminology and Jev quickstart are discussed in the final case study. Choice, Score, Noul, and distribution-derived confidence are specific API details.

  根拠https://github.com/sdras/system-one-explainer/blob/main/README.md#L31

  ラベルreadme-claim

• SeeAPI/awesome-jev-use-cases

  README states:

  根拠https://github.com/SeeAPI/awesome-jev-use-cases/blob/main/recipes/model-routing/run.py#L14

  ラベルcall-site-verified

• sgzmd/rss-morning

  README states: Jev classification on cheap RSS metadata (TypeSafe System One)

  根拠https://github.com/sgzmd/rss-morning/blob/main/README.md#L16

  ラベルreadme-claim

• shaharia-lab/jev-cli

  README states:

  根拠https://github.com/shaharia-lab/jev-cli/blob/main/crates/jev-cli/src/help/docs.rs#L813

  ラベルcall-site-verified

• shamazharikh/qwen-rlcd

  README states: Training uses RLCD (Reinforcement Learning for Calibrated Decisions). The objective is decisions plus calibrated probabilities. TypeSafe argues that RLHF hurts calibration through sycophancy and mode dropping.

  根拠https://github.com/shamazharikh/qwen-rlcd/blob/main/README.md#L160

  ラベルreadme-claim

• shantanugoel/ask-jev-skill

  README states: Hermes skill: call TypeSafe Jev (jev-latest) as a typed tiebreaker. Jev is not a chat model. It returns Choice / Score / Noul answers with probabilities so an agent can act or escalate.

  根拠https://github.com/shantanugoel/ask-jev-skill/blob/main/scripts/askjev.py#L15

  ラベルcall-site-verified

• shantanugoel/mario-jev

  README states: Jev receives structured RAM observations and answers focused questions about movement, starting a jump, and sustaining

  根拠https://github.com/shantanugoel/mario-jev/blob/main/src/mario_jev/policy.py#L3

  ラベルcall-site-verified

• sharziki/semdecide

  README states: Typed semantic decisions for Unix pipelines and CI, powered by TypeSafe AI Jev.

  根拠https://github.com/sharziki/semdecide/blob/main/src/reflex_guard/cli.py#L180

  ラベルcall-site-verified

• shhivv/arc-cua

  README states: not use decision models, JEV or TypeSafe; it is a standalone driver.

  根拠https://github.com/shhivv/arc-cua/blob/master/src/arc_cua/policies/typesafe.py#L26

  ラベルcall-site-verified

• shinshin86/aituber-onair

  README states: | typesafe | TypeSafe AI | jev-latest | https://api.typesafe.ai/v1/systemone |

  根拠https://github.com/shinshin86/aituber-onair/blob/main/packages/comment-intelligence/README.md#L291

  ラベルreadme-claim

• shitianfang/jev-use

  README states: hand the tasks that need no text output to Jev — faster steps, fewer

  根拠https://github.com/shitianfang/jev-use/blob/main/README.md#L7

  ラベルreadme-claim

• shivendrasoni/vector-cache

  README states: The default judge is TypeSafe Jev (vendor-quoted ~100 ms per call).

  根拠https://github.com/shivendrasoni/vector-cache/blob/main/README.md#L98

  ラベルreadme-claim

• ShuhanSun/jev-oas-sentinel

  README states: jev-oas-sentinel compares two OpenAPI documents in two layers:

  根拠https://github.com/ShuhanSun/jev-oas-sentinel/blob/main/src/jev_oas_sentinel/jev.py#L19

  ラベルcall-site-verified

• sileod/tasksource

  README states: (pick one option), score (an ordered scale) and noul (yes/no or a probability). This is the Jev / System One request format.

  根拠https://github.com/sileod/tasksource/blob/main/README.md#L76

  ラベルreadme-claim

• socai-io/jev-social

  README states:

  根拠https://github.com/socai-io/jev-social/blob/main/src/classifier.js#L222

  ラベルcall-site-verified

• sonnylazuardi/superterminal

  README states: The ranking comes from Jev,

  根拠https://github.com/sonnylazuardi/superterminal/blob/main/README.md#L58

  ラベルreadme-claim

• sorrycc/typesafe-snake

  README states: TypeSafe's Jev (a System One decision model, not an LLM) plays Snake in real time.

  根拠https://github.com/sorrycc/typesafe-snake/blob/master/server/index.ts#L2

  ラベルcall-site-verified

• sosopop/jev_stock

  README states: My current view is that JEV is more useful here as a disciplined reasoning layer than as a proven quantitative trading model.

  根拠https://github.com/sosopop/jev_stock/blob/main/jev_hk_predict.py#L29

  ラベルcall-site-verified

• SoundBlaster/Jev4Mellea

  README states: A small Python adapter that brings TypeSafe Jev's semantic checks into

  根拠https://github.com/SoundBlaster/Jev4Mellea/blob/main/src/mellea_jev/providers/categorical.py#L45

  ラベルcall-site-verified

• SREGym/SREGym

  README states: Jev is disabled by default. To enable it for Codex, set TYPESAFE_API_KEY and run:

  根拠https://github.com/SREGym/SREGym/blob/main/clients/jev/client.py#L14

  ラベルcall-site-verified

• sriganesh/jevibe-check

  README states: Usage shows request counts and estimated cost. Set token prices under Cost estimate, or use Clear stats to reset the counters. Your key and settings are kept. Check TypeSafe for actual charges.

  根拠https://github.com/sriganesh/jevibe-check/blob/main/extension/background.js#L133

  ラベルcall-site-verified

• steven-shoemaker/hunch

  README states: To reach Jev through a gateway your company already uses, pass gateway="openrouter" (reads OPENROUTER_API_KEY) or gateway="vercel" (reads AI_GATEWAY_API_KEY) to configure. Everything else works the same.

  根拠https://github.com/steven-shoemaker/hunch/blob/main/src/hunch/client.py#L37

  ラベルcall-site-verified

• steven-shoemaker/hunch-js

  README states: To reach Jev through a gateway your company already uses, pass gateway: "openrouter" (reads OPENROUTER_API_KEY) or gateway: "vercel" (reads AI_GATEWAY_API_KEY) to configure. Everything else works the same.

  根拠https://github.com/steven-shoemaker/hunch-js/blob/main/src/client.ts#L1

  ラベルcall-site-verified

• STRML/omp-jevens-classifier

  README states: Jev-judged permission checks for OMP's bash tool, and for eval payloads that spawn

  根拠https://github.com/STRML/omp-jevens-classifier/blob/main/jev.ts#L64

  ラベルcall-site-verified

• sufianetaouil/every

  README states: every needs a Jev API key. If you don't have one, join the hosted-tier waitlist:

  根拠https://github.com/sufianetaouil/every/blob/main/every/judge.py#L16

  ラベルcall-site-verified

• sumleo/prompt2jev

  README states: Natural language, an LLM prompt, or the code that runs one in. A Jev decision out: typed questions and a runnable script.

  根拠https://github.com/sumleo/prompt2jev/blob/main/skills/prompt2jev/scripts/prompt2jev.py#L27

  ラベルcall-site-verified

• superagents-lab/jev-search

  README states: Search the web in plain language. A decision model such as TypeSafe's Jev, Cloudflare's Clef or OpenAI's GPT-6 Luna chooses sources, time ranges and search terms, then ranks the results returned through Search1API. You g

  根拠https://github.com/superagents-lab/jev-search/blob/main/README.md#L5

  ラベルreadme-claim

• supercorp-ai/supercov

  README states: export TYPESAFE_API_KEY=... # get one at https://typesafe.ai

  根拠https://github.com/supercorp-ai/supercov/blob/main/crates/supercov-cli/src/quality/tests.rs#L2768

  ラベルcall-site-verified

• szczyglis-dev/py-gpt

  README states: Jev / System One (inline) - integrates TypeSafe AI Jev for fast, typed semantic decisions over structured state, including classification, routing, selection, verification, and scoring.

  根拠https://github.com/szczyglis-dev/py-gpt/blob/master/src/pygpt_net/plugin/jev/worker.py#L106

  ラベルcall-site-verified

• tacticocc/Jevbridge

  README states: ACP and MCP adapter that bridges TypeSafe Jev with any LLM.

  根拠https://github.com/tacticocc/Jevbridge/blob/main/src/types.ts#L111

  ラベルcall-site-verified

• tamnd/kime

  README states: The base URL defaults to http://127.0.0.1:8000 in place of https://api.typesafe.ai.

  根拠https://github.com/tamnd/kime/blob/main/ts/README.md#L22

  ラベルreadme-claim

• tanghaojin/x-smart-tags

  README states: 扩展通过 Background Service Worker 直接调用 https://api.typesafe.ai/v1/systemone。

  根拠https://github.com/tanghaojin/x-smart-tags/blob/main/README.md#L42

  ラベルreadme-claim

• tanxarx/awesome-jev

  README states: A curated list of resources, open-source clones, integrations, and engineering playbooks for Jev — TypeSafe AI's non-generative "System 1" decision model.

  根拠https://github.com/tanxarx/awesome-jev/blob/main/README.md#L3

  ラベルreadme-claim

• TauricResearch/TradingAgents

  README states: export TYPESAFE_API_KEY=... # Jev social-post screening (optional)

  根拠https://github.com/TauricResearch/TradingAgents/blob/main/tradingagents/agents/post_screen.py#L68

  ラベルcall-site-verified

• tbdavid2019/ai-hedge-fund-API

  README states: Fallback Tier-3: TypeSafe Official Service (https://api.typesafe.ai/v1/systemone)

  根拠https://github.com/tbdavid2019/ai-hedge-fund-API/blob/main/README.md#L57

  ラベルreadme-claim

• The-Vibe-Company/quivr

  README states: That is far under Jev's budget of 32k tokens for the state plus the longest

  根拠https://github.com/The-Vibe-Company/quivr/blob/main/plugins/alerts/README.md#L292

  ラベルreadme-claim

• TholeG/typesafe-chess

  README states: Two virtual chess players, both powered by TypeSafe's System One model Jev.

  根拠https://github.com/TholeG/typesafe-chess/blob/main/README.md#L3

  ラベルreadme-claim

• TianyuCodings/JevHarness

  README states: | docs/ | Task authoring, a recorded Jev call, comparison video, and website screenshots |

  根拠https://github.com/TianyuCodings/JevHarness/blob/main/README.md#L212

  ラベルreadme-claim

• timpratim/macbrow

  README states: LiveKit Agents for the voice loop, TypeSafe's Jev

  根拠https://github.com/timpratim/macbrow/blob/main/README.md#L26

  ラベルreadme-claim

• tinyhumansai/opencompany

  README states: asked of Jev — TypeSafe's System One routing model — through the TinyHumans

  根拠https://github.com/tinyhumansai/opencompany/blob/main/README.md#L114

  ラベルreadme-claim

• TKY-27/JevSlop

  README states: Eight atomic detail Score questions plus a Jev overall Score and choice in one systemOne request.

  根拠https://github.com/TKY-27/JevSlop/blob/main/functions/api/evaluate.ts#L1

  ラベルcall-site-verified

• togethercomputer/tev1

  README states: Jev-inspired implementation and does not use Jev's answers as training labels.

  根拠https://github.com/togethercomputer/tev1/blob/main/README.md#L102

  ラベルreadme-claim

• TokenTrim/jev-routing-experiment

  README states: Jev is not a text model. It returns typed decisions (Choice / Score / Noul) with calibrated

  根拠https://github.com/TokenTrim/jev-routing-experiment/blob/main/jev_router/jev.py#L17

  ラベルcall-site-verified

• tonone-ai/tonone

  README states: | JEV_API_KEY or TYPESAFE_API_KEY | https://api.typesafe.ai/v1/systemone | jev-latest | jev |

  根拠https://github.com/tonone-ai/tonone/blob/main/lib/jev/README.md#L220

  ラベルreadme-claim

• tonsky/tgadmin

  README states: With Jev

  根拠https://github.com/tonsky/tgadmin/blob/main/README.md#L8

  ラベルreadme-claim

• tpaulshippy/shady-town

  README states: Decisions and thresholds stay in code. Built on TypeSafe's Jev via

  根拠https://github.com/tpaulshippy/shady-town/blob/main/README.md#L88

  ラベルreadme-claim

• trancong12102/agentskills

  README states: | answer audit (optional) | JEV_API_KEY | Create a key at console.typesafe.ai/keys, then export JEV_API_KEY=… (TYPESAFE_API_KEY also works). Turns on Jev's two extra checks in scripts/audit-answers.py. |

  根拠https://github.com/trancong12102/agentskills/blob/main/README.md#L38

  ラベルreadme-claim

• trpc-group/trpc-agent-go

  README states: For Jev, set JEV_BASE_URL=https://api.typesafe.ai, configure JEV_API_KEY

  根拠https://github.com/trpc-group/trpc-agent-go/blob/main/internal/systemone/README.md#L63

  ラベルreadme-claim

• truespar/sentio

  README states: provider = "jev"

  根拠https://github.com/truespar/sentio/blob/main/README.md#L998

  ラベルreadme-claim

• turenlabs/turenos

  README states: The detector uses one request with Noul questions for danger and confirmation, a Score for risk, and a Choice for category. Set TYPESAFE_ENDPOINT only if using a compatible HTTPS endpoint; plain HTTP is accepted only for

  根拠https://github.com/turenlabs/turenos/blob/main/command-guard/README.md#L41

  ラベルreadme-claim

• tyler-dot-earth/patdown

  README states: The judge is swappable. The default backend uses Effect Decision / DecisionModel with TypeSafe; rules and CLI commands use a provider-neutral interface.

  根拠https://github.com/tyler-dot-earth/patdown/blob/main/scripts/probe-jev-input-size.mjs#L25

  ラベルcall-site-verified

• tylergibbs1/sift

  README states: Load dist/ as an unpacked extension in chrome://extensions, then enter your TypeSafe API key on

  根拠https://github.com/tylergibbs1/sift/blob/main/scripts/smoke.ts#L74

  ラベルcall-site-verified

• tylerjharden/ailerix

  README states: POST /api/v1/systemone — Jev-compatible Choice / Score / Noul API

  根拠https://github.com/tylerjharden/ailerix/blob/main/src/components/playground-client.tsx#L45

  ラベルcall-site-verified

• typakon4/jev-layer

  README states:

  根拠https://github.com/typakon4/jev-layer/blob/main/src/providers/demo.mjs#L36

  ラベルcall-site-verified

• TypeLLM/TypeLLM

  README states: TypeLLM brings type-safe generation to existing autoregressive LLMs without changing their architecture or weights. Inspired by TypeSafe AI's Jev, it lets models retain their native thinking and free-form generation whil

  根拠https://github.com/TypeLLM/TypeLLM/blob/main/README.md#L48

  ラベルreadme-claim

• TypeSafeAI/typesafe-playground

  README states: Small experiments. Clear decisions. This is an independent community project, not an official TypeSafe AI product, production agent harness, or claim that a typed answer is correct.

  根拠https://github.com/TypeSafeAI/typesafe-playground/blob/main/lib/serverJev.ts#L52

  ラベルcall-site-verified

• UditAkhourii/quicksilver

  README states: Jev, TypeSafe's System One model. Jev returns typed

  根拠https://github.com/UditAkhourii/quicksilver/blob/main/README.md#L21

  ラベルreadme-claim

• uehaj/sys1grep

  README states: sys1grep: endpoint api.typesafe.ai/v1/systemone (default), model jev-latest (default)

  根拠https://github.com/uehaj/sys1grep/blob/main/README.md#L229

  ラベルreadme-claim

• uezo/aiavatarkit

  README states: Gates — Smart Turn, Namo Turn, Jev, OpenAI Decisions, LLM turn gate, filler-only, session hold, custom gates

  根拠https://github.com/uezo/aiavatarkit/blob/main/aiavatar/sts/vad/turn_end_gates/jev.py#L172

  ラベルcall-site-verified

• unownone/jevsume

  README states: ATS-friendly resume review powered by Jev (TypeSafe System One). The frontend extracts resume text the way a parser would, then a Cloudflare Worker runs typed JEV questions and composes a JevScore.

  根拠https://github.com/unownone/jevsume/blob/main/readme.md#L3

  ラベルreadme-claim

• uu889/laya-opencv

  README states: | TypeSafe 官方 | https://api.typesafe.ai | TypeSafe 控制台的密钥 |

  根拠https://github.com/uu889/laya-opencv/blob/main/README.md#L386

  ラベルreadme-claim

• v-modal/awesome-jev-tools

  README states: A curated list of public projects and developer patterns built on Jev, TypeSafe AI's System One model for typed decisions.

  根拠https://github.com/v-modal/awesome-jev-tools/blob/main/README.md#L7

  ラベルreadme-claim

• valentynkit/awesome-jev-typesafe

  README states:

  根拠https://github.com/valentynkit/awesome-jev-typesafe/blob/main/scripts/jev-client.mjs#L5

  ラベルcall-site-verified

• valentynkit/jev-commit

  README states: that the two agree. This does, in one call to Jev, before the commit lands.

  根拠https://github.com/valentynkit/jev-commit/blob/main/jev_commit/jev.py#L18

  ラベルcall-site-verified

• valentynkit/jev-skip

  README states: Gateway shim rather than the direct API. Re-measure on api.typesafe.ai before pinning

  根拠https://github.com/valentynkit/jev-skip/blob/main/README.md#L19

  ラベルreadme-claim

• vinilana/live-jev

  README states: TypeSafe's Jev (a "System One" decision model)

  根拠https://github.com/vinilana/live-jev/blob/main/README.md#L4

  ラベルreadme-claim

• vnmoorthy/siege

  README states: Built at CoreWeave Hacks: Agent Loops (Weights & Biases, AGI House, TypeSafe AI), San Francisco, September 13 to 14, 2026.

  根拠https://github.com/vnmoorthy/siege/blob/main/README.md#L248

  ラベルreadme-claim

• w3cj/jev-chat

  README states: Jev does not produce text. Every text argument, such as a search query, a task title or a city, is

  根拠https://github.com/w3cj/jev-chat/blob/main/apps/server/src/jev/client.ts#L1

  ラベルcall-site-verified

• walidboulanouar/awesome-jev-use-cases

  README states:

  根拠https://github.com/walidboulanouar/awesome-jev-use-cases/blob/main/README.md#L2

  ラベルreadme-claim

• waynesutton/ask-jev-ai

  README states: | Judge | Jev, TypeSafe System One, through the Convex AI Gateway Decisions endpoint (typesafe/jev-1.13) on a deployment token. No key. TypeSafe direct is the fallback when a key is set. | Decisions, docs.typesafe.ai |

  根拠https://github.com/waynesutton/ask-jev-ai/blob/main/README.md#L40

  ラベルreadme-claim

• WeSecureYou/Jev-test

  README states: The project works immediately without credentials. When JEV_API_KEY is absent, it logs a warning and uses deterministic, occupation-aware mock data with the same validated shape as a live Jev response.

  根拠https://github.com/WeSecureYou/Jev-test/blob/main/src/services/jevClient.ts#L9

  ラベルcall-site-verified

• wfzyx/von

  README states: von serve exposes /v1/systemone, byte-compatible with the TypeSafe specification. Anything that talks to Jev talks to Von by changing the base URL; the SDKs do the same via VON_BASE_URL.

  根拠https://github.com/wfzyx/von/blob/master/js/src/client.ts#L10

  ラベルcall-site-verified

• whyashthakker/awesome-jev-use-cases

  README states: git clone https://github.com/whyashthakker/awesome-jev-use-cases.git

  根拠https://github.com/whyashthakker/awesome-jev-use-cases/blob/main/shared/providers.js#L16

  ラベルcall-site-verified

• Wizhill05/typesafe-image-diffusion

  README states: Jev (via TypeSafe System One) is a judgment model. You give it state, ask multiple choice questions, and it picks answers with confidence scores. It classifies things. It does not draw things.

  根拠https://github.com/Wizhill05/typesafe-image-diffusion/blob/main/fix_synthesizer.py#L3

  ラベルcall-site-verified

• wobsoriano/is-jeven

  README states: Uses TypeSafe AI's Jev model under the hood to determine if a number is even.

  根拠https://github.com/wobsoriano/is-jeven/blob/main/index.js#L19

  ラベルcall-site-verified

• wustep/jev-playground

  README states: npm test # vitest: renderer invariants, Jev request shapes, proxy, MIDI round-trip, inbox/match stubs

  根拠https://github.com/wustep/jev-playground/blob/main/server/jevHandler.ts#L20

  ラベルcall-site-verified

• Xiaooolong/vev

  README states: the official Jev API (shape reference, costs a few cents)

  根拠https://github.com/Xiaooolong/vev/blob/main/conformance/README.md#L17

  ラベルreadme-claim

• xinyao27/jevonian

  README states: Jev picks the route for jevonian/auto — model and thinking depth in a single call.

  根拠https://github.com/xinyao27/jevonian/blob/main/internal/brain/types.go#L45

  ラベルcall-site-verified

• xiwan/acp-bridge

  README states: | Auto agent routing | Send a task to the virtual agent auto and let Jev (TypeSafe System One) pick the best-fit agent with a calibrated confidence score; falls back safely |

  根拠https://github.com/xiwan/acp-bridge/blob/main/README.md#L48

  ラベルreadme-claim

• y0usaf/pi-jev

  README states: The gate sends the working directory, the tool name, the last user message (first 1200 characters), and the tool's arguments to api.typesafe.ai. For write and edit those arguments contain file content. The output judge s

  根拠https://github.com/y0usaf/pi-jev/blob/main/README.md#L145

  ラベルreadme-claim

• yairshy/decido

  README states: ordinary Python. Start with Jev or supply your own decision provider.

  根拠https://github.com/yairshy/decido/blob/main/src/decido/providers/typesafe.py#L111

  ラベルcall-site-verified

• yan-labs/yan-skills

  README states: | jev | Typesafe(JEV / System One) | https://api.typesafe.ai/v1/systemone,Authorization: Bearer 鉴权(protocol: "typesafe-systemone",不是 Anthropic Messages/OpenAI 协议) | ⚠️ 不支持 run/run-many——它是结构化决策 API,不生成文本、不支持多轮工具调用,只能用下面「

  根拠https://github.com/yan-labs/yan-skills/blob/main/agent-fleet/README.md#L123

  ラベルreadme-claim

• Yifan-Lan/awesome-jev-robustness

  README states: entries · jev-1.13.0 in 111, jev-1.14.0 in 1, other decision models (Laya, Kev, Clef, Decision-1, OpenAI Decisions, open Jev-style checkpoints) in 31 · snapshot 2026-10-25 · numbers are the authors' own

  根拠https://github.com/Yifan-Lan/awesome-jev-robustness/blob/main/README.md#L5

  ラベルreadme-claim

• Yinsongxu/LLM2Jev

  README states: Turn your LLM / MLLM into a Jev-style decision model. Zero training required.

  根拠https://github.com/Yinsongxu/LLM2Jev/blob/main/demos/web/server.py#L19

  ラベルcall-site-verified

• ZeroGold/call-coach-ai

  README states: You need a TypeSafe API key. There are three ways to run Call Coach:

  根拠https://github.com/ZeroGold/call-coach-ai/blob/main/lib/coach.mjs#L13

  ラベルcall-site-verified

• ZeroTang05/VideoAdGuard

  README states: | TypeSafe | https://api.typesafe.ai/v1 | jev-latest | 官方接口 |

  根拠https://github.com/ZeroTang05/VideoAdGuard/blob/main/README.md#L138

  ラベルreadme-claim

• zhengxuyu/litjev

  README states: A reproduction of Jev: turn any Qwen model into a fast decision model.

  根拠https://github.com/zhengxuyu/litjev/blob/main/scripts/dump_prompt.py#L10

  ラベルcall-site-verified

• zhuyansen/jev-search-rerank-eval

  README states: Does a TypeSafe Jev rerank beat embedding search? Measured, with the judge-bias removed.

  根拠https://github.com/zhuyansen/jev-search-rerank-eval/blob/main/README.md#L3

  ラベルreadme-claim

• zsavage8/padflow-jev-evals

  README states: them. It exists so a decision model (TypeSafe Jev, or anything else) can be

  根拠https://github.com/zsavage8/padflow-jev-evals/blob/main/scripts/run_baseline.py#L7

  ラベルcall-site-verified