Browser & computer use
18 entries in this category. Each names the specific decision and links to openable evidence.
AndrewPrifer/jimothy
README states: Train small, insanely fast local classifiers from Jev-compatible examples. Run locally in your browser or Node.js.
Evidencehttps://github.com/AndrewPrifer/jimothy/blob/master/README.md#L3
Labelsreadme-claim
Bodila51/muse-jev-playbook
README states: Use TypeSafe AI's Jev as a cheap, fast decision layer inside an AI agent workflow — triage, classify, score, and gate work before expensive steps (browser, deep research, retries, subagents).
Evidencehttps://github.com/Bodila51/muse-jev-playbook/blob/main/README.md#L3
Labelsreadme-claim
browser-use/jev-ultrafast
README states:
Evidencehttps://github.com/browser-use/jev-ultrafast/blob/main/jev_ultrafast/model.py#L119
Labelscall-site-verified
droidrun/mobile-jev
README states: One goal. A real Android phone. Jev makes the decisions.
Evidencehttps://github.com/droidrun/mobile-jev/blob/main/scripts/mobile-agent/policy.mjs#L224
Labelscall-site-verified
ethanplusai/jev-chat-for-twitch
README states: Sign up at typesafe.ai and create an API key. The key stays in your browser profile. You pay TypeSafe directly for what you use.
Evidencehttps://github.com/ethanplusai/jev-chat-for-twitch/blob/main/README.md#L28
Labelsreadme-claim
filedcom/playjev
README states: Stagehand, but with Jev. Natural-language browser automation for Playwright, powered by Jev.
Evidencehttps://github.com/filedcom/playjev/blob/main/src/core/playjev.ts#L305
Labelscall-site-verified
forvela/jev-agent-browser
README states: Give a parent agent a bounded browser task. Jev chooses the next typed action. agent-browser executes it. Ambiguity, repetition, or a stuck state becomes a structured handoff back to the parent.
Evidencehttps://github.com/forvela/jev-agent-browser/blob/main/src/decision.js#L1
Labelscall-site-verified
idovmamane/dejevu
README states: Jev? Déjà vu. Browser agents that run on instinct, no Jev needed. One look at the page. One call to any open model. One action.
Evidencehttps://github.com/idovmamane/dejevu/blob/main/dejevu/policy.py#L209
Labelscall-site-verified
jkudish/jev-browser
README states:
Evidencehttps://github.com/jkudish/jev-browser/blob/main/src/questions.ts#L4
Labelscall-site-verified
Liyucheng1997/332_lab-jev-chat
README states: 采集:一个 App 一个适配器(app/src/main/java/com/jev/probe/capture/ChatAppAdapter.kt),服务按前台包名分发;适配器只负责把当前窗口变成「标题 + 消息列表(谁说的、说了什么)」,下游判断 / 悬浮窗 / 填入全部通用。微信适配器读聊天气泡(com.tencent.mm:id/bkl),按气泡位置判断谁说的;QQ 适配器读正文节点(com.tencent.mobileqq:id
Evidencehttps://github.com/Liyucheng1997/332_lab-jev-chat/blob/main/README.md#L50
Labelsreadme-claim
michaelswissa/jevry
README states: The historical beta.1 study ran three five-turn sessions per implementation, with the same fixture tasks and resolved Claude model. Jevry also used Jev for browser decisions.
Evidencehttps://github.com/michaelswissa/jevry/blob/main/desktop/engine.ts#L205
Labelscall-site-verified
ndolinschi/swarmrouter
README states: Pick which agent/skill handles a task (research / code / browser / support / writer) + confidence — visual swarm map powered by TypeSafe Jev.
Evidencehttps://github.com/ndolinschi/swarmrouter/blob/main/src/app/api/decide/route.ts#L7
Labelscall-site-verified
phyous/tsai-civ2
README states: An original Civilization II browser harness for TypeSafe Jev, with live decision
Evidencehttps://github.com/phyous/tsai-civ2/blob/main/civ2/session.py#L486
Labelscall-site-verified
pst2154/Nemotron_Jev
README states: Ask typed questions about text or JSON and inspect model-derived probability distributions in a browser. One container runs the model, the original Decision Lab explorer, and a TypeSafe-shaped API.
Evidencehttps://github.com/pst2154/Nemotron_Jev/blob/master/README.md#L3
Labelsreadme-claim
rokcso/bluenoise
README states: BlueNoise collects no telemetry or analytics. It does not read cookies, access account credentials, call X APIs, or take account actions such as blocking, muting, following, or posting. Settings and rules are stored thro
Evidencehttps://github.com/rokcso/bluenoise/blob/main/README.md#L34
Labelsreadme-claim
Sac-Y/Jev-cu
README states: 把 Computer Use 的「下一步点哪里」交给 Jev(TypeSafe System One):Jev 从界面文字候选中选元素、动作、完成度与风险,Codex Computer Use 负责读取界面与执行,本地策略门槛拦截敏感操作。只传文字,不传截图。
Evidencehttps://github.com/Sac-Y/Jev-cu/blob/main/scripts/jev-decide.mjs#L19
Labelscall-site-verified
yikangy873-gif/jev-desktop
README states: Jev Desktop adds a bounded decision loop to Codex Computer Use for browser tabs and native macOS apps.
Labelscall-site-verified
Ying-Kai-Liao/jev-browser
README states: jev-browser runs locally. It has no server of its own and collects no telemetry. To decide
Evidencehttps://github.com/Ying-Kai-Liao/jev-browser/blob/main/README.md#L362
Labelsreadme-claim
Keep going on this site
Try, docs, cases, and pricing link to each other. The next page is on this site.
- Site directoryThe full list of the pages above.
- EcosystemCommunity projects by category.
- Jev coverageEach entry names the decision and its evidence.
- Daily casesThe next batch of coverage cases, each day.
- CommunityPublic posts, each with its source.
- BuildsNotes on model and site updates.
- PrimitivesChoice, Noul, and Score together.
- PatternsRouting, fan-out, and composite scoring.
- SDKsHTTP, Python, and JavaScript.
Use clavue-jev
- Try a judgmentSend a state and read the answer.
- Live scenesTetris, six rapid judgments, and a sort. Copy the call and run it again.
- Judgment LabSupport triage, release review, compliance routing — edit and run on the spot.
- Compareclavue-jev and jev-1.13.0 answer the same question.
- DemosJudgment examples you can edit.
- PricingInput is $0.042 per million tokens. Output is free.
- APICall POST /v1/judge with a key.
Docs
- IntroductionWhat this model is for.
- Quick startThe smallest call that runs.
- System OneWritten here: a state plus one typed question. The model is clavue-jev.
- StateThe text you send in.
- ConfidenceHow sure the answer is, and when to act on it.
- ChoicePick one mutually exclusive option.
- NoulA yes/no question, returned as the probability of yes.
- ScoreRate along a scale.
- Intent routingOne Choice call dispatches the request.
- Confidence routingAct when confidence is high. Hand off when it is low.
- Fan-outAsk several questions of one state.
- Composite scoringFold several scores into one ranking.
- HTTP APIThe request shape without an SDK.
- Python SDKCall a judgment from Python.
- JavaScript SDKCall a judgment from JavaScript.
Cases
- Use-case mapWhich primitive people use for which job.
- GuardrailsA yes/no check before an action runs.
- Function callingPick which function to call.
- Classify with confidenceDo not act when the class is unsure.
- Semantic findFind the passages that match.
- Parallel questionsAsk the whole set in one call.
- Hierarchical classificationCoarse class first, then the fine one.
- Citation checkWhether a sentence is supported by the source.
- Entity alignmentWhether two names are the same thing.
- Skill suggestionChoose which skill fits the moment.