Directory · #109 trending MCP repo
Awesome Jev Live
wh000wh000/awesome-jev-live
Awesome Jev — evidence-graded index of TypeSafe System One: SDKs, MCP tools, agents, apps and open models. 20 languages, rebuilt every 2 hours.
- Stars
- 378
- This week
- —
- Forks
- 15
- Open issues
- 6
- Latest release
v2026.09.21
Documentation
README
Awesome Jev Live
The evidence-graded Jev index that rebuilds itself every two hours.
English · 简体中文 · 繁體中文 · 日本語 · 한국어 · Español · Français · Deutsch · Português (Brasil) · Русский · Italiano · العربية · हिन्दी · Türkçe · Tiếng Việt · ไทย · Bahasa Indonesia · Polski · Nederlands · Українська
[!NOTE] Live index · Last sync:
2026-11-07T16:50:18+08:00(UTC+8) · Entries: 714 · Added in the latest update: 52 · Implementation languages: 25
Every entry below was collected, filtered and re-checked by the pipeline in this repository. Numbers and timestamps come from the sources, not from a hand-written snapshot.
🔥 Picks of the moment — the top of each category
One entry per category, ranked by evidence grade and stars, recomputed on every update. A ranking, not an endorsement; every pick links through to its full card below. Projects that published a screenshot or a recording are preferred, so the strip stays visual.
| 🏛️ typesafe-ai/skills ⭐2883 · ✅ official Agent skills for building with TypeSafe's System One API |
🧰 john-rocky/coreai-kit
⭐123 · Swift · 👁️ observed
Swift SDK for running chat, vision and speech models on iPhone and Mac with Apple's Core AI. Model download and caching, FoundationModels integration, and runnable…
|
🤖 Bodila51/grok-bot-jev
⭐132 · Python · 👁️ observed
Connect TypeSafe Jev to Grok Bot as a cheap decision layer - usage gates, skill template, examples
|
🛡️ dharmeshgurnani/CodeOtter
⭐3 · JavaScript · 👁️ observed
Autonomous, self-hosted AI code reviews, calibrated scoring gauges, and merge gates from any open model.
|
🧪 Mapika/decider
⭐1056 · Python · 👁️ observed
A family of System One-style models fine-tuned from Qwen3.5, designed for one-pass typed decisions with calibrated probabilities.
|
🔬 r-ms/mini-jev
⭐62 · Python · 🔎 inferred
mini-Jev: what a Jev-style typed-decision interface looks like on a frozen Qwen3-4B — read the option letter's logits instead of generating JSON. Preregistered…
|
🎮 superagents-lab/jev-search
⭐556 · TypeScript · 👁️ observed
Search the web with TypeSafe's Jev: source selection, query understanding and relevance ranking. Built with Search1API.
|
📰 9 known defects (jaggedness) ⭐0 · ✅ official 字面理解、**计数不可靠**、日期比较、多级间接、无关细节大 state、对抗内容、指令与 criteria 矛盾、结构性不变量、生成。官方建议"拆成原子问题在代码里组合"。 |
Contents
- What is Jev
- How entries are graded
- Official SDKs and developer tools — 7
- Community clients, SDKs and adapters — 126
- Agent tooling: MCP, hooks, gates and coding agents — 209
- Routing, guardrails and approvals — 60
- Evaluation, calibration and benchmarks — 84
- Open reproductions, weights and architecture research — 45
- Applications, games, robotics and interactive demos — 46
- Writing, discussions and sibling lists — 83
- Other projects — 54
- Projects by implementation language
What is Jev
Jev is TypeSafe AI's first System One model. It does not write prose. It takes a state plus questions whose answer space you define in advance, and returns typed values with probability distributions your code can branch on.
| Shape | state + typed questions → constrained answers + probabilities → your code |
| Primitives | Choice (pick one of ≤255 options), Score (a rubric 2–10), Noul (a probabilistic yes/no) |
| Endpoint | POST https://api.typesafe.ai/v1/systemone, model jev-1.13.0 / alias jev-latest |
| Good fits | routing, triage, scoring, moderation, verification and low-latency gates inside a bounded workflow |
| Known limits | counting is unreliable, multi-level indirection is weak, and official material names nine classes of jaggedness. Schema-valid output is not the same as a correct decision — calibrate on your own data. |
How entries are graded
Most lists in this space assert inclusion. This one says how much it actually verified, then lets you filter accordingly.
| Grade | What it means |
|---|---|
official |
Published by TypeSafe AI itself. |
observed |
A public artifact that can be opened and read — real source, real configuration, or an explicit TypeSafe/Jev declaration in the repository name or topic. |
inferred |
Matched on an ambiguous signal plus corroborating vocabulary, but not yet read line by line. |
unverified |
Looks related, nothing independently confirmed. Listed for discovery only. |
Official SDKs and developer tools
Everything published by TypeSafe itself. Start here.
🏛️ typesafe-ai/skills · ⭐2883 · ✅ official · 14d
📝 Summary
Agent skills for building with TypeSafe's System One API
💡 The vendor's own agent skills. Because it is updated continuously, it is the closest thing to a specification of how TypeSafe intends Jev to be driven from an agent.
📌 Basic facts
| Field | Value |
|---|---|
| Category | Official SDKs and developer tools |
| Evidence | official |
📊 Data
| Metric | Value |
|---|---|
| Stars | 2883 |
| Last push | 2026-10-23 |
| First listed | 2026-09-18 |
🏛️ typesafe-ai/system-one-adapter-python · ⭐407 · Python · ✅ official · 45d
📝 Summary
Drop-in TypeSafeClient replacement backed by LLM APIs
💡 Drop-in replacement that backs the same interface with an ordinary LLM provider. This is the honest way to A/B a typed decision against a prompt, on your own data, before committing to either.
🔧 Found used in code: README.md
📌 Basic facts
| Field | Value |
|---|---|
| Category | Official SDKs and developer tools |
| Evidence | official |
| Language | Python |
📊 Data
| Metric | Value |
|---|---|
| Stars | 407 |
| Last push | 2026-09-22 |
| First listed | 2026-10-23 |
🏛️ typesafe-ai/typesafe-sdk-python · ⭐309 · Python · ✅ official · 23d
📝 Summary
The official Python library for the TypeSafe API
💡 Synchronous and asynchronous clients. The fastest path from an API key to a typed decision, and the reference the community clients are compared against.
🔧 Found used in code: README.md
📌 Basic facts
| Field | Value |
|---|---|
| Category | Official SDKs and developer tools |
| Evidence | official |
| Language | Python |
📊 Data
| Metric | Value |
|---|---|
| Stars | 309 |
| Last push | 2026-10-14 |
| First listed | 2026-10-23 |
🏛️ typesafe-ai/typesafe-sdk-js · ⭐284 · TypeScript · ✅ official · 23d
📝 Summary
The official TypeScript/JavaScript library for the TypeSafe API
💡 TypeScript client where the answer type is inferred from the question you asked, so a mismatched return type is a compile error rather than a runtime surprise.
📌 Basic facts
| Field | Value |
|---|---|
| Category | Official SDKs and developer tools |
| Evidence | official |
| Language | TypeScript |
📊 Data
| Metric | Value |
|---|---|
| Stars | 284 |
| Last push | 2026-10-14 |
| First listed | 2026-10-23 |
🏛️ TypeSafeAI/jev-harness · ⭐38 · TypeScript · ✅ official · 4d
📝 Summary
A custom coding harness for TypeSafe AI's Jev: an LLM proposes, Jev answers narrow questions, code decides, every step leaves a receipt.
📌 Basic facts
| Field | Value |
|---|---|
| Category | Official SDKs and developer tools |
| Evidence | official |
| Language | TypeScript |
📊 Data
| Metric | Value |
|---|---|
| Stars | 38 |
| Last push | 2026-11-03 |
| First listed | 2026-11-03 |
🏷 agent-harness · agentic-coding · ai-agents · ai-governance · ai-safety · code-review · coding-agent · developer-tools
🏛️ typesafe-ai/WorkflowEvals · ⭐21 · Python · ✅ official · 38d
📝 Summary
evals.typesafe.ai workflow code published
📌 Basic facts
| Field | Value |
|---|---|
| Category | Official SDKs and developer tools |
| Evidence | official |
| Language | Python |
📊 Data
| Metric | Value |
|---|---|
| Stars | 21 |
| Last push | 2026-09-29 |
| First listed | 2026-10-23 |
🏛️ typesafe-ai/typesafe-ai.github.io · ⭐2 · HTML · ✅ official · 155d
📝 Summary
No upstream description was published.
📌 Basic facts
| Field | Value |
|---|---|
| Category | Official SDKs and developer tools |
| Evidence | official |
| Language | HTML |
📊 Data
| Metric | Value |
|---|---|
| Stars | 2 |
| Last push | 2026-06-04 |
| First listed | 2026-10-23 |
Community clients, SDKs and adapters
Typed clients for the System One endpoint, across as many languages as the community has got round to.
🧰 jexp/neo4jev · ⭐182 · Jupyter · 👁️ observed · 49d · ⭐+1
📝 Summary
Typesafe.ai System One Model Jev navigating a Neo4j graph by using a classifier over neighbouring relationships
📌 Basic facts
| Field | Value |
|---|---|
| Category | Community clients, SDKs and adapters |
| Evidence | observed |
| Language | Jupyter |
📊 Data
| Metric | Value |
|---|---|
| Stars | 182 (+1) |
| Last push | 2026-09-18 |
| First listed | 2026-10-28 |
🧰 keltokhy/jgrep · ⭐140 · Python · 👁️ observed · 20d
📝 Summary
grep, but the pattern is a description. Filters lines by meaning with TypeSafe's Jev decision model: ~200 ms and a thousandth of a cent per line.
🔧 Found used in code: README.md
📌 Basic facts
| Field | Value |
|---|---|
| Category | Community clients, SDKs and adapters |
| Evidence | observed |
| Language | Python |
📊 Data
| Metric | Value |
|---|---|
| Stars | 140 |
| Last push | 2026-10-17 |
| First listed | 2026-10-29 |
🏷 cli · grep · jev · openrouter · semantic-search · typesafe
🧰 qkal/Canny · ⭐134 · TypeScript · 👁️ observed · 45d
📝 Summary
Stops AI coding agents from claiming work is done without evidence. Deterministic hooks decide, TypeSafe's Jev advises. Append-only ledger, zero runtime dependencies.
🔧 Found used in code: README.md
📌 Basic facts
| Field | Value |
|---|---|
| Category | Community clients, SDKs and adapters |
| Evidence | observed |
| Language | TypeScript |
📊 Data
| Metric | Value |
|---|---|
| Stars | 134 |
| Last push | 2026-09-22 |
| First listed | 2026-10-29 |
🏷 agent-hooks · ai-agents · claude-code · claude-code-hooks · codex · codex-cli · guardrails · jev
🧰 john-rocky/coreai-kit · ⭐123 · Swift · 👁️ observed · 2d
📝 Summary
Swift SDK for running chat, vision and speech models on iPhone and Mac with Apple's Core AI. Model download and caching, FoundationModels integration, and runnable examples with documented OS, SDK and model requirements.
🔧 Found used in code: Sources/CoreAIOps/CoreAI+SystemOne.swift
📌 Basic facts
| Field | Value |
|---|---|
| Category | Community clients, SDKs and adapters |
| Evidence | observed |
| Language | Swift |
📊 Data
| Metric | Value |
|---|---|
| Stars | 123 |
| Last push | 2026-11-05 |
| First listed | 2026-10-26 |
🏷 apple · apple-core-ai · asr · core-ai · coreai · foundation-models · ios · ios-27
| 🖼 Image | 🎬 Video |
|---|---|
![]() |
![]() animated recording |
🧰 yzfly/awesome-jev-zh · ⭐90 · HTML · 👁️ observed · 0d · ⭐+1
📝 Summary
Jev / TypeSafe System One 中文精选列表:官方资料、SDK、爆款应用、Agent 工具、开源复现与独立评测,附中文上手指南,每日自动收录 GitHub 热门项目。
🔧 Found used in code: docs/quickstart.md
📌 Basic facts
| Field | Value |
|---|---|
| Category | Community clients, SDKs and adapters |
| Evidence | observed |
| Language | HTML |
📊 Data
| Metric | Value |
|---|---|
| Stars | 90 (+1) |
| Last push | 2026-11-07 |
| First listed | 2026-09-18 |
🏷 ai-agent · awesome · awesome-list · chinese · jev · llm · structured-output · system-one
🧰 obie/ruby_decision_model · ⭐58 · Ruby · 👁️ observed · 25d
📝 Summary
Ruby client for decision models such as Typesafe Jev
🔧 Found used in code: README.md
📌 Basic facts
| Field | Value |
|---|---|
| Category | Community clients, SDKs and adapters |
| Evidence | observed |
| Language | Ruby |
📊 Data
| Metric | Value |
|---|---|
| Stars | 58 |
| Last push | 2026-10-12 |
| First listed | 2026-10-29 |
🧰 bodepudimuneendra-netizen/ConsciousDB · ⭐54 · Python · 👁️ observed · 4d
📝 Summary
A database-agnostic Agentic GraphRAG framework using swappable System One models (local Laya / cloud Jev). A plug-and-play intelligence layer featuring a complete 4-phase pipeline, continuous evaluation and custom A* traversal for any graph database.
📌 Basic facts
| Field | Value |
|---|---|
| Category | Community clients, SDKs and adapters |
| Evidence | observed |
| Language | Python |
📊 Data
| Metric | Value |
|---|---|
| Stars | 54 |
| Last push | 2026-11-02 |
| First listed | 2026-11-02 |
🏷 astar-algorithm · eval-harness · fine-tuning · graphdatabase · graprag · jev · laya · neo4j
🧰 MrJev/awesome-jev · ⭐19 · Python · 👁️ observed · 0d
📝 Summary
A curated list of projects, integrations, and resources for Jev, TypeSafe AI's System One model.
📌 Basic facts
| Field | Value |
|---|---|
| Category | Community clients, SDKs and adapters |
| Evidence | observed |
| Language | Python |
📊 Data
| Metric | Value |
|---|---|
| Stars | 19 |
| Last push | 2026-11-06 |
| First listed | 2026-09-18 |
🏷 ai-agents · awesome · awesome-list · jev · mcp · typesafe
🧰 ckaraca/awesome-jev · ⭐17 · Python · 👁️ observed · 4d
📝 Summary
A curated list of tools, integrations, and experiments built on Jev, TypeSafe AI's System One model for fast, typed decisions.
📌 Basic facts
| Field | Value |
|---|---|
| Category | Community clients, SDKs and adapters |
| Evidence | observed |
| Language | Python |
📊 Data
| Metric | Value |
|---|---|
| Stars | 17 |
| Last push | 2026-11-02 |
| First listed | 2026-09-19 |
🏷 ai-agents · awesome · awesome-list · computer-use · jev · llm · system-one · typesafe
🧰 Premo-Cloud/typesafe-sdk-java · ⭐12 · Java · 👁️ observed · 16d
📝 Summary
Community Java SDK for Jev, TypeSafe's System One model: typed questions in, typed answers with calibrated probabilities out. Java 17+, Spring Boot starter (unofficial)
📌 Basic facts
| Field | Value |
|---|---|
| Category | Community clients, SDKs and adapters |
| Evidence | observed |
| Language | Java |
📊 Data
| Metric | Value |
|---|---|
| Stars | 12 |
| Last push | 2026-10-21 |
| First listed | 2026-10-28 |
🏷 java · jev · sdk · spring-boot · system-one · system-one-models · typesafe
🧰 PromtEngineer/jev-harness · ⭐9 · Python · 👁️ observed · 43d
📝 Summary
A Pi agent harness built around TypeSafe's Jev (System One model): router, context picker, gate, verifier. Tested on Neon Postgres branches.
📌 Basic facts
| Field | Value |
|---|---|
| Category | Community clients, SDKs and adapters |
| Evidence | observed |
| Language | Python |
📊 Data
| Metric | Value |
|---|---|
| Stars | 9 |
| Last push | 2026-09-24 |
| First listed | 2026-10-28 |
🧰 sumleo/prompt2jev · ⭐9 · Python · 👁️ observed · 46d · **NEW**
📝 Summary
Agent skill and CLI that turn natural language, an LLM prompt, or the code that runs one into a TypeSafe Jev decision: typed state, Choice/Score/Noul questions, and a runnable script
🔧 Found used in code: docs/sdk/python.md
📌 Basic facts
| Field | Value |
|---|---|
| Category | Community clients, SDKs and adapters |
| Evidence | observed |
| Language | Python |
📊 Data
| Metric | Value |
|---|---|
| Stars | 9 |
| Last push | 2026-09-21 |
| First listed | 2026-11-07 |
🏷 agent-skill · claude-code · decision-making · jev · llm · typesafe
| 🖼 Image | 🎬 Video |
|---|---|
| no media published | ![]() animated recording · Open video |
🧰 clouatre-labs/decisions-judge-mcp · ⭐5 · JavaScript · 👁️ observed · 5d
📝 Summary
Typed decisions for AI agents as an MCP tool: yes/no probability (noul), choice, and score in one fast request. Backed by the TypeSafe System One model.
📌 Basic facts
| Field | Value |
|---|---|
| Category | Community clients, SDKs and adapters |
| Evidence | observed |
| Language | JavaScript |
📊 Data
| Metric | Value |
|---|---|
| Stars | 5 |
| Last push | 2026-11-02 |
| First listed | 2026-10-12 |
🏷 agent · ai · cli · decision-making · decisions · judge · llm · mcp
| 🖼 Image | 🎬 Video |
|---|---|
![]() |
![]() animated recording |
🧰 elbruno/ElBruno.AI.Decisions · ⭐5 · C# · 👁️ observed · 3d
📝 Summary
.NET 10 decision SDK: Choose, Score, and Assess with Microsoft Foundry (Microsoft-Decision-1), native local Ollama System One models, and TypeSafe AI Jev. One API, probabilities instead of paragraphs.
📌 Basic facts
| Field | Value |
|---|---|
| Category | Community clients, SDKs and adapters |
| Evidence | observed |
| Language | C# |
📊 Data
| Metric | Value |
|---|---|
| Stars | 5 |
| Last push | 2026-11-03 |
| First listed | 2026-11-03 |
🏷 dotnet · dotnet10 · jev · microsoft-extensions-ai · nuget
| 🖼 Image | 🎬 Video |
|---|---|
![]() |
no media published |
🧰 avalonreset/legends-jev · ⭐3 · Python · 👁️ observed · 1d
📝 Summary
Optical character recognition (OCR), storefront signage verification, bounding-box grounding, and typed computer-use decision CLI powered by TypeSafe Jev 1.14.
📌 Basic facts
| Field | Value |
|---|---|
| Category | Community clients, SDKs and adapters |
| Evidence | observed |
| Language | Python |
📊 Data
| Metric | Value |
|---|---|
| Stars | 3 |
| Last push | 2026-11-05 |
| First listed | 2026-11-04 |
🏷 bounding-box · browser-agent · claude-code · codex · computer-use · cto-legends · decision-model · image-text-extraction
| 🖼 Image | 🎬 Video |
|---|---|
![]() |
no media published |
🧰 moeinheidari7829/Kev-Bio · ⭐3 · Python · 👁️ observed · 0d
📝 Summary
Official code for "Every Pixel Is an Option: A System One Model for Medical Ultrasound Biometry" (arXiv 2611.03360)
📌 Basic facts
| Field | Value |
|---|---|
| Category | Community clients, SDKs and adapters |
| Evidence | observed |
| Language | Python |
📊 Data
| Metric | Value |
|---|---|
| Stars | 3 |
| Last push | 2026-11-07 |
| First listed | 2026-11-07 |
🏷 biometry · echocardiography · jev · jev-ai · kev · medical-imaging · system-one · ultrasound
| 🖼 Image | 🎬 Video |
|---|---|
![]() |
no media published |
Asset hot-linked from the upstream repository because no redistribution licence was declared.
🧰 Bonzokoles/36_chambers · Python · 👁️ observed · 15d
📝 Summary
Federated knowledge retrieval engine for AI agents. Routes queries across specialized backends (FTS5, ChromaDB, graph) via a registry-first orchestrator with TypeSafe/Jev AI routing and evidence re-ranking. Read-only by default, security-gated by design.
📌 Basic facts
| Field | Value |
|---|---|
| Category | Community clients, SDKs and adapters |
| Evidence | observed |
| Language | Python |
📊 Data
| Metric | Value |
|---|---|
| Stars | 0 |
| Last push | 2026-10-22 |
| First listed | 2026-10-23 |
🏷 ai-agents · ai-infrastructure · chromadb · embeddings · federated-search · fts5 · information-retrieval · jev
| 🖼 Image | 🎬 Video |
|---|---|
![]() |
no media published |
Asset hot-linked from the upstream repository because no redistribution licence was declared.
🧰 socai-io/jev-social · ⭐202 · JavaScript · 🔎 inferred · 5d · ⭐+1
📝 Summary
Open-source, local-first social media research agent for Instagram, TikTok, and LinkedIn. Jev routes read-only steps; socai CLI captures cited browser evidence.
📌 Basic facts
| Field | Value |
|---|
…
Releases
🧰
🤖
🛡️
🧪
🔬
🎮 





