Jev directory
CommunityJev Integrations & Adapters
n8n Nodes TypeSafe AI — npmjs.com
An n8n community node for the TypeSafe AI System One (Jev) API. It adds Ask Yes/No, Ask Choice, Ask Score, Evaluate Questions and Get Many Models operations, so workflows can branch on calibrated probabilities instead of parsed free text.
- Reading time
- 2 min
- Published by
- Community
- Added
- Nov 2026
This is one of several Jev listings for the same project. The canonical entry is n8n Nodes TypeSafe AI — DomMonte.
Highlights
- Exposes Ask Yes/No, Ask Choice, Ask Score, Evaluate Questions and Get Many Models operations.
- Authenticates with a TypeSafe AI API credential sent as an Authorization: Bearer header.
- Defaults to the jev-latest model; versioned IDs such as jev-1.13.0 can be pinned for tuned thresholds.
- Evaluate Questions asks several questions about the same state in one request, the cheapest option.
- Self-hosted installs need an Owner or Admin role; n8n Cloud allows it only once n8n verifies the package.
Questions about this entry
Short answers drawn from the linked project or page.
On a self-hosted instance, go to Settings > Community Nodes, select Install, enter n8n-nodes-typesafe-ai and accept the community-node risk notice; the TypeSafe AI node then appears in the node panel. For queue mode or shell installs, create ~/.n8n/nodes, run npm i n8n-nodes-typesafe-ai inside it and restart n8n.
A TypeSafe AI API credential holding an API key from the TypeSafe dashboard. The key is sent as an Authorization: Bearer header, and the credential's Test button verifies it by calling GET /v1/models.
No, not yet. n8n Cloud users can install community nodes only after the package has been verified by n8n. Until then the package is available on self-hosted instances only.
Ask Yes/No returns the probability the answer is yes (a noul). Ask Choice picks one option and returns the full probability distribution plus a confidence value. Ask Score rates a state against ordered levels and returns a probability-weighted score that can land between levels. Evaluate Questions asks several questions about one state in a single request.
The Options > Model field defaults to jev-latest. You can pin a versioned ID such as jev-1.13.0 if you have tuned thresholds against that specific version.
Plain text, or a JSON object or array such as a chat log, a record or your app's current state. Expressions like {{ $json }} that resolve to an object are sent as JSON. The State Format option is Auto by default, falling back to text when the state does not parse as an object or array, and can be forced to Text or JSON.
Simplify is on by default and returns the simplified response, for example a noul, choice, probabilities, confidence or score plus model and usage. Turning it off returns the raw API response body instead.
Reactions & coverage
Posts, threads, and videos about this entry from around the web.
Facebook: Intake Router keeps diagnosis out of the answer space at the front desk.
FacebookClinicnotes
X: OpenRouter Decision Model Rankings
X: herdr + jev, an agent router
X: A group chat bot with decision room routing
More Jev resources like this
Flow Steward extension that exposes Jev through standard project connections, plus an importable support-triage example. Adds an Evaluate with JEV connector that takes state and named choice, score, or noul questions and returns answers with usage.
Jev Integrations & Adapters#github#routing#classification
CommunityFlow-Steward/ext-typesafe-ai
TypeScript service and library that places Jev between a PBX such as Asterisk or FreePBX and call destinations, classifying caller intent into predefined routes. Each route maps to an explicit PBX action, with low-confidence or unhandled calls sent to a fallback route.
Jev Integrations & Adapters#typescript#routing#github
Communityrayanweragala/jev-call-router
LangChain.js integration for TypeSafe System One models. It wraps the jev classifier as a Runnable that returns typed choice, score, and noul answers with calibrated probabilities, plus agent middleware for model routing and tool gating.
Jev Integrations & Adapters#typescript#sdk#routing
Community
Community n8n node that sends text or JSON items to Jev decision models and routes them by the answer. Supports Choice, Score, and Noul questions, backed by either a local Ollaya daemon or the hosted TypeSafe AI API.
Jev Integrations & Adapters#routing#local#typescript
Community
Effect provider package that serves the DecisionModel capability with Jev, the System One model answering typed questions with probabilities instead of text. Defines classify, rate, and probability decisions answered in one call, with typed AiError mapping.
Jev Integrations & Adapters#typescript#sdk#classification
Community
n8n community node that sends a state and a map of Noul, Choice, and Score questions to the System One endpoint in one request, returning calibrated probabilities and confidence for workflow branching.
Jev Integrations & Adapters#github#typescript#guardrails
Communityzampierid4p/n8n-nodes-typesafe
From the Jev guides
Original write-ups on what this entry is for.
From the Jev AI model community
Posts from builders shipping with Jev right now.