Skip to content
Metatext Decide — typed decisions for your agents, with a probability on every answerGet early access
Metatext
Sign Up

n8n-nodes-jev

n8n community node for Jev that adds a typed smart if-statement: choice, score and noul with probabilities in workflows.

ProjectbetaMIT
View on GitHub ↗

Overview

n8n-nodes-typesafe-jev adds Jev to self-hosted n8n as a node for typed decisions. It supports Choice (classify into your labels), Score (fractional score against an ordered rubric) and Noul (probability that a yes/no question is true, with an optional boolean threshold), and you can ask multiple questions of the same primitive in one call. Outputs keep every native answer, the original input, model version, token usage and measured latency.

The README frames Jev as a smart if-statement for emails, tickets, prompts and JSON state, while noting that the ~100 ms headline is TypeSafe's target and that TypeSafe reports roughly 70-500 ms end to end. It uses the official @typesafe-ai/sdk, so it is not eligible for n8n Cloud verification. Examples include a prompt-guard workflow and a cost comparison with a cheap LLM.

What the decision model decides

TypeQuestionAnswers
ChoiceWhich of your labels fits this item?—
ScoreWhere does this item sit on your rubric?—
Yes / noIs this yes/no condition true?yes · no

How it works

  1. Install the community node and add a TypeSafe Jev API credential.
  2. Drop the Jev node after your trigger; state defaults to the incoming JSON.
  3. Add questions of one primitive type.
  4. Branch the workflow on the answer, probability or threshold.

Strengths

  • No output parsing; answers are typed.
  • Keeps latency, usage and model version for auditing.
  • Multiple questions per call.

Caveats

  • Self-hosted n8n only; not n8n Cloud verified.
  • Requires a TypeSafe account and key.
  • Latency is not guaranteed; 70-500 ms per TypeSafe.

Getting started

In self-hosted n8n: Settings > Community Nodes > Install, enter n8n-nodes-typesafe-jev, then add your TypeSafe API key as a credential.

At a glance

Type
Project
Status
beta
Language
TypeScript
Runtime
n8n community node
Has code
yes
Has tests
yes
License
MIT

Use cases and integrations

How curated lists describe it

Frequently asked questions

What can the node decide?

Choice, Score and Noul questions about each item's state.

Does it work on n8n Cloud?

No, it uses a runtime dependency so it is not n8n-verified; it targets self-hosted n8n.

How fast is it?

The README cites TypeSafe's 70-500 ms end-to-end range and warns ~100 ms is not a guarantee.

Related projects

  • Sim

    Open-source workspace for building, deploying and monitoring AI agents and workflows, listed as integrating TypeSafe System One.

    ProjectproductionTypeScript
  • agentrun

    Workflow DSL for existing agents that uses Jev for the steps that are decisions and calls an agent only when investigation is needed.

    LibrarybetaTypeScript
  • n8n-nodes-jev-classification

    n8n Jev Classification node that branches items by category with a Needs Review branch, plus parallel and batched requests.

    ProjectbetaTypeScript
  • n8n-nodes-typesafe

    n8n node that sends a state and a map of typed questions to /v1/systemone in one request and returns per-question answers.

    ProjectbetaTypeScript
  • n8n-nodes-typesafe-ai

    n8n community node for TypeSafe's System One API with yes/no, choice and score questions, retry handling and an AI Agent tool.

    ServicebetaTypeScript
  • jev-orchestrator

    Python library to batch, gate, trace and calibrate thousands of typed Jev decisions, maximising questions per call.

    ProjectprototypePython