clarity-judge
Writing checker running separate named Jev checks such as hedging, filler, tone and passive voice, each with its own verdict.
Overview
Clarity Judge evaluates writing against separate, named checks with Jev: hedging, em dash overuse, clarity, filler phrases, tone, passive voice and actionability. You inspect each verdict, confidence signal and supporting sentence instead of one opaque score. It does not rewrite text or judge factual correctness, and it describes itself as an independent community project, not an official TypeSafe product.
The public demo at judge.jev.works deliberately has no server key, so visitors see deterministic simulated results labelled as demo mode unless they paste their own TypeSafe key, which stays in their browser's localStorage and is sent only as a header to the deployment's own endpoint. Running locally with Next.js and a key in .env.local gives real verdicts.
What the decision model decides
| Type | Question | Answers |
|---|---|---|
| Yes / no | Does this text hedge excessively / overuse em dashes / use filler phrases / rely on passive voice?(inferred) | yes · no |
| Score | How clear / actionable is this text, and is the tone appropriate?(inferred) | — |
How it works
- Paste writing
- Each named check is sent to Jev
- Show per-check verdicts, confidence and supporting sentences
- Writer revises manually
Strengths
- Separate, inspectable checks
- Public demo cannot spend anyone's credits
- Key stays in your browser
Caveats
- Demo mode shows simulated results
- Does not check facts
Getting started
git clone https://github.com/TypeSafeAI/clarity-judge.git, set TYPESAFE_API_KEY in .env.local, then pnpm dev.At a glance
- Type
- Demo
- Status
- beta
- Language
- TypeScript
- Runtime
- Next.js web app
- Has code
- yes
Use cases and integrations
How curated lists describe it
Multi-axis writing quality checker powered by TypeSafe AI's Jev model. Separate named checks, each with its own verdict and confidence.
hellogumbo/awesome-jev
Frequently asked questions
Are demo results real?
No, without your own key they are labelled simulated results.
Does it rewrite my text?
No, it only evaluates.
Is it official TypeSafe software?
The README says it is an independent community project.
Related projects
Analytics engine that rates published posts on pre-registered rubrics with Jev and tests ratings against real engagement.
ProjectbetaTypeScriptDocument evaluation CLI and CI gate using Jev or OpenAI Decisions, with pre-registered validation and documented reliability limits.
CLIbetaPythonRussian-language Letta agent skill that uses Jev as a fast external judge, with rules for treating mid-range probabilities as 'unsure'.
ProjectprototypePythonCommunity tool that fetches a SaaS homepage as Markdown and asks Jev ten questions about first-visit clarity, with a leaderboard.
ServicebetaTypeScriptUpload a paper PDF and get a fast Jev-based judgment of understandability and completeness, with low-confidence answers flagged.
DemoprototypeTypeScriptComparative judgment of ASAP essays by Jev and two LLM judges, turning pairwise decisions into scores compared with human grades.
ServiceresearchPython