Skip to content

draftpulse

Experimental editor that scores X drafts live on six viral dimensions with one Jev call, with a heuristic fallback.

ProjectprototypeMIT

Overview

DraftPulse scores social media drafts for X as you type. On a pause of about 500 ms it evaluates the draft on six dimensions: hook strength, concrete specificity, reply propensity, shareability, dwell time and a penalty for AI slop such as clichés and boilerplate. It also detects the post's category and combines dimension ratings into an overall readiness score using configurable weights.

Jev evaluates the narrow rubric questions in a single call. Without an API key the tool falls back to a built-in deterministic heuristic, so it works out of the box. The README states a zero-scraping data policy: any calibration uses only the official X API, owned posts or licensed archives. It is marked experimental. The UI runs locally alongside a separate scoring API.

What the decision model decides

TypeQuestionAnswers
ScoreHow strong is the hook in the first 80 characters?(inferred)—
ScoreHow much AI slop risk does this draft carry?(inferred)—
ChoiceWhich category is this post?(inferred)—

How it works

  1. User types a draft
  2. After a pause, send the draft to Jev with six rubric questions
  3. Combine scores with configured weights
  4. Show dimension bars, category and readiness score

Strengths

  • Live, non-blocking feedback
  • Works without a key via heuristics
  • Configurable weights

Caveats

  • Experimental; virality scores are not validated predictions
  • Requires a key for Jev scoring

Getting started

npm install && cp .env.example .env && npm run dev, then open http://localhost:5174.

At a glance

Type
Project
Status
prototype
Language
TypeScript
Runtime
web app + local API
Has code
yes
License
MIT

Use cases and integrations

How curated lists describe it

Frequently asked questions

What does Jev score?

Hook strength, specificity, reply propensity, shareability, dwell time and slop risk.

Do I need a key?

No, a heuristic fallback works without one.

Does it scrape X?

No, the README states a zero-scraping policy.

Related projects

  • Analytics engine that rates published posts on pre-registered rubrics with Jev and tests ratings against real engagement.

    ProjectbetaTypeScript
  • Chrome extension that grades readable page sections on ten rubric scores with Jev via Vercel AI Gateway and gives the page an A-E grade.

    ProjectbetaTypeScript
  • Codex plugin that runs polls on simulated reader profiles with a System One model, recording typed answers for follow-up studies.

    ProjectbetaTypeScript
  • Community tool that fetches a SaaS homepage as Markdown and asks Jev ten questions about first-visit clarity, with a leaderboard.

    ServicebetaTypeScript
  • Scores a page for SEO, answer engines and generative engines, using code for measurable checks and Jev rubric scores for judgment.

    DemoprototypeTypeScript
  • Web demo that reads a draft social post and returns Jev's tone choice, a 0-3 virality score and a cringe probability in one call.

    AppprototypeTypeScript