draftpulse

Experimental: live X draft viral scorer powered by TypeSafe Jev

How it uses Jev

Live Debounced Scoring : Automatically evaluates drafts on typing pauses ( 500ms), giving instant visual feedback without blocking the editor. Six Viral Dimensions : Scores hook strength, concrete specificity, reply propensity, bookmark shareability, dwell time, and slop risk. TypeSafe Jev Integration : Evaluates multiple narrow rubric questions in a single structured LLM call via TypeSafe Jev. Deterministic Fallbac…

Highlights

  • Composite Scoring Formula — Combines individual dimension ratings into an overall readiness score using configurable weights in server/weights.ts .
  • Category Auto-Detection — Automatically identifies the post genre (Build in Public, Technical Insight, Hot Take, Story, Question) to apply appropriate evaluation rubrics.
  • Server-Side Key Isolation — Keeps TYPESAFE API KEY strictly in the backend Express service; secrets are never bundled or exposed to the client browser.
  • Calibrated Feedback Loops — Designed around the data-loop plan to continuously refine heuristics based on actual observed post performance.

Use cases

  • "Validate this product launch hook before posting."
  • "Detect AI slop patterns and overused buzzwords in this thread draft."
  • "Estimate bookmark shareability for this technical guide summary."

Getting started

Prerequisites Node.js 18+ and npm Optional: TypeSafe API Key (from TypeSafe)

Source: community catalog. Details and metrics reflect the published community snapshot and are not a performance endorsement. Documentation details extracted from GitHub · the README.

Keep exploring

Similar to draftpulse

Browse all
jevibe-check project preview
Applications
A live tone labeler for Bluesky posts and drafts, using TypeSafe's Jev API.
mimicry project preview
Applications
Rewrite AI drafts in your own voice with a bounded TypeSafe feedback loop.
jev-trade project preview
Applications
Live Jev trader on Hyperliquid
jev-model-router (lucianfialho) project preview
Cost-optimized OpenRouter model router using TypeSafe's Jev, with a live full-catalog scorer instead of a hardcoded model list
eve project preview
Integrations
Vercel's open agent framework, which ships Jev as the default evaluation model in its experimental evaluate path.
jev-gate project preview
Benchmarks & research
Not every coding task needs your best model. Experimental Jev-powered model routing for Claude Code — V3 prototype runs today, V4 routes at the task boundary.
tsai-civ2 project preview
Games & simulations
TypeSafe Jev plays original Civilization II in a browser, with live action probabilities. Experimental full-game harness.
JevRouter project preview
Agent tooling
A lightweight Jev-powered router for models, tools, and subagents
x-scanner project preview
Browser & computer use
Chrome extension that labels every post you scroll past on X with typed Jev judgments and a live cost counter