Skip to content
JevDirectory.orgTraffic count unavailable. View site traffic

Jev directory

CommunityJev Tools & CLIs

Jev Writer

Content analytics and LLM-as-a-judge harness that rates published posts against pre-registered rubrics using Jev's calibrated probabilities, then tests those ratings against real engagement numbers under power gates and false-discovery corrections.
See highlightsOpen github.com
Stars
3
Language
JavaScript
License
MIT
Reading time
2 min
Published by
Community
Author
kaos599
Last verified
Oct 2026

kaos599/jev-writer

User repository on GitHub

View on GitHub

Find out which qualities of your writing actually predict engagement. Rates every post you have published against a pre-registered rubric using Jev's calibrated judgments, then tests those ratings against your real engagement numbers. Refuses to report findings your sample cannot support.

Forks
0
Last pushed
Updated Sep 2026

Repo stats from the GitHub API, cached Oct 2026.

Tagscommunitycatalog-importlogicrwanalyticswriting
Works forJev for EvaluationJev for Scoring
Used inJev for Sales & Marketing
Languages
JavaScript
Platforms
CLIOpenRouterVercel AI Gateway
Capabilities
Open sourceRequires API key

Highlights

  • Rates each post with thirty Jev judgments against a pre-registered rubric, then correlates them against impressions, reach rate and conversion rate.
  • Enforces an observational power gate, date-confound controls and Benjamini-Hochberg corrections, and refuses to state findings the sample cannot support.
  • Ships as an agent skill or CLI: doctor checks your key with one live call while `run ./exports` builds, rates, analyses and renders in one step.
  • Works with LinkedIn exports today and any platform you write a rubric pack for; bring a Vercel AI Gateway, OpenRouter or TypeSafe key.
  • The README stresses it analyses posts and does not write them; the CI badge reports 83 passing tests, and a companion professional-skills package pairs with it.

Quickstart

bash
npx github:Kaos599/jev-writer doctor
npx github:Kaos599/jev-writer run ./exports

Watch out

Grounded on the truncated README and GitHub metadata reviewed 2026-10-15; the contents list is visible but the known-limits and costs sections were cut off. Ratings are Jev judgments correlated with observed engagement, not a causal model, and LinkedIn is the only platform integrated today.

Questions about this entry

Short answers drawn from the linked project or page.

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.

Follow @typesafeai