Skip to main content

Free kit, 11 files

AI News Content Kit

Score an AI story before you cover it, then turn it into a post or article with your sources checked.

Unlock all 21 tools, free

Opening AI News Content Kit.

Sample

Real output from a made-up example

Made by the same code as the tool

A made-up launch, scored by the same code your story goes through. Nothing here happened: the company, the product and every link are invented.

Score and verdict

82out of 100
Strong

Worth covering now, if the sources hold up.

  1. Strong 80+
  2. Promising 65+
  3. Mixed 50+
  4. Limited under 50
  • Relevance to your audience28 / 35
  • Usefulness and demo potential20 / 25
  • Primary-source confidence20 / 20
  • Freshness10 / 10
  • Demand signal4 / 10

ProceedProceed: primary source confirmed, fresh, strong audience fit

  • Primary source confirmed: you read it, and it is an official or research site
  • Exact time, 3.5 hours before the check, inside your 3 days window.
  • Strong audience fit: you can say who changes what, and how
  • All 3 claims are supported by a source you read
  • Strong score (82): Worth covering now, if the sources hold up.

The story

Fernhill Labs launches Pilot 2, an AI agent that drafts your weekly report

Exact time, 3.5 hours before the check, inside your 3 days window.

Sources, each with a tier

5 links, 4 unique pages: the repeat was the same page with tracking added.

  1. S1

    fernhill.example.com/blog/introducing-pilot-2

    PrimaryPrimaryRead2 links collapsed

  2. S2

    docs.fernhill.example.com/pilot-2/pricing

    PrimarySupportingRead

  3. S3

    techledger.example.net/2026/10/04/fernhill-pilot-2-launch

    SecondarySupportingRead

  4. S4

    social.example.org/fernhillhq/posts/1048

    Aggregator or socialBackgroundNot read yet

Claim ledger

Each claim points at a source and a place in it.

  1. 1

    Pilot 2 is available on every paid plan from October 4.

    SupportedS1Announcement, "Availability" section

  2. 2

    Fernhill says Pilot 2 drafts a weekly report from connected spreadsheets in about two minutes.

    SupportedS1Announcement, caption under the demo video

  3. 3

    Pilot 2 costs $20 a month per seat on the Team plan.

    SupportedS2Pricing page, Team column

Free, one email

Score your own story

  • Paste a headline and links, and watch repeats collapse into one story
  • Rate five weighted dimensions and get a score, a band and a verdict
  • Log each claim against its source and copy a research brief
  • The kit: scorecard, source checklist, claim ledger and post and article templates

Continuing subscribes the email on that account to Work Worth Doing, my newsletter, the same as the form below, and unlocks every free tool in this browser. We only receive your email address. Unsubscribe in one click.

This unlocks every free tool on the site. You'll also get Work Worth Doing, Demetri Panici's newsletter, every week: what I built and what changed in AI. We'll email you a link to confirm. Unsubscribe in one click.

Already subscribed? Enter the same email to unlock this browser. You won't be signed up twice.

The scorecard

How the score works

I adapted this from the scorecard behind my own AI news dashboard. Five dimensions share 100 points. The weights are fixed, so a story that is exciting but thinly sourced cannot outscore one that is useful and well sourced.

Relevance to your audience
35points
Would your audience do something differently because of this?
Usefulness and demo potential
25points
Could you show it, test it or walk through it?
Primary-source confidence
20points
How well does the original source back the story?
Freshness
10points
Is it new inside your window?
Demand signal
10points
Is your audience asking about it?

Bands. Strong 80 and up, Promising 65 and up, Mixed 50 and up, Limited under 50.

  • Only an exact publication time inside your window earns freshness points. A date alone, or no date, scores zero, because a made-up midnight would be a guess.
  • One source is primary, and it has to be one you read. An unread link is a suggestion and cannot back a claim.
  • The tier guess comes from a short list of domains that is not authoritative. Every tier is yours to change.
  • The score says how well a story fits your audience and how well it is sourced. It does not predict how a post will perform.

Set it up

Set it up where you already work

One ZIP, three places it goes. Pick the one you use.

  1. Claude app

    In Settings, then Capabilities, turn on code execution and file creation. Then open Customize, then Skills, press +, choose Create skill, then Upload a skill, and pick the ZIP.

  2. Claude Code or Codex

    Unzip it and move the news-content-kit folder into ~/.claude/skills/ for Claude Code, or into ~/.agents/skills/ for Codex.

  3. Any other chat assistant

    Open the paste-in file from the ZIP and paste it at the start of a chat, or into the assistant's custom instructions.

The same email opens these

More free tools

See all 21
  • Download, 9 files

    YouTube Packaging Kit

    See your title and thumbnail close to how viewers will, on a phone and a desktop, before you publish.

  • Runs in your browser

    Post Length Checker

    Write a post once and see how it fits on X, LinkedIn, Threads, Instagram, TikTok, YouTube and Bluesky.

  • Runs in your browser

    Which AI should I pay for?

    Tell it what you actually do. Get the best tool for each job, the one plan to pay for, and what you give up.