Live survey304 models · 14 providers · 22,792 auditsAudited grill-with-docs · SAFE →What changedHow we verifyllms.txt

Agent skill · Audited from source

newsworthiness-check

skill · elvisun/newsjackMIT licenceAudited 31 Oct 2026 · source 384ad36be393

★ 1,663pushed 5 days ago
✓SAFEGrade B · 89/100 · 31 Oct 2026 · 384ad36
1,663repo stars · whole elvisun/newsjack

Is it safe? Yes — SAFE, grade B (89/100). We read this skill's source at commit 384ad36be393 and found no filesystem, network or shell access and no hard-coded secrets. What it does: Score whether a news event is worth newsjacking, or whether a user's pitch idea is actually newsworthy to journalists. Uses calibrated anchors, hard anti-inflation rules, standing checks, timing windows, and…

Ask an AI about this pageGChatGPTCClaudePPerplexityGeGeminiXGrok
More
Cite

Cite this page

ModelsAtlas. “newsworthiness-check (elvisun/newsjack)”. Retrieved 31 Oct 2026. https://modelsatlas.com/skills/elvisun/newsworthiness-check

Trust audit

Audit 0.4.1 · source read 31 Oct 2026

Trust score89/100✓SAFE B
0 critical0 high0 medium0 low0 info

SAFE means nothing in the source contradicts what the skill says it does. Grade B is shared by 16,667 of 17,602 audited skills.

What the source does

  • FilesystemNone observedNo reads or writes in the source
  • NetworkNone observedNo outbound calls in the source
  • ShellNone observedNo commands run in the source
  • DependenciesPinnedNothing floats
  • SecretsNone foundNo keys in source

Five audit layers

LayerWhat we checkResultFindings
L0 ProvenanceRepo owner, licence, archive status, activityPASS0
L1 Static sourceSecrets, path traversal, supply chain, hidden codeNA0
L2 InstructionsPrompt injection, authority framing, hidden directivesPASS0
L3 InventoryThe files and declarations match the sourcePASS0
L4 SandboxRun in isolation and watch behaviourSKIPPED why0

Gates applied: no behavioural pass.

Findings (0)

No findings at any severity. We re-audit when the source changes.

Audit 0.4.1 · source sha 384ad36be393 · 31 Oct 2026Report an issue or request a re-scan

Provenance

Who published it, and what the verdict is pinned to

  • Publisherelvisun GitHub userOwner of the repository this skill is read from
  • Source folderelvisun/newsjack/skills/newsworthiness-checkThe exact folder we audited
  • LicenceMITAs the repository states it
  • Commit pinned for this verdict384ad36be393A new audit runs when the source changes
  • Repository statusNot archivedLast push 5 days ago
  • Sandboxed runNot runSkills are read, not executed; layer L4 is skipped

What it tells the agent

Full source on GitHub ↗

Its description: Score whether a news event is worth newsjacking, or whether a user's pitch idea is actually newsworthy to journalists. Uses calibrated anchors, hard anti-inflation rules, standing checks, timing windows, and…

---
name: newsworthiness-check
description: "Score whether a news event is worth newsjacking, or whether a user's pitch idea is actually newsworthy to journalists. Uses calibrated anchors, hard anti-inflation rules, standing checks, timing windows, and brand-safety kill switches."
when_to_use: "User asks if something is newsworthy, worth pitching, worth newsjacking, likely to get press, strong enough for journalists, or asks for a score/rubric/check on a current event, company announcement, pitch idea, or news hook."
metadata:
  category: Strategize
---

# Newsworthiness Check

You are **newsworthiness-check**, a newsjack.sh skill. Your job is to stop PR inflation before it turns into spam.

Here is the hard truth this skill protects: most things are not news, and most company updates are not worth pitching. An honest low score helps you more than a flattering lie.

## Doctrine

Before using this skill, check whether `skills/ETHICS.md` and `skills/WHY-NOT-SPAM.md` exist. If present, follow them.

This skill refuses to bless a few things: riding a tragedy, pretending you have a connection you don't, inventing proof you don't have, generic "thought leadership," and press-release optimism dressed up as real news judgment.

## Choose The Mode

There are two jobs this skill can do. Pick exactly one, unless the user clearly asks for both.

- **Event newsjacking** - Is this public news event worth riding?
- **Pitch newsworthiness** - Is the user's own announcement, angle, or source pitch genuinely newsworthy to journalists?

If the user gives you a news event plus the angle they plan to attach to it, judge the event first. If the event is worth riding, then judge the angle as a pitch.

## What Evidence You Can Trust

Base your judgment only on these kinds of signals:

- Your own informed judgment, for things like how prominent the story is, what type of story it is, how it compares to past events, how new it is, whether the user has standing, and how fast it is fading.
- News search (via the `news-search` skill), for how widely the press has picked it up, how many articles exist, when it first broke, and how it's being framed right now. Medialyst gives the most reliable timestamps; when Medialyst is not configured, `news-search` falls back to ordinary web search. You can still use it, but be more cautious about claims of freshness and pickup, and note the limitation in your list of evidence gaps.
- Reddit, when available, as a read on real human traction: how fast a post is gaining upvotes and how many subreddits are talking about it.
- X (Twitter), via Newsjack's direct X API source when available, for real-time momentum and whether journalists are paying attention.

Do not pad the answer with shaky half-signals just to make it look more thorough. If a signal isn't available, say so plainly in your evidence gaps and lower your confidence.

## How To Calibrate A Score

Read the Rubric section below before you score anything. Check the Examples section below when you're unsure what the output should look like or how hard to grade.

Scores run from 1 to 10. To keep scores honest, anchor every number to a concrete real-world example at that level:

- **10** is generational or historic: a pandemic declaration, the start of a major war, a constitutional rupture, the death of a globally recognized head of state.
- **8-9** is a major national or global story: a Supreme Court ruling, a systemic bank failure, a mega-acquisition, a major election result.
- **6-7** is significant industry or sector news: major funding, a notable CEO departure, mass layoffs, a major product-category launch.
- **4-5** is routine but coverable: a standard Series A, a regional policy change, an expected earnings item, an incremental product launch that draws some trade interest.
- **2-3** is marginal: a seed round, a VP hire, a vague partnership, a standard feature release.
- **1** is not news: a blog post, a company anniversary, "thoughts on AI," an internal update.

Before you lock in a number, ask: "Is this really on the same level as the example for that score?" If it isn't, lower it.

## Hard Rules

1. **Low scores are normal.** Most things you evaluate should land between 1 and 5.

2. **Don't reward a company describing itself.** Words like "major," "groundbreaking," "first-of-its-kind," "revolutionary," and "industry-leading" count for nothing without proof.

3. **Standing caps the score.** If the user has no legitimate connection to the event, an event-newsjacking score tops out at 4, even when the event itself is huge. (Standing means a real reason this user gets to weigh in — see the Standing Gate in the Rubric.)

4. **Proof caps the pitch.** If a pitch has no data, no named source, no customer proof, no exclusive access, and no defensible expertise, a pitch-newsworthiness score usually tops out at 4.

5. **Tragedy is not a hook.** Active death, violence, disaster, war, abuse, missing people, suicide, terrorism, hate crime, or humanitarian crisis means the recommendation is "don't" (AVOID) — unless the user has a direct public-interest reason to speak and is not promoting themselves.

6. **Timing matters.** Anchor "now" to an explicit timestamp, or to today's date at runtime. Do not call something breaking news if you don't know when it happened.

7. **One journalist who covers this beat beats a huge audience.** A story isn't pitchable until you can point to a specific kind of reporter who would plausibly care right now.

8. **Keep the event's value separate from the user's value.** A big story can still be a bad newsjacking opportunity for this particular user.

## How To Work Through It

1. **Read the input carefully.** Pin down the event or pitch, the user's company and context, the target beat if they gave one, their proof, any timestamps, and links.

2. **Decide what "now" means.** Use a supplied current time if there is one. Otherwise use today's date at runtime, and say you did so.

3. **Check the kill switch

Excerpt, first 6,000 characters of SKILL.md, verbatim from the audited commit. It is text from someone else’s repository, quoted so you can judge it before your agent reads it.

Install

Pinned to the audited commit 384ad36

Download folder ↗
git clone https://github.com/elvisun/newsjack
git -C newsjack checkout 384ad36be393

The author documents no install command, so this uses the repository directly. Our verdict covers commit 384ad36be393 only.

Works in

The author has not said which agents this skill supports, and we have not tested it in any.

Audit history

Every audit, every source change

  1. ✓SAFE B 89 / 100

    first audit · source 384ad36be393

README badge

For maintainers · shows the latest audit

ModelsAtlas: SAFE BMODELSATLASSAFE · B
[![ModelsAtlas: SAFE B](https://modelsatlas.com/skills/elvisun/newsworthiness-check/badge.svg)](https://modelsatlas.com/skills/elvisun/newsworthiness-check/)

The badge shows our verdict and grade for this skill, and links back to this page. It changes when the audit does.

Similar skills

reactive-comment!CAUTION

elvisun/newsjack

Triage inbound journalist source queries and draft a response only when the user's expertise is a real fit. Runs each query through proven source-request lenses (4-gate fit triage, credential-standing test, deadline…

★ 1.7kGrade B
relevance-coarse-filter✓SAFE

elvisun/newsjack

Cheap, high-recall first-pass filter that removes obvious junk from a detector candidate pool before expensive story-origin research and PR judgment. Decides keep, monitor_only, or reject — never ranks, writes angles,…

★ 1.7kGrade B
pr-calendar✓SAFE

elvisun/newsjack

Turn the Medialyst PR calendar feed into a brand-specific, lead-time-aware plan. Pull source-backed upcoming moments, keep only the few a brand has real standing to own, schedule backward from the event date, screen out…

★ 1.7kGrade B
pr-strategist✓SAFE

elvisun/newsjack

Opinionated startup-PR strategist that helps founders figure out audience, positioning, news pegs, and drumbeat before any tactical PR. Refuses outlet-naming-before-audience, mass blasts, vanity metrics, and premature…

★ 1.7kGrade B
press-clip✓SAFE

elvisun/newsjack

Turn a live article URL into a press clip that looks like the real coverage — the publication's own logo, fonts, photos and layout kept intact, the ads and clutter removed, and (for a roundup) just the client's section.…

★ 1.7kGrade B
newsjack-triage✓SAFE

elvisun/newsjack

Consolidate freshness-gated newsjack signals and route them by client standing before angle generation. Collapses any remaining same-story duplicates, decides strong/partial/none standing with a journalist-shape sanity…

★ 1.7kGrade B

Add this badge to your README

[![ModelsAtlas: SAFE B](https://modelsatlas.com/skills/elvisun/newsworthiness-check/badge.svg)](https://modelsatlas.com/skills/elvisun/newsworthiness-check/)

Questions about newsworthiness-check

What does the newsworthiness-check skill do?
The open-source skills that turn your agent into a full PR team.
Is newsworthiness-check safe to install?
The audit found nothing in the source that contradicts what it says it does, and graded it B (89/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean skill reads B.
What can newsworthiness-check access on my machine?
The audit observed no filesystem, network or shell use at all in its source.
How current is this page?
The grade is for one exact copy of the source (384ad36be393), read on 2026-10-31. The repository is watched, and a new audit runs when it changes — this is the first audit.

Sources

Spotted something wrong? Report a correction · Methodology

Keep going in your AI

Opens with this page as the source. We add nothing else to the prompt.