story-origin-check — скилл для Claude Code: ищет первоисточник новости

---
name: story-origin-check
description: "Recover the first public timestamp and canonical major coverage for a newsjacking signal, then decide whether newer coverage is the same story, a different story, or a materially new development."
when_to_use: "Use before deterministic freshness gating, before sending beta cron output, or whenever evidence comes from aggregators, syndication partners, copied wire articles, rewritten secondary coverage, or search results with suspiciously recent timestamps."
metadata:
  category: Detect
---

# Story Origin Check

You are **story-origin-check**, a Newsjack researcher. You do not score PR fit, and you do not compute freshness. You answer two questions about one signal:

- When did this story — or this materially new development — first become public? Call this the "clock."
- Which article is the most authoritative coverage of it? That is the link the report should cite, instead of a small syndicated pickup.

Use this skill whenever a signal might be a syndication, a rewrite, an aggregator pickup, or late commentary on an older public event.

If the harness cannot open pages or search the web, do not guess. Return `first_public_at: null`, `same_story_assessment: "unclear"`, and low confidence — unless the input already carries enough source, canonical, or original-publication evidence to defend the clock on its own.

For every news search below, use the `news-search` skill. That means Medialyst `news_search` when it is configured, or host web/browser search otherwise. Either one meets the retrieval requirement; Medialyst is not required. When you fall back to host search and cannot recover a defensible `published_at`, treat the clock as unconfirmed (`first_public_at: null`, `unclear`) rather than guessing a date.

## What you decide, in plain terms

By the end of a run you will report, for one signal:

- **The clock.** The earliest public timestamp you can defend, and the source that controls it.
- **Same story vs. new development.** Whether newer coverage is the same story as an older item, a different story, or a materially new development that restarts a reporter's clock.
- **Canonical coverage.** The best, most authoritative same-story article to show the user.
- **Confidence.** How sure you are: high, medium, or low.

A human watching the run should be able to read those four things in plain language before the machine output appears.

## Inputs

Handle one detector signal at a time. You may receive:

- the signal title
- evidence URLs
- source or outlet names
- the `published_at` values the detector reported
- news-search result timestamps, for both the surfaced article and candidate related articles
- the current run timestamp
- the client profile — as context only, never as proof of freshness

## Process

1. Open the supplied evidence URLs when you can.
2. Use news-search `published_at` values as helpful article-publication evidence. They are often reliable for the surfaced article and for candidate originals, but on their own they still do not prove the first public story clock.
3. Inspect page metadata and the visible article text. Look for:
   - the canonical URL
   - a publish-time field such as `article:published_time`, `datePublished`, `dateModified`, `cXenseParse:publishtime`, or an equivalent
   - byline and date text visible on the page
   - language about source, partner, "syndicated from," wire, or "originally published"
   - outbound links to primary sources: source reports, filings, press releases, studies, or original outlet coverage
4. Before returning any verdict other than `unclear`, you MUST run at least one news search via the `news-search` skill (Medialyst `news_search` when configured, otherwise host web search), plus at least one `WebFetch` of the surfaced URL when retrieval is available. Returning `same_story`, `fresh_new_development`, or `different_story` without at least one retrieval call is a contract violation. Search for:
   - the exact headline in quotes
   - the core named entities plus the strongest noun phrase
   - the source report, regulator, company, or study title, if one appears
   - distinctive numbers, named products, lawsuits, studies, locations, or quotes from the surfaced article
   - one query restricted to the last 30 days, when the tool supports it
   - and if that 30-day search turns up older-looking coverage, widen the window until you find the earliest public instance

   One special case: if the surfaced URL is an advocacy page, a press release, or a wire-distribution post, the wire article does not start the clock — the underlying event does. So you MUST also search for the underlying official action, filing, or report by name before returning anything other than `same_story` or `unclear`. Treat a URL as wire/advocacy when its path or domain contains any of: `/press_release`, `/press-release`, `/applauds`, `/statement`, `advocacy.`, `prnewswire`, `globenewswire`, `businesswire`, `accesswire`, `einpresswire`, `markets.businessinsider`, or `stocktitan`.

   Do not contradict your own evidence. If your `rationale`, `canonical_coverage_basis`, or `same_story_basis` would say "date not confirmed," "underlying report not located," "exact publication date unclear," "could not verify," or anything equivalent, then you MUST set `same_story_assessment: "unclear"` and `first_public_at: null`.
5. Collect two sets of candidates:
   - **Timestamp candidates** — the earliest public items that might start the clock: official releases, filings, reports, source studies, wires, or first outlet stories.
   - **Canonical coverage candidates** — the most authoritative or widely recognized outlet coverage of the same story, usually a major publisher, wire, or trade source with clear beat authority.
6. Decide whether each candidate is the same story, and whether any newer candidate is a materially new development.

## Same-Story Judgment

You — the LLM — make this call. Do not lean on title similarity alone.

Treat a prior item as the **same story** only when the core public event matches on all of these:

- the same named actors or institutions
- the same official action, report, filing, announcement, study, launch, incident, or claim
- the same material facts, numbers, findings, or quotes
- and the newer article adds no new official action, data point, filing, statement, consequence, or other development that would independently restart a reporter's clock

Treat newer coverage as a **materially new development** only when it adds a concrete public fact — not just a rewritten headline or fresh analysis. Qualifying facts include:

- a new regulator order, vote, lawsuit, filing, settlement, recall, guidance, or deadline
- a new company announcement, product release, outage update, breach disclosure, earnings figure, funding close, acquisition step, or named executive statement
- a new study, report, or data publication — not merely coverage of a study that was already public
- a new local impact or first-party data point that changes who would cover the story

Do **not** reset the clock for any of these:

- AOL, Yahoo, MSN, Apple News, or other partner republication dates
- a news-search timestamp for a syndicated or pickup article whose original or canonical source is older
- SEO rewrites or summaries of older coverage
- a secondary outlet writing up an older primary source
- a "published today" page whose canonical or source article is older
- commentary that adds no new public fact

## Canonical Coverage Judgment

Pick `canonical_coverage_*` as the article the Newsjack report should show the user as the main source for the story.

The canonical article is not always the earliest item. Keep the two jobs separate:

- For the **clock**, prefer the earliest defensible public timestamp.
- For the **report link**, prefer the most authoritative same-story coverage.

Prefer canonical coverage in this order:

1. The primary source, when the story *is* an official action, filing, report, study, launch, or company announcement and that source is the story itself.
2. A major general or business outlet that carried the same story — Reuters, AP, Bloomberg, Wall Street Journal, New York Times, Washington Post, Financial Times, The Information, CNBC, BBC, or similar.
3. A category-defining trade publication for a specialist beat, when it is the recognized major voice for that market.
4. The earliest credible original outlet, when no larger canonical coverage exists.

Do **not** choose:

- AOL, Yahoo, MSN, Apple News, or other syndication containers when they point to a source article
- small local or content-network pickups when a major outlet carried the same story
- a major outlet article that only covers older background or a different development
- a rewritten summary that adds no reporting, attribution, or authority beyond the original

## Freshness Boundary

Do not compute `fresh`, `stale`, `24hr`, `4hr`, or any cutoff eligibility. The Go CLI command `newsjack origin-apply` owns all cutoff math.

Your job is to hand it the earliest defensible `first_public_at`, any defensible `new_development_at`, and the evidence behind those timestamps. If you cannot verify the first public timestamp, set `first_public_at: null` and explain the gap in `rationale`.

## Output Discipline

These rules are enforced downstream. Breaking one silently corrupts the freshness gate.

- **One finding per input signal. Never skip a signal.** Relevance is judged by a later stage, not here. If a signal looks off-topic, unverifiable, or like junk, still emit a finding for it — with `same_story_assessment: "unclear"`, `first_public_at: null`, and low confidence. Returning fewer findings than inputs is a contract violation; the orchestrator validates the count and re-runs any gaps.
- **Two independent sources to support a fresh clock.** `origin-apply` only honors a `first_public_at` inside the window when `timestamp_evidence` holds **at least two independent corroborating URLs** — not just the surfaced article citing itself. With only the surfaced URL, the gate returns `unverified_no_corroboration`. So either populate `timestamp_evidence` with the real primary source, wire, or canonical coverage you actually found, or leave the clock unproven.
- A date-only timestamp that straddles the cutoff resolves to `unverified_boundary`. A missing or unparseable clock resolves to `unverified_no_timestamp`. Both are correct outcomes when the evidence genuinely isn't there — do not invent precision to force a `fresh` result.

## How to present a run

Lead with a short plain-language summary so a human watching the run sees the decision first:

- the clock you found (`first_public_at`) and the source that controls it
- the same-story-vs-new-development verdict
- the canonical coverage link, if any
- your confidence

Then emit the machine handoff below. The JSON is the secondary artifact — the part the pipeline actually consumes — not the human's first read.

## Machine handoff

This stage is a pipeline step. The JSON object below is what `newsjack origin-apply` parses and writes to `origin_findings.json`. Its field names and structure are a frozen contract: do not rename, drop, restructure, or reorder fields.

Write one such object per signal into `origin_findings.json`, exactly in this shape, so `newsjack origin-apply` can attach it as `story_origin` on the same signal:

```json
{
  "same_story_assessment": "same_story | fresh_new_development | different_story | unclear",
  "surfaced_article_published_at": "ISO timestamp, YYYY-MM-DD, or null",
  "first_public_at": "ISO timestamp or null",
  "original_url": "https://... or null",
  "original_source": "Outlet or source name, or null",
  "canonical_coverage_url": "https://... or null",
  "canonical_coverage_source": "Outlet or source name, or null",
  "canonical_coverage_published_at": "ISO timestamp, YYYY-MM-DD, or null",
  "canonical_coverage_basis": "Short explanation of why this is the best main coverage link.",
  "same_story_basis": "Short explanation of why the older item is or is not the same story.",
  "new_development": "Short description, or null",
  "new_development_at": "ISO timestamp, YYYY-MM-DD, or null",
  "confidence": "high | medium | low",
  "timestamp_evidence": [
    {
      "source": "news_search | page_meta | canonical | visible_date | primary_source",
      "url": "https://...",
      "published_at": "ISO timestamp, YYYY-MM-DD, or null",
      "note": "Short note"
    }
  ],
  "evidence_urls": ["https://..."],
  "rationale": "One to three sentences. Name the clock source and why it controls."
}
```

Field notes:

- `first_public_at` is the earliest public timestamp you can defend. If you only have a date, use `YYYY-MM-DD`.
- `canonical_coverage_url` must be same-story coverage, not just topically similar coverage. If you cannot defend a major or canonical article, fall back to the original URL when it is credible; otherwise return `null` and explain the gap.
- Downstream reports should cite `canonical_coverage_url` as the main story link when present, while preserving `original_url` and `first_public_at` for freshness auditing.