---
title: Litmus — 20 deterministic trust signals for every page | NanoParse
url: https://www.nanoparse.app/litmus
description: "Every NanoParse parse returns twenty data points: source authority, authorship, freshness, structural trust, paywalls, syndication, key figures and more. Deterministic, inspectable, $0.002."
---

Litmus · 20 checks · deterministic

# Twenty data pointsbefore your agent cites a page.

Every parse returns clean Markdown plus Litmus — twenty rule-based checks your agent can inspect before it trusts a page. No model in the path, no temperature, no second opinion that changes overnight.

[Try it free](https://www.nanoparse.app/#parse)[See the response schema](https://www.nanoparse.app/quickstart)

litmus.json

v2

Verdict

Safe to build on

Cite

sourcenews · 0.82

authorshipJane Chen · jsonld

freshness2026-01-14 · current

trusthigh · 0.92

paywallnone

reading_time7 min

20 of 20 signals0 flags

Reading the verdict

## Four patterns to check first.

The signals only matter if they change what happens next. These four cover nearly every call an agent makes.

Verdict order

drop→review→cite

A hard failure wins. If a page is gated, missing or erroring, no amount of polish elsewhere upgrades it to citable.

Cite01

### Safe to build on

verdict is "cite" — authorship present, citations dense, density high, no flags. The page says what it claims and shows its work.

Cite it, and log the verdict.

Drop02

### Wrong page, not wrong source

page_status fires, or a paywall is detected, or content is truncated. The agent read a gate, not the article.

Do not cite; retry elsewhere or escalate.

Review03

### Right content, wrong copy

syndication is detected, or redirects crossed domains. The content is real, but you are holding a copy.

Cite the original via canonical.

Review04

### Assertion dressed as evidence

Thin content, heavy hedging, no citations, no named authorship. The prose asserts more than the page can support.

Fine to mention, not fine to build on.

Every signal, with its payload

## The full twenty, in four groups.

Source, trust, validity and read. Each one is deterministic — a given page and a given Litmus version produce the same signals every time, so you can cite them in your own logs.

Signal groups

01Source602Trust503Validity604Read3

20 of 20 active

Rule-based and extractive. No model, no temperature, no second run that disagrees with the first.

### Source

6 signals

Who wrote it, what it claims to be, who it links to, and whether it shows its work.

01source

Page type, authority score, and the reasoning behind it.

`type: "news" · authority: 0.82`

`type: "blog" · authority: 0.21`

02authorship

03content_type

04outbound_authority

05citations

06structured_data

## All 20, spelled out

One line per signal: what it measures, a good example and a bad one. Same reference as the explorer above, in plain reading order.

### Source

6 signals

Who wrote it, what it claims to be, who it links to, and whether it shows its work.

sourcePage type, authority score, and the reasoning behind it.`type: "news" · authority: 0.82``type: "blog" · authority: 0.21`

authorshipA named person, an organisation, or nobody — and where it was found.`present: true · kind: "person"``present: false · kind: "unknown"`

content_typeWhat the page actually is, not what the URL suggests.`type: "news_article"``type: "thin_listing"`

outbound_authorityA score plus the domains it links out to — primary sources or more blogs.`score: 0.71 · top: "who.int"``score: 0.22 · top: "blog.example"`

citationsReferences section, external links, and links per 1,000 words.`has_references: true · per_1k: 3.2``has_references: false · per_1k: 0`

structured_dataJSON-LD schema, OpenGraph, and the canonical URL.`json_ld: true · og: true``json_ld: false · og: false`

### Trust

5 signals

Is it substance, or hedging and bare assertion? These signals count what is actually on the page.

structural_trust_scoreHigh, medium or low — with the factors behind it, so it can be inspected.`level: "high" · trust_factors: 3``level: "low" · trust_factors: 1`

hedge_languageDensity of "reportedly", "some say", "allegedly" — claim vs rumour.`density: "low" · markers: 2``density: "high" · markers: 14`

content_densityThin-content flag plus ratio of substantive sentences.`thin_content: false · ratio: 0.78``thin_content: true · ratio: 0.31`

boilerplate_ratioHow much is nav, ads and furniture. High ratio plus low density is the content-farm shape.`ratio: 0.18``ratio: 0.74`

quote_densityAttributed quotes and attribution verbs per 1,000 words.`quotes: 3 · per_1k: 0.7``quotes: 0 · per_1k: 0`

### Validity

6 signals

Is it current, is it reachable, is it gated, is it the original, is it the page you asked for.

freshnessPublished date, age, Last-Modified, staleness, and corrections.`published: "2026-01-14" · stale: false``published: "2019" · stale: true`

page_statusSoft 404, login wall, captcha, JS shell. A 200 is not proof you got the article.`kind: null — a real page``kind: "soft_404" — not the article`

paywallDetected or not, with a note — a gate never passes as content.`detected: false``detected: true`

syndicationIs this a copy of something published elsewhere? Cite the original.`detected: false``detected: true`

is_archivedPresence in the Wayback Machine and archive.today.`wayback: true``wayback: false`

redirectsFinal URL, hop count, and whether it crossed domains.`hops: 0 · cross_domain: false``hops: 2 · cross_domain: true`

### Read

3 signals

What it says and how long it takes — enough to triage before you spend context.

tldrA one-line summary of what the page says — extractive, not generated.`"A 20-year cohort study links…"``"" — nothing extracted`

key_figuresThe numbers on the page, each with the context it appeared in.`7.3 yr — "increase in life expectancy"``0 figures`

reading_timeMinutes, to budget context before you spend it.`minutes: 7``minutes: 1`

The whole payload

## One response. Markdown and the verdict, together.

No second endpoint, no second charge. The Markdown is the page; the signals are its vitals. Every field is inspectable, and every unknown is returned as null or "unknown" rather than guessed.

Nearly everything comes from the rendered page

DOM, Markdown, response headers — measured, not inferred.

Two signals look outside the page

is_archived queries the Wayback Machine; source.authority leans on a domain reference list.

Versioned, so runs can be compared

The Litmus version is pinned to the response. Same page plus same version equals the same signals.

Fails visibly

A missing signal is null or "unknown", never a confident fabrication.

litmus_version 220 groupsdeterministic

POST /fetch — 200 response

Copy

```
{ "url": "https://example.org/cohort-study", "markdown": "# Study Finds Correlation Between Diet and Longevity …", "litmus": { "litmus_version": "2", "verdict": "cite", "flags": [], "source": { "type": "news", "authority": 0.82, "rationale": "established publisher" }, "authorship": { "present": true, "name": "Jane Chen", "kind": "person", "source": "jsonld" }, "content_type": { "type": "news_article" }, "outbound_authority": { "score": 0.71, "top_domains": [ "who.int", "thelancet.com" ] }, "citations": { "has_references": true, "external_links": 12, "external_per_1k_words": 3.2, "primary_domain_links": 5 }, "structured_data": { "json_ld": true, "schema_types": [ "NewsArticle" ], "open_graph": true, "canonical": "https://example.org/cohort-study", "meta_description": true }, "structural_trust_score": { "level": "high", "trust_factors": [ "High-authority source", "Named authorship", "Dense citations" ] }, "hedge_language": { "density": "low", "markers": 2 }, "content_density": { "thin_content": false, "ratio": 0.78 }, "boilerplate_ratio": { "ratio": 0.18, "confidence": "high" }, "quote_density": { "attributed_quotes": 3, "attribution_verbs": 5, "per_1k_words": 0.7 }, "freshness": { "published": "2026-01-14", "age_days": 276, "stale_warning": false, "correction_noted": false }, "page_status": { "is_error_page": false, "kind": null, "confidence": "high" }, "paywall": { "detected": false, "note": null }, "syndication": { "detected": false, "canonical_url": null }, "is_archived": { "wayback": true, "archive_today": false }, "redirects": { "final_url": "https://example.org/cohort-study", "hops": 0, "cross_domain": false }, "tldr": { "text": "A 20-year cohort study links the Mediterranean diet to a 7.3-year increase in average lifespan." }, "key_figures": [ { "value": "7.3 yr", "context": "increase in life expectancy on the Mediterranean diet" }, { "value": "24,200", "context": "participants across seven countries" } ], "reading_time": { "minutes": 7 } }}
```

scroll

abbreviated markdown · full signal block

![](https://www.nanoparse.app/brand/nanoparse-logo-official.svg)

## Run it on your own URL.

No account, no wallet. Three free tries, then $0.002 per parse. A test costs less than one wrong citation.

[Run a test — $0.002](https://www.nanoparse.app/#parse)[Read the quickstart](https://www.nanoparse.app/quickstart)

10 free parses a month · no account · no API key
