The Networkr Standards

What we enforce. What we reject. How you can verify.

Version 1.0Published 2026-04-18·Changelog

This page describes what Networkr enforces in every article published on every site in its network. We treat it as a binding commitment, not marketing. Every claim ends with a way for you to verify it yourself.

What we are

Networkr generates SEO articles, cross-links them across a network of sites that share editorial standards, and auto-heals posts based on Google Search Console verdicts. We are AI-assisted. We disclose it on every article.

What we reject, by name
  • Hallucinated URLs in articles — stripped pre-publish via sitemap allowlist.
  • Unverifiable factual claims — extracted and checked by a second-pass model; low-confidence claims block publish.
  • Thin content — minimum 800 words; quality-scored above that; randomized target per article between 1,200–3,500 to avoid formulaic footprint.
  • PBN footprints — shared branding, shared voice, duplicate niches, synchronized publish windows, shared hosting. Blocked at admission; re-audited weekly.
  • Doorway pages, scraped content, keyword-stuffed pages — detected at crawl; admission rejected or existing site quarantined.
  • Cloaking, sneaky redirects, link schemes — as defined by Google Search Essentials. Non-negotiable.
  • Strategy leakage from the user's own site — posts may only reference what is publicly visible on the user's homepage and README. Rejected drafts that reference not-yet-exposed material are logged, never published.
The pre-publish quality gates

Every article passes all of these before publish. Any single failure blocks publication. No manual override.

01

Length

≥800 words. Randomized target per article between 1,200–3,500 to avoid formulaic footprint. Short articles fail silently — they do not publish.

02

Link integrity

Every internal URL verified against the destination site's live sitemap. Hallucinated links are stripped, not published with a warning. Count logged in the post's lineage page.

03

Fact check

Claims extracted by a second-pass model, web-verified against live sources, low-confidence claims blocked. Source citations attached to the lineage.

04

Schema validation

Article, BreadcrumbList, Person (author), and Organization (publisher) JSON-LD emitted and validated against schema.org before publish.

05

Exposure-manifest compliance

No post references a product feature, tech decision, or claim that is not already publicly visible on the user's own homepage or README. Rejected drafts are logged, never published.

06

Banned-word filter

Per-site configurable blocklist. Common slop phrases and corporate filler (“leverage”, “unlock”, “in today's fast-paced world”) can be site-banned and enforced at generation.

07

Visual-personality match

Cover image style is derived from the user's site audit (dominant colors, mood, tone). Generic fallback covers are rejected — a pastel blog never gets a dark-techy cover.

Full specification: /standards/quality-gates. Per-article results: click any post → View lineage.
The post-publish audit

After publish, each post is monitored via Google Search Console:

  • Within 2–7 days, we pull Google's indexation verdict for the URL.
  • Verdicts of Crawled, not indexed trigger a regenerate-with-quality-boost pass. The original post's lineage records the event.
  • Verdicts of Duplicate, Google chose different canonical trigger a uniqueness review and differentiation pass.
  • Verdicts indicating user-side issues (robots.txt, noindex, 5xx) are reported to the user, never silently fixed.

The result: we iterate based on Google's own judgment, not ours. Every iteration is logged and publicly viewable on the post's lineage page.

Network safety — enforcement timeline

Sites join the network only after passing a trust crawler. Admitted sites are re-audited weekly. When a site trips detection, this timeline runs — publicly, with a case number.

T+0 → T+1h
Quarantine

Cross-linking to and from the site is disabled immediately. Existing articles remain readable. The site is not deleted; it is frozen.

T+24h
Notification

The network admin receives an email with specific signal flags, evidence hashes, and a case number for appeal.

T+48h
Human review + appeal

Appeals filed at /appeal are processed by a human reviewer. Reviewer decisions are logged publicly with case number and reasoning.

On confirmed abuse
Permanent removal

Site permanently removed from network. Public entry added to /spam-index with reason codes and evidence hashes. Decision linkable, auditable, reversible.

False positives cost us more than false negatives. Detection thresholds err toward not-flagging. Every flag that gets appealed and reversed is published in the transparency report as an error we made — not hidden.

What we publish, publicly, for accountability
SignalWhere
Per-post lineage — sources, gates, model, timestamps, strip report/blog/<slug>/lineage
Flagged domains with reason codes/spam-index
Quarterly transparency report — sites admitted, quarantined, removed, appeals/transparency-report
Every quality gate specification, versioned/standards/<gate-name>
Every model change that affects content/changelog?tag=model
Policy changes to this page/standards/history
Google alignment map

Each row names a Google guideline and the exact Networkr mechanism that enforces it. This is the line-by-line proof that we build to the spec Google already publishes.

Google guidelineNetworkr mechanism
Helpful, reliable, people-first contentFact-check gate · exposure manifest · word-count floor
Scaled content abusePer-site voice uniqueness · randomized length · fact-check block · public lineage
Site-reputation abuseEvery site passes an independent trust crawl · no third-party content hosted under members' domains
Sneaky redirectsCrawl detection at admission · weekly re-audit
Link spamCross-link scoring requires topical relevance · DA-band match · complement-not-competitor classification
E-E-A-TReal named authors per site · Person schema with sameAs · optional Build News mode providing git-verifiable first-person experience
AI-generated content disclosureAI-assisted disclosure footer on every post · included in Article schema · links to content policy
Structured data accuracySchema validated against schema.org before publish · lineage logs the validation result
How to verify any of this

Pick any post from any site in our network. Open its /blog/<slug>/lineage page. You will see:

  • The research sources used
  • The model and pipeline version
  • Each quality gate's result (pass / fail + details)
  • The link-integrity report (stripped vs kept URLs)
  • The publish timestamp and any post-publish heal events

If anything on this page doesn't match what you see in a lineage page, email standards@networkr.dev. Confirmed inconsistencies are corrected within 72 hours and logged in /standards/history.

What this page doesn't promise
  • ·We do not promise ranking improvements. We promise that we do not violate Google's guidelines and that we publish the audit trail to prove it.
  • ·We do not promise zero errors. We promise public accountability when we make them.
  • ·We do not promise that every post will be indexed. We promise that we respond to Google's verdicts and iterate.
Related

Standards v1.0 · Published 2026-04-18 · Permalink to this version. See history for all prior versions.