InTheWake

Use when about to claim work is complete, fixed, or passing, before committing or creating PRs - requires running verification commands and confirming output before making any success claims; evidence before assertions always

The dedication under which all work in this household is done — every task offered to God, excellence as worship. Automatic posture skill, present and byte-identical in every repository and sibling workspace. Preserve the invocation; getting it right is the worship, not getting it fast.

Hard ban on any reference to the external domain getbets.us on cruisinginthewake.com. Enforced via PreToolUse hook (.claude/hooks/no-getbets-guard.js) and git pre-commit hook (.githooks/pre-commit). Fires automatically on every Write/Edit/MultiEdit and every commit; no manual invocation.

Voice & Presence — fires during cruise content writing. Guards the sound of the prose so it reads as written by someone who has actually sailed the route, not by a brochure or a model. Shapes voice markers, rhythm, vocabulary, and pastoral honesty. For post-draft diagnostics, see voice-audit.

Post-draft diagnostic for InTheWake content. Scans for cruise-marketing tells, AI authorship-cluster signals, and absence-of-authenticity gaps. Assesses authenticity risk by clustering signals across layers (Layer 1 strong, Layer 2 supporting, Layer 3 counter-signals). Required: every voyage-pack ships with a voice-audit attestation in its .factcheck.json sidecar. Fires before committing content edits or before publishing a new page. For during-writing standards, see like-a-human. For corpus measurement, see voice-dna.

External red-team critique pass. Sends a commit range or proposed change to a second model (defaulting to a different LLM) with the explicit charter to find every flaw "as though you're trying to take the engineer's job." Grok-native usage notes.

Foundational integrity guardrail. Automatic on every file modification and any non-trivial task. Demands verified claims with specific evidence, explicit scope, anomaly dispositions, and rejection of narrow or vague assertions. The single most important discipline on this project. Grok-native adaptation.

Prevents destructive, high-risk, or integrity-violating operations on the InTheWake project. Grok-native version emphasizing explicit confirmation, reversible-by-default patterns, and refusal of "just this once" rationalizations.

Factual claim discipline. Every factual claim in shipped content (voyage packs, ship pages, port pages, restaurant pages) must trace to a primary source verified in-session, recorded in a .factcheck.json sidecar. Blocks three confabulation modes: training-data echo, copy-propagation across files, and superlative-by-default. Pre-commit hook gates voyage-pack .md commits on sidecar presence + freshness. Peer to careful-not-clever, not subordinate.

Discovers voice patterns from the InTheWake content corpus. Measures rhythm, vocabulary, and experiential fingerprints across port guides, ship profiles, logbook entries, and restaurant pages — data-driven voice profiling that feeds like-a-human and voice-audit.

Generates new cruise port pages to gold standard (dubai.html). Enforces ICP-2 v2.1, LOGBOOK_ENTRY_STANDARDS v2.300, and all 21 audit detections.

Sends a commit range (default: origin/main..HEAD) to an external LLM with the framing "find every little thing wrong as though you're trying to take the engineer's job." Returns a structured findings report under audit-reports/. Designed as the external-audit backstop named in careful-not-clever v1.8.1-alpha §"The limit of this rule" — the rule acknowledges that an author cannot reliably catch their own claim- evidence gaps, and points at an external pass as the real check.

Blocks cross-page image reuse on this site. Same image bytes may not back two different ships, two different ports, two different sections, or two different entities of any kind. The Cordelia-on-Carnival pattern (an image of the wrong ship served as the right ship's hero) cost this project 194 manual repairs in v2; this skill exists to make sure that never happens again.

Pre-publication quality gate that evaluates reader emotional experience. Complements voice-audit (mechanical tells) and like-a-human (during-writing standard) with feeling-level assessment. Applied during content review before publishing any page.

Per-page repair of internal numeric inconsistencies — when a single ship page lists 2+ different guest-count numbers in non-'maximum' contexts. Resolves to passengers_double_occupancy per Policy 0.2 of admin/POLICY_DECISIONS.md. Voice-preserving: only the integer changes. Companion to validator rule DATA-004.

Rewrites <meta name="ai-summary"> tags that ICP-018 flags as boilerplate. Each rewrite carries 2 ship-specific facts (class/year/tonnage/distinctive feature) + 1 voice-aligned editorial line, ≤80 words, no boilerplate phrases. Companion to validator rule ICP-018.

Enforces atomic commits, checkpoint summaries, rate-limit recovery, and safe resume patterns. Writes session pulses + cognitive memory so interrupted sessions can restart cleanly.

Auto-submit pages to IndexNow when created or edited. Instantly notifies Bing, Yandex, Naver, Seznam, and Yep — no waiting for crawlers.

Deep multi-LLM research pipeline that produces validated ship and port pages. 4 phases: reconnaissance → triage → deep research → synthesis → page generation.

Researches and writes authentic logbook entries for cruise venue pages. Handles review research, negative flips, fact verification, voice consistency, and authenticity checks across 5 cruise lines (RCL, NCL, Virgin, MSC, Carnival). Fires when editing restaurant/venue pages or writing logbook content.

Guides GA4 setup, conversion tracking, and UTM parameters for cruisinginthewake.com. Focuses on key metrics that matter for a cruise planning site.

Deep audience personas for InTheWake — who reads this site, what they need, what they fear, what language they use. Feeds content creation, tool design, and port guide writing.

Final typographic polish before publishing. Checks curly quotes, bullet consistency, em-dash usage, broken links, missing alt text, and visual consistency across pages.

Round-robin multi-LLM consultation tuned for InTheWake PWA development. GPT proposes, Gemini refines, Grok challenges — full-context debate with wheat/chaff justifications, low-hanging fruit, and PWA checklist.

Playwright-based test scenarios for InTheWake's 9 interactive JavaScript tools. These calculators handle real money decisions for real travelers.

Validates deployment artifacts (sitemap.xml, manifests, schemas, llms.txt) are consistent with actual site pages. Catches orphaned files, missing entries, and stale caches.

Plans content updates before each cruise season. Knows Alaska/Caribbean/Mediterranean schedules, identifies stale content per season, and generates task lists tied to the calendar.

Cross-repository cognitive memory system with semantic search. Persists knowledge across sessions using TF-IDF recall, memory versioning, knowledge graph edges, and confidence decay. Memory is cognition, not storage.

Validates internal links, anchor references, image paths, and cross-page connectivity across the cruise planning site. Catches broken links before visitors do.

Generates and updates cruise port pages from POI schema data, enforcing site standards, voice, and accessibility requirements. Knows the 388-port structure and content patterns.

Audits WCAG 2.1 AA compliance across cruise content pages, interactive tools, and port guides — validates alt text, color contrast, heading hierarchy, ARIA labels, keyboard navigation, and screen reader compatibility

Scans last-reviewed meta tags, flags stale pages, and generates a freshness report — tuned for cruise content with seasonal awareness

Validates JSON-LD structured data, Open Graph, Twitter Cards, and AI meta tags for cruise and travel content

Quick multi-LLM second opinion. Sends a single prompt to GPT, Gemini, or Grok with a role-based system prompt and returns structured feedback.

Full multi-LLM pipeline orchestration. Runs the cruising pipeline that coordinates GPT, Gemini, and Grok as consultants while Claude remains lead developer. Auto-detects cruising mode from this repository.