菜单
Scans last-reviewed meta tags, flags stale pages, and generates a freshness report — tuned for cruise content with seasonal awareness
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
| name | content-freshness |
| version | 1.0.0 |
| description | Scans last-reviewed meta tags, flags stale pages, and generates a freshness report — tuned for cruise content with seasonal awareness |
| type | audit |
| triggers | [{"command":"/freshness"},{"session_start":true}] |
Purpose: Identify stale content before visitors or AI crawlers encounter outdated information.
Integrates with: ICP-2 (which requires last-reviewed meta tags on all pages).
Soli Deo Gloria
When triggered by /freshness or at session start, perform the following audit across all HTML pages in the repository.
last-reviewed Meta TagsSearch every .html file for:
<meta name="last-reviewed" content="YYYY-MM-DD">
For each page, extract the date and compute days since last review relative to today.
Missing tags: Flag any HTML page that lacks a last-reviewed meta tag entirely — this is an ICP-2 compliance gap.
The default staleness threshold is 90 days. The user may override this (e.g., /freshness --threshold 60).
Flag every page where days_since_review > threshold.
Not all stale pages are equally urgent. Rank flagged pages using this priority order:
| Priority | Content Type | Detection | Why |
|---|---|---|---|
| P0 — Critical | Port pages (~388) | Path contains /ports/ or filename matches port patterns | Cruise lines change itineraries seasonally; stale port info causes booking confusion |
| P1 — High | Ship profiles (~298) | Path contains /ships/ or filename matches ship patterns | Ship amenities, deck plans, and deployments change regularly |
| P2 — High | Interactive tools & calculators | Path contains /tools/, /calculators/, or pages with calculator/form elements | Price data goes stale; wrong numbers erode trust |
| P3 — Medium | Seasonal content | See Step 4 below | Time-sensitive by definition |
| P4 — Normal | General articles & guides | Everything else | Still matters, but lower blast radius |
Within each priority tier, sort by staleness (most days since review first).
InTheWake has seasonal cruise content that must be reviewed before each season begins, regardless of the 90-day threshold.
| Season | Content Pattern | Must Be Fresh By |
|---|---|---|
| Alaska Summer | Pages referencing Alaska, Inside Passage, Glacier Bay, Juneau, Ketchikan, Skagway | April 1 |
| Caribbean Winter | Pages referencing Caribbean, Bahamas, Eastern/Western Caribbean | October 1 |
| Mediterranean Summer | Pages referencing Mediterranean, Greek Isles, Adriatic | March 1 |
| Holiday Sailings | Pages referencing holiday cruises, Christmas, New Year's | October 15 |
| Repositioning | Pages referencing transatlantic, repositioning cruises | March 1 and September 1 |
Logic: If today's date is within 60 days before a season's "Must Be Fresh By" date, flag any matching page that has not been reviewed since the previous season ended. These get elevated to P3 even if they are within the normal 90-day window.
For port and ship pages, check whether the content reflects the current cruise season:
last-reviewed date is older than 6 months, flag as price data potentially stale.Output a structured report in this format:
=== CONTENT FRESHNESS REPORT — InTheWake ===
Generated: [today's date]
Threshold: [N] days
Total HTML pages scanned: [count]
Pages missing last-reviewed tag: [count]
Pages within threshold: [count]
Pages STALE: [count]
Seasonal alerts: [count]
--- P0: CRITICAL (Port Pages) ---
[filepath] — last reviewed [date] ([N] days ago)
[filepath] — last reviewed [date] ([N] days ago)
...
--- P1: HIGH (Ship Profiles) ---
...
--- P2: HIGH (Interactive Tools) ---
...
--- P3: MEDIUM (Seasonal Alerts) ---
[filepath] — last reviewed [date] — SEASONAL: [season name] starts [date], review needed by [date]
...
--- P4: NORMAL (General Content) ---
...
--- MISSING TAGS (ICP-2 Compliance Gap) ---
[filepath] — no last-reviewed meta tag found
...
After the report, provide a brief summary:
last-reviewed tags.Users can customize behavior:
/freshness — Run with defaults (90-day threshold, all priorities)./freshness --threshold 60 — Use a 60-day threshold./freshness --priority P0 — Only show Critical items./freshness --seasonal-only — Only show seasonal alerts./freshness --missing-only — Only show pages missing last-reviewed tags.Soli Deo Gloria
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.