Content Freshness Updater

What it does

Scans older published posts for dated language, specific years, "current" or "latest" claims, statistics, and references to tools or prices, flagging each with why it may be stale and what to check, without silently changing any factual claim on its own. It prioritizes posts by a mix of age and traffic so the highest-impact updates surface first.

Install

This skill isn't packaged in an installable registry. It's a plain SKILL.md file you can copy directly. Copy it from the SKILL.md tab and save it at the path for your assistant:

Use it in your AI assistant

Claude Code

.claude/skills/content-freshness-updater/ (or ~/.claude/skills/content-freshness-updater/ for all projects)

GitHub Copilot (VS Code)

.github/skills/content-freshness-updater/

OpenAI Codex

.codex/skills/content-freshness-updater/

Cursor

.cursor/skills/content-freshness-updater/ (or ~/.cursor/skills/content-freshness-updater/ globally)

Compatibility

Any WordPress site with posts older than about a year. No special plugin required; benefits from analytics access to prioritize by traffic, but works without it using publish date alone.

The complete SKILL.md for this skill:

---
name: content-freshness-updater
description: Use when a user wants to audit old posts for outdated stats, dates, prices, or references before a refresh. Triggers on "check this post for outdated info", "what needs updating on this old post", or "find stale content on the site".
---

# Content Freshness Updater

## When to use
Posts published more than roughly a year ago need a check for stale statistics, outdated year references, superseded tool or product mentions, or pricing that has likely changed.

## Procedure
1. Pull a list of posts past a given age threshold (default 12 months), sorted by traffic if analytics access is available, otherwise by age.
2. For each candidate post, scan for specific years ("in 2024"), phrases like "currently" or "the latest version", named tools or products, prices, and statistics with a citation or source.
3. Flag each finding with the exact text, why it may be stale, and what to verify (a new version release, a price change, an updated statistic), rather than guessing a replacement value.
4. Do not change any factual claim automatically; present the flagged list for the user to verify and update themselves, since only they can confirm the current correct value.
5. Note the post's last-updated date so the user can decide whether a full refresh or a targeted patch is the right scope.

## Guardrails
- Never invent an updated statistic, price, or version number to replace a stale one; flag it for the user to verify against a real source.
- Do not flag evergreen phrasing that only looks dated (a post titled "2024 Guide" that is explicitly about that year historically) as an error; distinguish intentional dating from accidental staleness.