ai-engineering-fluency: Skill for Claude Code

.claude/skills/whats-new-catalog/SKILL.md

whats-new-catalog is a skill for Claude Code from rajbos/ai-engineering-fluency. It costs 167 tokens per session (1,980 once invoked), scanned B, original, MIT.

A release-catalog checker for a VS Code extension's hand-written “What's New” page and feature notifications. It compares shipped versions and linked views with the catalog file.

In plain words
What is it for?
Use it before or after a release to check that the What's New screen and its “Take me there” buttons describe the software accurately.
Why use it?
It catches missing release notes, entries still labelled “Unreleased”, and links to views or tabs that no longer exist—problems that normal builds and tests may miss.

Skill for Claude Code

Written for Claude Code: installed under .claude/.

This is rajbos/ai-engineering-fluency's own configuration. It tells Claude Code how to work on ai-engineering-fluency itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything ai-engineering-fluency configures →

Needs its repository: it reads a path above its own folder, which exists only inside the repository. The line is Full feature documentation: [docs/features/WHATS-NEW.md](../../../docs/features/WHATS-NEW.md).

Reuse

Borrowing it

Nothing to install: this file belongs to rajbos/ai-engineering-fluency. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/rajbos/ai-engineering-fluency/main/.claude/skills/whats-new-catalog/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/rajbos/ai-engineering-fluency

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for whats-new-catalog

README.md
[![agentmods](https://agentmods.dev/badge/skills/rajbos/ai-engineering-fluency/whats-new-catalog/github.svg)](https://agentmods.dev/skills/rajbos/ai-engineering-fluency/whats-new-catalog)
Your own site
<a href="https://agentmods.dev/skills/rajbos/ai-engineering-fluency/whats-new-catalog"><img src="https://agentmods.dev/badge/skills/rajbos/ai-engineering-fluency/whats-new-catalog/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for whats-new-catalog

Your own site · 80×15
<a href="https://agentmods.dev/skills/rajbos/ai-engineering-fluency/whats-new-catalog"><img src="https://agentmods.dev/badge/skills/rajbos/ai-engineering-fluency/whats-new-catalog.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 167 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,980 The whole file, excluding the scripts and references it only reads on demand.
Security scan B 1 finding. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00167 $0.01980
Opus 5.5 $0.00067 $0.00792
Sonnet 5.5 $0.00033 $0.00396
Haiku 4.5 $0.00017 $0.00198

Measured yesterday against content hash aeeac6e34711, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade B, and why

whats-new-catalog scanned grade B with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Subtle steeringmediumPrompt injection

Instructions that bias recommendations or shape behaviour without the user noticing.

| **Tab-tracking gaps** — a catalogued `tab` on a view that never posts `viewTabOpened` | The announcer can never tell the user opened it, so it keeps offering something they already found. Invisible in review; the main
.claude/skills/whats-new-catalog/SKILL.md · 149 lines

How it starts

The opening of the file, as written. The whole thing — 149 lines — stays where its author put it; the contents beside it link to each section on GitHub.

What's New Catalog Skill

vscode-extension/src/whatsNew/catalog.ts is hand-written prose that drives two user-facing things:

  • the What's New view (the last 5 releases, in plain English), and
  • the one-a-day new-feature notifications — which surface gets pointed out, and where its "Take me there" button goes.

Because it is prose, nothing about it compiles-or-breaks. A release can ship with no entry, an entry can point at a tab that was renamed, and everything still builds and every test still passes. The view just quietly starts lying.

This skill covers the mechanical half. The editorial half — what to say — is yours, and the guidance for it is below.

Full feature documentation: docs/features/WHATS-NEW.md

The detector

node .github/skills/whats-new-catalog/whats-new-catalog.js          # human-readable
node .github/skills/whats-new-catalog/whats-new-catalog.js --json   # machine-readable

Dependency-free apart from the repo's own TypeScript — it transpiles and evaluates catalog.ts rather than pattern-matching its prose, so multi-line descriptions containing braces and quotes can never confuse it. Run pnpm install in vscode-extension/ first if it complains.

What it checks

Check Why it matters
Version coverage — package.json's version has a catalog entry Without one, everything in that release is invisible in the view and can never be announced
Date hygiene — only the newest entry may be date: null An older undated entry makes the view say "Unreleased" about something users have run for months
Orphan surfaces — every surface resolves to a real view, and a real tab on it An orphan is a "Take me there" button that silently does nothing
Tab-tracking gaps — a catalogued tab on a view that never posts viewTabOpened The announcer can never tell the user opened it, so it keeps offering something they already found. Invisible in review; the main reason this script exists

Read the full file on GitHub · 149 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. yesterday First seen · 149 lines · 167 tokens per session scan B aeeac6e34711

Subscribe to this mod's changes

whats-new-catalog is a skill published in the GitHub repository rajbos/ai-engineering-fluency (119 stars, last pushed yesterday), licensed MIT. It adds 167 tokens to every session and 1,980 once invoked, about $0.0007 per session on Opus 5.5. A static security scan graded it B with 1 finding (subtle steering). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-26.