Subwave News Dispatch logo

Subwave News Dispatch

CommunityPopular
perminder-klair
subwave-news-dispatch

Write a news post for the SUB/WAVE "Dispatches" page, a short human-friendly tutorial about a feature, fix, or release. Use this skill whenever the user wants to "write a news post / dispatch / news article", "announce a feature on the news page", "add a changelog entry to the site", "post an update about <feature>", "write release notes for the site", or "tell people about <what shipped>" for SUB/WAVE. The skill knows where the markdown lives (web/content/news), the frontmatter schema, the what/how/why structure, and that every draft must be run through the `no-ai-slop` skill before saving. It does NOT publish or deploy; it just writes the .md file, and the /news page picks it up on the next build.

Overview

Publisherperminder-klair
Repositorysubwave
Skill namesubwave-news-dispatch
Stars
1.4K
Forks
122
Bundled files
Instructions only
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by perminder-klair on GitHub. Read the source before you install it.

Installation

Install the Subwave News Dispatch AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/perminder-klair/subwave/tree/develop/.claude/skills/subwave-news-dispatch
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/perminder-klair/subwave.git /tmp/subwave
mkdir -p .claude/skills
cp -r /tmp/subwave/.claude/skills/subwave-news-dispatch .claude/skills/subwave-news-dispatch
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Subwave News Dispatch in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Subwave News Dispatch on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Subwave News Dispatch is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

SUB/WAVE news dispatch

Write one news article for the Dispatches page (/news). Each dispatch is a short, plain-English mini-tutorial about something that shipped: what it is, how to use it, and why it helps. Not a changelog line.

Why this skill exists

The news page reads markdown straight off disk (web/lib/news.ts reads web/content/news/*.md), so adding a post is just writing one file in the right shape. The easy-to-get-wrong parts are the frontmatter schema, keeping the tone human (short tutorial, not release-note jargon), and remembering to humanize the draft. This skill encodes all three.

Where the file goes

web/content/news/YYYY-MM-DD-slug.md
  • The YYYY-MM-DD- prefix is only for sorting files on disk; it's stripped from the URL. The page slug is the frontmatter slug if present, otherwise the filename minus that date prefix. So 2026-05-30-clone-a-voice.md serves at /news/clone-a-voice.
  • Pick the date from when the feature actually shipped. For a real commit: git show -s --format=%cs <sha>. Evergreen "this is what SUB/WAVE does" posts use the Spotlight category and any sensible recent date.
  • The eight seed articles already in web/content/news/ are the reference set. Match their length and voice.

Frontmatter schema

yaml
---
title: Make your DJ sound like anyone     # the headline; sentence case, no trailing period needed
date: 2026-05-30                          # ISO yyyy-mm-dd; drives ordering (newest first)
category: Feature                         # Release | Feature | Fix | Announcement | Spotlight
version: v0.2.0                           # optional, shows as a dateline tag
author: The SUB/WAVE desk                 # optional, byline on the article page
excerpt: One or two sentences shown on the index card and used as the meta description.
---
  • category describes the nature of the post, not the raw conventional-commit type. A fix(web) commit that listeners will notice is usually a Feature or Fix dispatch; a chore: release is a Release; an evergreen feature explainer is a Spotlight.
  • excerpt is required in practice. It's the deck on the index and the meta description / OG description. Keep it concrete.
  • Watch the YAML. title and excerpt are parsed as YAML, so a colon-then-space inside the value (a new bit: a moon phase) makes the parser read it as a nested key and the build fails. Avoid raw : in those values (reword, or use a comma), or wrap the whole value in double quotes. The same goes for a value that starts with a quote, [, {, @, or `.

Structure of the body

Aim for ~150-350 words. Three beats, in this order:

  1. What's new. One or two lines. What shipped, in plain words.
  2. How to use it. The concrete steps. Real commands, the exact filename/folder, where to click in admin. Use fenced code blocks for commands and paths. This is the part readers came for.
  3. Why it helps. The benefit, in one short paragraph. What it lets them do now that they couldn't before.

Use ## subheads (sentence case) to break those up. Look at the seed articles for the rhythm. The first paragraph gets an automatic vermilion drop cap, so open with prose, not a heading.

Writing rules

  • No em-dashes. Reword with a comma, a colon, parentheses, or a full stop instead. This is a hard rule for the wire.
  • Short sentences. Vary the rhythm. First person ("we traced it to the Opus mount", "nothing for you to do") is fine and reads human.
  • No changelog-speak, no inflated significance, no rule-of-three padding.
  • Be specific: real paths, real commands, the real button name in admin. Pull the detail from git log, the PR, or the relevant manual page under web/components/manual/.
  • Straight quotes, sentence-case headings, no decorative emoji.

Required: de-slop before saving

ALWAYS run the finished draft through the no-ai-slop skill in edit mode, apply its edits, then write the .md file. This is not optional. It's the step that keeps the wire from reading like AI release notes, and it's where the em-dash, throat-clearing opener, and rule-of-three habits get caught.

Two things to hold onto when you apply its edits: the no-em-dash rule above is absolute here even though no-ai-slop allows 1-2 in longer drafts, and the wire's first-person voice ("we traced it to the Opus mount") is a voice signal to preserve, not throat-clearing to cut.

Verify

bash
cd web && npm run lint && npm run build
  • Confirm the new slug appears under ● /news/[slug] in the build output (it's statically prerendered). A frontmatter typo shows up here as "Failed to collect page data for /news/[slug]".
  • The article shows on /news automatically, newest first, and the sitemap (web/app/sitemap.ts) picks it up via getNewsSlugs(). No other file to touch.
  • Optional spot check: cd web && npm run dev, open http://localhost:7700/news, confirm the headline, dateline, and body render.

Worked example (abridged)

File: web/content/news/2026-05-30-clone-a-voice.md

markdown
---
title: Make your DJ sound like anyone
date: 2026-05-30
category: Feature
version: v0.2.0
author: The SUB/WAVE desk
excerpt: PocketTTS can now clone a voice from a single WAV. Drop a clip in the voices folder, point a persona at it, and that's the voice on air.
---

Voice cloning used to be a Chatterbox-only trick. PocketTTS can do it too now...

## What's new

PocketTTS does zero-shot cloning from one reference WAV...

## How to use it

Drop the `.wav` into the voices folder:

state/voices/morning-host.wav


Then open a DJ persona in admin and set its voice to that filename. Save.

## Why it helps

Your station gets a real on-air identity...

Frequently asked questions

What does the Subwave News Dispatch AI skill do?

Write a news post for the SUB/WAVE "Dispatches" page, a short human-friendly tutorial about a feature, fix, or release. Use this skill whenever the user wants to "write a news post / dispatch / news article", "announce a feature on the news page", "add a changelog entry to the site", "post an update about <feature>", "write release notes for the site", or "tell people about <what shipped>" for SUB/WAVE. The skill knows where the markdown lives (web/content/news), the frontmatter schema, the what/how/why structure, and that every draft must be run through the `no-ai-slop` skill before saving...

Why use Subwave News Dispatch on TypingMind?

Because you install it once and use it with any model. Subwave News Dispatch is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Subwave News Dispatch in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/perminder-klair/subwave/tree/develop/.claude/skills/subwave-news-dispatch. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Subwave News Dispatch?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Subwave News Dispatch?

As many as you like. As long as a model supports skills, you can use Subwave News Dispatch with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Subwave News Dispatch AI skill free?

Yes. It is published on GitHub by perminder-klair under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇