Changeset logo

Changeset

OrganizationPopular
reactive
changeset

Create user-focused changesets (changelog entries) for semver bumps, release notes, breaking changes, and docs; prefer impact and code examples over implementation detail

Overview

Publisherreactive
Repositorydata-client
Skill namechangeset
Stars
2K
Forks
98
Bundled files
Instructions only
LicenseApache-2.0
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 reactive on GitHub. Read the source before you install it.

Installation

Install the Changeset 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/reactive/data-client/tree/master/.agents/skills/changeset
  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/reactive/data-client.git /tmp/data-client
mkdir -p .claude/skills
cp -r /tmp/data-client/.agents/skills/changeset .claude/skills/changeset
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Changeset 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 Changeset 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 Changeset 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.

Create Changesets

Overview

Generate changesets, update documentation, draft blog entries, and update skills for user-facing changes.

Steps

  1. Analyze changes

    • Compare current branch to master to identify modified packages
    • Group changes by impact type (feature, fix, breaking)
    • Identify transitive dependencies that expose the change
    • Check for modified skills in .agents/skills/ and packages/*/.agents/skills/
  2. Determine affected packages

    • Include directly modified packages
    • Add transitive dependents when interfaces/behaviors are exposed:
      • @data-client/endpoint → also select @data-client/rest, @data-client/graphql
      • @data-client/core or @data-client/normalizr → also select @data-client/react, @data-client/vue
  3. Run changesets

    • Run yarn changeset once per distinct change
    • Select all affected packages (direct + transitive)
    • Choose appropriate version bump (patch/minor/major)
      • For packages under 1.0, use minor for breaking changes
    • When writing the markdown body, follow Changeset body quality below (user-visible outcome, code examples when helpful, no internal implementation narrative)
  4. Update documentation

    • Update primary docs in docs/ for any changed public APIs
    • Search for all usages of changed APIs across docs/, packages/*/README.md, and root README.md
    • Update all found references to reflect new behavior, signatures, or deprecations
    • For new exports: add to relevant API reference pages
  5. Update release blog

    • Find the latest draft blog post in website/blog/ (has draft: true in frontmatter)
    • Add the change following @website/blog/.cursor/rules/blog-posts.mdc:
      • Place in appropriate conceptual category section
      • Include code example if applicable
      • Link to PR/commit and relevant docs
      • For breaking changes: add to summary's Breaking Changes section with anchor link
  6. Update skills

    • If code changes affect workflows described in existing skills, update those skills to match
    • If new APIs or patterns are introduced that agents should know about, add them to the relevant skill
    • Skill changes don't need changesets — they are development tooling, not published packages

Writing perspective

All user-facing text (changesets, blog entries, docs) should be written from the library user's point of view. Answer: what did the user see go wrong, and what works for them now? Avoid internal names (conditional types, branch names, helper types like SoftPathArgs, file paths, PR numbers) unless the audience is maintainers reading a technical appendix — changeset bodies are for consumers reading the changelog.

Changeset body quality

  1. Lead with impact — One short title line, then 1–3 sentences on behavior: errors gone, typings improved, new capability, migration note.
  2. User vocabulary — Name public APIs (RestEndpoint, resource(), hook names). Do not explain how the fix was implemented.
  3. When to add code — Prefer a minimal example when the change is TypeScript-only or subtle: show the pattern that was broken and now works (subclass, extend, option object). Skip examples for trivial renames or obvious one-line fixes.
  4. Examples — Realistic imports and types; omit unrelated options. For fixes, you can show one “now types correctly” snippet instead of a long before/after if the before state was “TypeScript error on …”.
  5. Breaking changes — Still say what the user must do; use Before/After sections with code when the migration is non-obvious. Before marking anything breaking, follow .cursor/rules/breaking-changes.mdc (ship a compatible version, recommend the future-proof form, track the cleanup).

Changeset format

  • First line: Action verb ("Add", "Fix", "Update", "Remove")
  • Breaking: Prefix with BREAKING CHANGE: or BREAKING:
  • Body: User outcome first; implementation almost never belongs here
  • New exports: Use "New exports:" with a bullet list

Code examples in changesets

  • Fixes: Optional // Before: / // After: comments in one block, or two small blocks — keep them copy-paste plausible
  • Breaking changes: Use #### Before and #### After headers with complete snippets
  • Multiple scenarios: Short intro line per scenario, or separate fenced blocks with a one-line label above each

Markdown Formatting

Follow @.cursor/rules/markdown-formatting.mdc for all markdown content.

Frequently asked questions

What does the Changeset AI skill do?

Create user-focused changesets (changelog entries) for semver bumps, release notes, breaking changes, and docs; prefer impact and code examples over implementation detail

Why use Changeset on TypingMind?

Because you install it once and use it with any model. Changeset 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 Changeset in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/reactive/data-client/tree/master/.agents/skills/changeset. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Changeset?

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 Changeset?

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

Is the Changeset AI skill free?

Yes. It is published on GitHub by reactive under the Apache-2.0 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 👇