Releasenotes logo

Releasenotes

OrganizationPopular
exceptionless
releasenotes

Evaluate semantic-version impact and generate formatted changelogs from git history since the last release tag. Use when preparing or creating a release, choosing the next version, or drafting release notes that categorize breaking changes, features, fixes, and other changes.

Overview

Publisherexceptionless
RepositoryExceptionless
Skill namereleasenotes
Stars
2.5K
Forks
508
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 exceptionless on GitHub. Read the source before you install it.

Installation

Install the Releasenotes 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/exceptionless/Exceptionless/tree/main/.agents/skills/releasenotes
  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/exceptionless/Exceptionless.git /tmp/Exceptionless
mkdir -p .claude/skills
cp -r /tmp/Exceptionless/.agents/skills/releasenotes .claude/skills/releasenotes
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Releasenotes 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 Releasenotes 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 Releasenotes 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.

Generate a changelog for all changes from the most recent release until now.

Mandatory release gates

These gates apply even when the user initially asks to create or publish a release and supplies the version.

  1. Start from freshly fetched main. Before choosing the release range, generating notes, or selecting a target SHA, run git fetch origin --prune --tags. Stop if the fetch fails. Resolve the release endpoint from refs/remotes/origin/main; never use the current checkout, local main, HEAD, a previously cached origin/main, or an older commit merely because it is already available locally.
  2. Show the exact notes and stop. Present the complete release notes to the user in a markdown code block, then end the turn without creating a tag, draft release, or published release. The initial release request, a supplied version, or approval of the production-deployment side effect does not approve unseen release notes. Only a subsequent user message approving the displayed notes authorizes publication.
  3. Re-fetch immediately before publication. After the notes are approved and immediately before creating the tag or GitHub release, run git fetch origin --prune --tags again and resolve refs/remotes/origin/main again. Stop if the fetch fails.
  4. Invalidate approval when main moves. Compare the freshly fetched origin/main SHA with the SHA used for the approved release range. If they differ, do not publish. Inspect the new commits, regenerate the notes, show the revised notes, and wait for a new subsequent approval.
  5. Target and verify latest main. Create the release with the exact SHA from the final successful fetch, for example gh release create <tag> --target <sha> --notes-file <file>. After creation, verify that the release tag resolves to that SHA. Do not claim success if the target verification fails.

Any change to the version, notes, or target SHA invalidates the prior approval.

Steps

  1. Complete the initial latest-main fetch gate above.
  2. Find the most recent release tag using git tag --sort=-creatordate and get commits and merged PRs through the freshly fetched refs/remotes/origin/main.
  3. Look at previous releases in this repo to match their format and style
  4. Evaluate every change and classify the highest semantic-version impact in the range:
    • Patch: Bug fixes, maintenance, internal refactors, documentation, minor UI polish, and small improvements that do not add a meaningful new user-facing capability.
    • Minor: Any substantive user-facing feature or meaningful new capability. Small improvements may remain patch-level when they do not materially expand what users can do.
    • Major: Breaking changes, deliberately incompatible behavior, or unusually large features that materially redefine the product or its public contracts.
  5. Use the highest-impact change to recommend the next version. Mixed ranges take the highest classification.
  6. If the range contains any substantive feature, breaking change, unusually large feature, or ambiguous version impact, explain the classification and recommend a minor increment for normal features or a major increment for breaking changes and unusually large features. If the user already supplied a version, honor it and state any mismatch with the recommendation without asking them to repeat or reconfirm it. Otherwise, ask what the next version should be before assigning a version, creating a tag, or publishing a release.
  7. If the range is patch-only, use the next patch version unless the user supplied a different version or the repository follows another established versioning scheme.
  8. Categorize changes into sections: Breaking Changes, Added, Changed, Fixed, Notes
  9. Focus on user-facing changes, important bug fixes, and breaking changes
  10. Include PR links and contributor attribution
  11. Complete the notes-review stop above. Publication is a later-turn action that must also pass the final latest-main gates.

When version impact is uncertain, ask rather than silently choosing the smaller increment. Do not treat commit labels alone as authoritative; evaluate the actual behavior described by the changes.

Output

State the recommended semantic-version impact and why. If a version decision is required, present the changelog with the version marked as pending and ask for the version. Otherwise, present the patch-version changelog in a markdown code block for review. Always end the drafting turn after presenting the notes; never publish in the same turn.

Frequently asked questions

What does the Releasenotes AI skill do?

Evaluate semantic-version impact and generate formatted changelogs from git history since the last release tag. Use when preparing or creating a release, choosing the next version, or drafting release notes that categorize breaking changes, features, fixes, and other changes.

Why use Releasenotes on TypingMind?

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

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/exceptionless/Exceptionless/tree/main/.agents/skills/releasenotes. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Releasenotes?

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 Releasenotes?

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

Is the Releasenotes AI skill free?

Yes. It is published on GitHub by exceptionless 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 👇