Back to Skills

release-notes

Generate user-facing release notes from tickets, PRDs, or changelogs. Creates clear, engaging summaries organized by category (new features, improvements, fixes). Use when writing release notes, creating changelogs, announcing product updates, or summarizing what shipped.

23,486stars2,489forksUpdated 7/11/2026

Security Assessment

Safe(100/100)
Security Score100/100

About release-notes

Release Notes Generator transforms technical inputs, such as JIRA or Linear tickets, PRDs, Git logs, or internal changelogs, into polished, user-facing release notes and changelogs. When source files are provided it reads them first; when a product URL is mentioned it can use web search to understand the product and its audience. The goal is clear, engaging summaries that lead with user benefit rather than implementation detail, suitable for announcing product updates or summarizing what shipped.

The process begins by gathering raw material and extracting, for each change, what changed, who it affects, and why it matters. Changes are then sorted into standard categories: New Features for entirely new capabilities, Improvements for enhancements to existing features, Bug Fixes for resolved issues, Breaking Changes for anything requiring user action such as migrations or API changes, and Deprecations for features being sunset. Each entry is written to lead with the user benefit, use plain language free of jargon, internal codenames, and ticket numbers, and stay to one to three sentences. Worked example transformations show how a technical note, such as implementing a Redis caching layer for dashboard endpoints, becomes a benefit-focused line about dashboards loading up to three times faster, or how a fixed race condition becomes a plain statement that orders no longer fail during high-traffic periods.

Output follows a consistent structure headed by product name and version or date, with a section per category and an explicit action-required callout for breaking changes. Screenshots or visuals are included when the user supplies them. Tone is tuned to the product's voice: professional for B2B, friendly for consumer products, and developer-focused for APIs. The result is saved as a Markdown document by default and can be converted to HTML or another format on request. The overall effect is to turn scattered, engineering-oriented records into a single readable narrative that communicates value to end users while still flagging anything that requires action on their part.

FAQ

What inputs can it work from?

JIRA exports, Linear tickets, PRDs, Git logs, and internal changelogs. If you mention a product URL, it can also use web search to understand the product and its audience.

How are the changes categorized?

Into New Features, Improvements, Bug Fixes, Breaking Changes (anything requiring user action), and Deprecations (features being sunset).

What writing principles does it follow?

Lead with the user benefit rather than the technical change, use plain language without jargon, codenames, or ticket numbers, and keep each entry to one to three sentences.

Can it match my product's tone of voice?

Yes. It adjusts tone to the product, using a professional style for B2B, a friendly style for consumer products, and a developer-focused style for APIs.

What format is the output?

A Markdown document by default, structured with a header and per-category sections. It can be converted to HTML or another format if you ask.

Install release-notes

Download and extract the skill files to your .claude/skills/ directory.

Quick Setup:

  1. Copy the skill folder to .claude/skills/
  2. Claude will automatically detect and use the skill