Skip to content

Repository files navigation

Professional release notes, generated instantly with AI.

Relog is a free SaaS tool that turns the commit history of any public GitHub repository into clean, structured, and professional release notes.

Paste a repository URL, add a release title, and Relog analyzes the repository's recent commits, categorizes the changes, and generates polished release notes that you can review, edit, preview, and export.

No login. No setup. No GitHub authorization.

Live Demo


📸 Screenshots

Landing Page

Relog Landing Page

Release Generator

Relog Generator

Generated Release Notes

Relog Release Notes

Screenshots are stored in /screenshots.


🌐 Live Demo

Try Relog directly in your browser:

→ Open Relog

No installation, account, GitHub login, or API key is required.


✨ Features

🤖 AI-powered release notes

Relog analyzes repository commits and transforms technical commit history into structured release notes that are easier for users and teams to understand.

🧠 Smart categorization

Changes are automatically organized into meaningful categories:

  • Features
  • Bug Fixes
  • Performance
  • Documentation
  • Breaking Changes
  • Chores

Relog also understands Conventional Commits, including feat, fix, refactor, and breaking changes.

✍️ Edit before publishing

Generated content isn't treated as final.

Review and refine the release notes using the built-in WYSIWYG editor before exporting them.

👀 Live Markdown preview

See how the generated release notes will look while editing.

📦 Export anywhere

Export release notes as:

  • .md files
  • Clipboard content

The generated Markdown can be used for:

  • GitHub Releases
  • Changelogs
  • Product announcements
  • Blog posts
  • Internal release communication

🔐 No account required

Relog doesn't require authentication.

There is no GitHub OAuth flow and no user account system standing between the user and the product.

⚡ Minimal setup

The workflow is intentionally simple:

Repository URL → Analyze → Generate → Edit → Export


🚀 How It Works

┌──────────────────────┐
│   Public GitHub Repo │
└──────────┬───────────┘
           │
           ▼
┌──────────────────────┐
│     GitHub API       │
│  Fetch commit data   │
└──────────┬───────────┘
           │
           ▼
┌──────────────────────┐
│  Commit Processing   │
│                      │
│ Conventional Commits │
│ Change extraction    │
│ Categorization       │
└──────────┬───────────┘
           │
           ▼
┌──────────────────────┐
│      AI Model        │
│  Generate structured │
│    release notes     │
└──────────┬───────────┘
           │
           ▼
┌──────────────────────┐
│      Tiptap Editor   │
│                      │
│ Edit → Preview       │
└──────────┬───────────┘
           │
           ▼
┌──────────────────────┐
│        Export        │
│ Markdown / Clipboard │
└──────────────────────┘

🏗️ Architecture

Relog uses a Next.js App Router architecture with a clear separation between application routes, UI components, editor functionality, reusable hooks, and shared utilities.

app/
├── app/
│   └── page.tsx              # Main application
│
├── page.tsx                  # Landing page
├── layout.tsx                # Root layout
├── error.tsx                 # Error boundary
└── not-found.tsx             # 404 handling

component/
├── section/
│   └── home/                 # Landing page sections
│
├── generateForm.tsx          # Repository input & generation flow
├── markdownEditor.tsx        # Release note editor
├── markdownRenderer.tsx      # Markdown rendering
├── header.tsx
├── footer.tsx
└── ui/                       # Reusable UI primitives

hook/
└── useScrolled.tsx           # Reusable interaction hook

lib/
└── ...                       # API, AI, validation & utility logic

image/
└── logo.png

The application is organized so that:

  • UI components handle presentation and interaction.
  • Editor components handle document editing and rendering.
  • Application logic handles GitHub and AI workflows.
  • Reusable hooks encapsulate browser interaction.
  • Shared utilities keep implementation details out of presentation components.

This separation keeps the product easier to maintain and extend as new release-note workflows are introduced.


💡 Key Technical Decisions

Next.js App Router

Next.js provides the application framework, routing, server-side capabilities, and production build system in one cohesive environment.

The App Router also allows sensitive API interactions to remain on the server instead of exposing implementation details directly to the browser.

GitHub API

Relog works directly with public GitHub repositories rather than requiring users to authenticate.

This keeps the first-use experience frictionless:

Paste a repository URL and start.

The project uses Octokit for GitHub API communication.

AI SDK + Groq

AI generation is separated from the presentation layer.

The project uses the AI SDK and Groq integration to transform processed repository information into structured release notes.

This abstraction makes the generation layer easier to evolve independently from the editor and UI.

Tiptap

Generated release notes need to remain editable rather than becoming a static AI response.

Tiptap provides a structured rich-text editing model while allowing Relog to maintain Markdown-oriented workflows.

Zod

User-provided input crosses multiple boundaries in the application.

Schema validation provides explicit contracts for data before it reaches downstream logic.

No authentication

Authentication was deliberately excluded.

Relog's primary workflow doesn't require persistent user accounts, and removing authentication reduces:

  • Friction
  • Infrastructure
  • User onboarding steps
  • Data-management responsibilities

The result is a product that can be used immediately.


🧩 Challenges & Solutions

1. Turning commit history into useful release notes

Challenge

Raw commit messages are often inconsistent, overly technical, or too terse to serve as public-facing release notes.

Solution

Relog combines commit analysis, Conventional Commit semantics, categorization, and AI generation to transform repository activity into a structured document.

The AI isn't simply asked to summarize a repository. It receives relevant change information and produces output organized around meaningful release categories.


2. Supporting both AI generation and human editing

Challenge

AI-generated text shouldn't be treated as immutable output.

Users may want to change wording, remove irrelevant changes, or add additional information before publishing.

Solution

Relog places the generated content inside a WYSIWYG editor, making AI generation the starting point rather than the final step.

Generate
   ↓
Review
   ↓
Edit
   ↓
Preview
   ↓
Export

3. Keeping the workflow authentication-free

Challenge

GitHub integrations commonly require OAuth authentication, which adds onboarding friction.

Solution

Relog focuses on public repositories and uses public GitHub API endpoints.

Users only provide the repository URL they want to analyze.


4. Converting structured content into portable Markdown

Challenge

The editor operates on structured rich-text content, while users ultimately need portable release notes.

Solution

Relog maintains a Markdown-oriented workflow so generated content can move from the editor into a standard .md file or clipboard without locking users into the application.


5. Designing a fast workflow around a potentially slow AI operation

Challenge

GitHub analysis and AI generation are inherently asynchronous operations.

A poorly designed flow can make users feel that the application has stalled.

Solution

The interface is designed around a clear generation workflow with focused states and progressive interaction, keeping the user aware of what is happening rather than leaving them staring at an unresponsive form.


🔒 Privacy & Data

Relog does not require user accounts or GitHub authorization.

The application is designed around processing public repository information without requiring users to create an account.

No login. No GitHub OAuth. No unnecessary profile data.

When using any AI-powered service, users should still avoid submitting information they do not want processed by the application's configured AI provider.


🛠️ Tech Stack

Technology Purpose
Next.js 16 Application framework & routing
React 19 UI
TypeScript Type-safe development
Tailwind CSS 4 Styling
Tiptap 3 Rich-text editing
AI SDK AI integration
Groq AI model provider
Octokit GitHub API integration
React Hook Form Form management
Zod Input validation
Motion UI animations
shadcn/ui / Radix UI UI primitives
Sonner Notifications

🚀 Getting Started

Prerequisites

  • Node.js 20+
  • npm, pnpm, yarn, or Bun
  • A compatible AI provider/API key configured through environment variables

Installation

Clone the repository:

git clone https://github.com/tasha-dev/relog.git
cd relog

Install dependencies:

npm install

Create your environment file:

cp .env.example .env.local

Configure the required environment variables in .env.local.

Then start the development server:

npm run dev

Open:

http://localhost:3000

Production build

npm run build
npm run start

🗂️ Project Structure

app/
  Next.js application routes, layouts, errors, and styles

component/
  Application components and reusable UI

component/section/
  Landing page sections

hook/
  Reusable React hooks

lib/
  Application logic, integrations, and utilities

image/
  Static project assets

type/
  Shared TypeScript types

🎯 Product Philosophy

Relog is built around a simple principle:

Developers shouldn't have to manually translate every commit into release communication.

The product deliberately reduces the process to a few steps:

Paste → Generate → Refine → Export

The goal isn't to replace developers' judgment.

It's to remove the repetitive part of release-note writing so they can spend their time reviewing and communicating what actually matters.


🤝 Contributing

Contributions are welcome.

If you'd like to make a significant change, please open an issue first to discuss the proposal.

Development workflow

  1. Fork the repository.
  2. Create a feature branch:
git checkout -b feature/amazing-feature
  1. Make your changes.
  2. Commit your changes:
git commit -m "Add amazing feature"
  1. Push your branch:
git push origin feature/amazing-feature
  1. Open a Pull Request.

📄 License

This project is licensed under the MIT License.

See LICENSE for details.


🌐 Links


Made with a focus on developer experience, simplicity, and better release communication.

Relog — Because your commits deserve to shine.

About

Professional release notes. Automatically generated. Zero hassle.

Resources

Stars

1 star

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages