Show HN: Conventional Comments in GitHub

https://github.com/pullpo-io/conventional-comments

2•francesc_holly•9mo ago

Hey HN, Cesc here, co-founder at Pullpo.

We spend a lot of time doing code reviews on GitHub. One recurring frustration was deciphering ambiguous comments. Misunderstandings slowed us down.

We're big fans of the https://conventionalcomments.org (discussed previously here: https://news.ycombinator.com/item?id=23009467) standard for adding clarity, but remembering and typing the prefixes (suggestion, issue(blocking), etc.) felt like friction.

So, we built a simple, free, open-source Chrome extension to make using this standard effortless within the GitHub UI.

How it works:

• It adds a small toolbar above GitHub comment boxes.

• You click buttons for labels (issue, suggestion, praise, nitpick, etc.) and optional decorators (blocking, non-blocking, if-minor).

• It automatically formats the comment prefix for you.

• There's a "Prettify" option to display prefixes as visual badges (using Shields.io, linked to a simple explainer on pullpo.io).

• It adapts to GitHub light/dark themes.

We built it because we needed it ourselves to improve our internal review process, and thought others might find it useful too. It's completely free and open-source (MIT license).

• Install Link -> https://chromewebstore.google.com/detail/gelgbjildgbbfgfgpib...

• GitHub Repo -> https://github.com/pullpo-io/conventional-comments

• Quick Demo Video -> https://youtu.be/jLzXlZ78rNE?si=KMzIH9Vb43glekEW

We just launched it on the Chrome Web Store. Would love to hear your feedback, suggestions, or any pain points you have with code review comments! Thanks, Cesc

Comments

badmonster•9mo ago

How does the extension hook into GitHub's DOM to inject the comment toolbar, and does it support dynamically loaded elements like in PR reviews with infinite scroll?

francesc_holly•9mo ago

Chrome extensions let you execute JS on certain sites, so we can querySelect all target textareas and insert the toolbar anywhere. And yes! That includes dynamically loaded elements like the PR review window. We can even apply light/dark themes that match GitHub's ;)

If you want specifics you can check out the open-source repo here: https://github.com/pullpo-io/conventional-comments

I suggest looking at `content.js`, constant `TARGET_TEXTAREA_SELECTORS` and function `initializeToolbarForTextarea`.

"Those rights don't count": Bovino says Pretti forfeited 2nd Amendment rights

OnePlus update blocks downgrades and custom ROMs by blowing a fuse

You can't parse HTML with regex

I'm posting this from a memory safe web browser

Bending Spoons, the little-known firm behind Vimeo's layoffs

The US Attorney General will remove ICE if voter files handed over

Proof Pack: deterministic replay and audit chain and PASS/FAIL verifier

PostgreSQL Shared Buffers Visualized

Your Aggregate Is Not a Table

Vimeo's Slow Fade: An Engineer's Front-Row Seat to the Fall of a Web Icon

The FDA Listened on Estrogen. Will They Now Fix Testosterone's Outdated Label?

Can AI Predict Stories? Learning to Reason for Long-Form Story Generation

If NotebookLM was a web browser

Centuries-Old Longevity Elixir Reveals Gaps in Modern Medicine

The Podcaster Poking at France's Biggest Secrets

Yes, It's Fascism

Show HN: QR Code Editor with static and dynamic QR codes (client-side first)

Show HN: Free database of NYC startups hiring (updated weekly)

Move Faster

The Goal Gradient Effect: Progress is a hell of a drug

Bill Gates-backed startup aims to revive Moore's Law with optical transistors

Show HN: Oura (Activity Tracker) MCP Server with Claude

Water-Mills and Wheels

Introduction to Mechanistic Interpretability – By Sarah

Show HN: AI-rganize – CLI tool for organizing your files

Setapp denied Launchpad replacement for not using AI. This is not good

Tired of Manual Priority Weights in Proxies? I Used a Reverse Radix Tree

Seemore: Implement a Vision Language Model from Scratch

General Theory of Neural Networks – Rob Leclerc

Pure Mojo implementation of moonshine ASR model outperform PyTorch+ Keras by 6x