Weekend Chrome Extension for LinkedIn Comment Suggestions

Why LinkedIn Comment Suggestions Matter

When I walked into a networking event and heard everyone talk about “networking on LinkedIn, ” something clicked for me: I could build a tool that gives LinkedIn comment suggestions, letting introverts like me show up in feeds without the pressure of crafting the perfect line. I’m a tech person who started in sales, and even though I could “perform” in a room, real-life conversations still drain me. The idea was simple , if talking in rooms is hard, why not let a tiny piece of software do the talking for me in the feed?

The Problem: Introverts & the LinkedIn Feed

Most of us who prefer listening over speaking have a love-hate relationship with LinkedIn. On one hand, the platform is a goldmine for visibility, job opportunities, and learning. On the other, the act of commenting feels like stepping onto a stage without a script. I’ve seen countless students and Gen-Z professionals stare at a post, type a generic “Great post! 🔥”, delete it, and then move on. The result? Low engagement, missed connections, and a feeling that you’re not showing up authentically.

Building the MVP in One Weekend

I set a deadline: Saturday morning to Sunday night. The stack was straightforward , a Chrome extension that injects a small UI under every LinkedIn post and a tiny backend that returns a few comment options. Here’s the rough workflow I followed:

1. Scaffold the extension , manifest.json with the required permissions (activeTab, https://www.linkedin.com/*).
2. Content script , runs on each LinkedIn feed page, finds the post container, and appends a Suggest button.
3. Popup UI , when the button is clicked, a small overlay appears with three pre-generated comment snippets.
4. Simple AI , I used OpenAI’s gpt-3.5-turbo via a lightweight serverless function (Vercel) to generate human-like suggestions based on the post’s text.
5. Copy-to-clipboard , one click copies the chosen suggestion into the comment box, ready for a quick edit or direct post.

The whole thing took about 12 hours of coding, testing, and tweaking the prompt to keep the tone conversational rather than corporate.

How the Extension Works (Step-by-Step)

• Detect the post , The content script watches for new DOM nodes using a MutationObserver. As soon as LinkedIn loads a new post, the script adds the Suggest button.
• Fetch suggestions , Clicking the button triggers a fetch request to the serverless endpoint, sending the post’s headline and first few lines.
• Render options , The response contains three short comments. They appear as clickable cards; each card copies its text to the clipboard.
• One-click posting , After copying, you can paste into the comment field or hit Ctrl+V. The extension also highlights the comment box so you know where to paste.

Lessons Learned in 24 Hours

• Prompt engineering matters , The first set of suggestions sounded like marketing copy. I refined the prompt to ask for “a friendly peer tone, no buzzwords, under 30 words.”
• Performance is key , LinkedIn is already heavy; adding extra network calls slowed the feed. Caching the last 5 suggestions locally cut the latency from ~800 ms to ~200 ms.
• User experience over features , I initially added a “save for later” button, but it cluttered the UI. Stripping it back to the core “Suggest” action made the extension feel lightweight.
• Open source vs. product , The biggest question now is whether to open-source the code or keep it private while I iterate. I’m leaning toward open-source because the problem is universal, and community contributions could improve the AI prompts and add language support.

Next Steps & Community Feedback

I’ve published the extension on the Chrome Web Store (unlisted for now) and shared the repo on GitHub. I’m looking for feedback on:

• Prompt tweaks , Do the suggestions feel natural for different industries?
• Feature ideas , Would a “tone selector” (e.g., formal, casual) be useful?
• Privacy concerns , How comfortable are users with a serverless function processing post text?

If you’re an introverted professional who wants to stay visible without the awkwardness of drafting comments, give the extension a spin and let me know what you think.

If you need a custom tool, a full-stack side project, or just want to chat about building something that solves a real pain point, I’m available for freelance work. Check out my portfolio at wabby.dev.