Igloo302

Igloo302 / interest-radar

Public

Hermes Agent Skill: 兴趣雷达 / Personal Radar - 相关性判断引擎

17
1
89% credibility
Found May 19, 2026 at 17 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
AI Summary

Interest Radar is a smart helper for AI assistants that tells you whether online content is worth your time. Instead of scrolling endlessly or relying on algorithms that track you, this tool privately analyzes any link or article you share and explains exactly why it might—or might not—be relevant to your interests. It works quietly in the background, learns from your feedback, and gets better at understanding your preferences the more you use it. The project emphasizes privacy (it doesn't collect or store your data) and works across different AI assistant platforms.

How It Works

1
🔍 You discover Interest Radar

You hear about a tool that helps your AI assistant understand what content actually matters to you.

2
📦 You add it to your AI assistant

You copy the skill files into your AI assistant so it can start making smart judgments about content.

3
🔗 You share a link with your assistant

You forward an article or link and ask your assistant to check if it's worth your time.

4
🧠 Your assistant thinks it through

Your AI assistant automatically analyzes the content and compares it against what it knows about your interests.

5
You get a clear verdict
✅
It's relevant

Your assistant confirms the content matches your interests and explains the connection.

❌
Not relevant

Your assistant politely explains why this content doesn't align with what you care about.

6
👍👎 You give feedback

You confirm or dismiss the judgment with a simple reply, and your assistant learns from your choice.

🎯 Your assistant gets smarter over time

With each piece of feedback, your AI assistant fine-tunes its understanding of what matters to you.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 17 to 17 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is interest-radar?

Interest Radar is a relevance judgment engine for AI agents. You send it any link or content, and it tells you whether that content matches your interests, why it thinks so, and whether you should bother reading it. The entire thing runs as a SKILL.md workflow definition with no Python dependencies or external runtime requirements. It works across multiple agent platforms including Hermes, Claude Code, Codex, and Cursor.

Why is it gaining traction?

The hook here is simplicity and portability. Most relevance filtering tools require you to run a separate service, manage APIs, or write custom code. Interest Radar just drops into your existing agent setup as a skill definition. The feedback loop is the real differentiator—confirming or dismissing judgments trains the system to match your preferences over time. Batch judgment capability lets other skills filter content before bothering you with it.

Who should use this?

Knowledge workers drowning in shared links and newsletters will get the most value. Developers building agent-based content pipelines could use the batch judgment interface to filter incoming information. Anyone tired of manually sorting signal from noise in their AI assistant's output.

Verdict

This is a clever concept with a credibility score of 0.9% and only 17 stars—extremely early stage. The no-dependency approach is genuinely innovative, but the documentation lacks practical examples and there's no visible test coverage. Worth watching, but wait for a more mature release before betting your workflow on it.

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.