▲ 214
Human signal
Here’s the tiny shell script I use to archive coding-agent sessions locally
It stores the transcript, git diff, and a 5-line summary so I can resume work without rereading everything.
Local-first AI slop blocker · Reddit-first beta
AI Slop Blocker for cleaner Reddit feeds.
AI Slop Blocker is a local-first feed filter for hiding low-effort AI posts, generic AI comments, recycled “expert” threads, and engagement bait. It starts with Reddit, explains every folded post, and avoids uploading your feed by default.
One intent, one promise: reduce obvious AI slop without pretending to prove who wrote every post.
AI slop blockerExact problem: low-effort AI posts wasting attention
Reddit-first betaStart where community signal and comment quality matter
Local by defaultScore visible posts in your browser before any cloud workflow
Reversible controlFold, badge, show anyway, whitelist, or raise the threshold
Direct answer
What is an AI slop blocker?
An AI slop blocker is a feed-quality filter for low-effort AI content. It should hide or fold generic AI posts, recycled listicles, bot-like comments, and engagement bait while leaving useful AI-assisted work alone.
The important distinction: this is not a magical AI detector. A practical AI slop blocker looks for low-value signals — vague claims, template phrasing, missing evidence, repeated hooks, and thin links — then lets you decide whether to fold, badge, hide, or whitelist.
- Primary use case: make Reddit feeds easier to scan.
- Privacy angle: score visible posts locally by default.
- Control angle: every hidden item should be explainable and reversible.
How it works
How AI Slop Blocker decides what to fold.
01
Scan the posts already on screen
The extension reads visible feed text in the browser. The MVP starts with Reddit because the pain is clear, communities are searchable, and feedback is fast.
02
Score low-value patterns, not “AI guilt”
It looks for generic AI phrasing, engagement bait, vague authority claims, repetition, thin affiliate/SEO links, and low information density.
03
Fold, badge, hide, or whitelist
You choose the threshold and action. Suspicious posts show clear reasons, and useful posts can be shown anyway or whitelisted.
▲ 6
In today’s fast-paced digital world, AI is revolutionizing everything 🚀
Here are 7 unbelievable ways to 10x your productivity. Number 5 will shock you. Thoughts?
▲ 47
Needs review
Ask HN: How do you tell useful AI summaries from recycled fluff?
AI Slop Blocker would badge this, not hide it. You stay in control.
More private than cloud classifiers
Cloud classifiers can expose sensitive reading context. AI Slop Blocker’s wedge is local-first scoring: visible posts are evaluated in the browser before any optional cloud workflow.
Less brittle than keyword muting
Keyword mutes miss disguised slop and hide useful AI discussions. AI Slop Blocker combines phrase patterns, structure signals, link behavior, and your own overrides.
Narrower than YouTube-only tools
YouTube/video tools solve a different problem. This project focuses on text-heavy feeds — Reddit first — where repeated comments and generic posts can be folded quickly.
SEO guide
What AI Slop Blocker should hide — and what it should keep
AI slop is not simply “content made with AI.” Useful summaries, code explanations, translations, and drafts can still help people. The problem is low-effort AI content: posts that look polished but add no evidence, no lived detail, no specific answer, and no reason to trust the author.
AI Slop Blocker is being designed for people who open Reddit to find human signal and instead hit recycled growth hooks, vague expert claims, bot-like replies, and thin link drops. The first version focuses on Reddit because users can validate the filter quickly in real communities; later versions can carry the rule set to X and LinkedIn.
What it should fold
- Generic “in today’s fast-paced world” AI phrasing.
- Engagement-bait posts with little detail or proof.
- Repeated listicle templates, emoji-heavy hooks, and fake urgency.
- SEO spam comments that point to thin external pages.
What it should not punish
- Clear human questions, even when they mention AI.
- Useful AI-assisted work with examples, screenshots, data, or code.
- Posts from trusted communities or whitelisted authors.
- Content you manually choose to show anyway.
Why local-first matters
A feed filter sees sensitive context: communities you read, work topics you follow, and people you interact with. That is why the MVP should score visible posts in the browser first, explain every folded item, and avoid uploading feed text by default.
How to judge if it works
The right metric is not “AI detection accuracy.” The right question is whether your feed becomes easier to scan without hiding useful posts. Good filters reduce obvious slop, explain decisions, and make false positives easy to reverse.
FAQ
AI Slop Blocker FAQ
Is this an AI detector?
No. It does not promise to prove whether text was generated by AI. It scores low-effort patterns that make feed posts less useful.
Will my Reddit or LinkedIn feed be uploaded?
The planned MVP is local-first. Visible feed text is scored in the browser by default, with no feed upload required.
What platform comes first?
Reddit first, because low-effort AI posts and generic comments are easy to validate in real communities. X and LinkedIn follow after the rule set is useful.