July 2, 2026
Clone Wars: Messy Code Strikes Back
Show HN: CLI tool for detecting non-exact code duplication with embedding models
A new code-cleanup app has developers cheering, questioning, and eyeing their messy projects
TLDR: Slopo is a new command-line app that tries to find hidden repeated code that normal tools miss, especially in big messy projects. Commenters were excited by the cleanup potential, but immediately grilled it on real-world accuracy, language differences, and whether it can handle the chaos of everyday coding.
A fresh Show HN launch called Slopo is serving one very relatable promise: finding the sneaky, lookalike chunks of code hiding in far-apart corners of a project. In plain English, it’s meant to catch the kind of repeated work that slips past normal copy-paste detectors. And the community reaction? Very into it, with just enough skeptical side-eye to keep things spicy.
The loudest vibe in the comments is basically: finally, a tool for the mess nobody wants to admit is there. One commenter called it a “great use case” for this kind of AI-powered matching, while another said it sounds perfect right before a big cleanup job. There’s also immediate workplace-chaos energy: one developer is already fantasizing about wiring it into a pre-push hook, aka making the tool judge your code before it leaves your laptop. That’s the kind of suggestion that makes neat freaks cheer and everyone else sweat.
But this wasn’t pure applause. The biggest questions were practical: how exactly does it slice up code in different languages, and does it work better on stricter languages than looser ones? That’s where the real drama lives—not “is this cool?” but can it survive contact with actual chaotic codebases. Even the creator admits the tool can surface “possible” duplicates that still need checking, which gave the thread a fun undercurrent of optimism mixed with “sounds amazing, now prove it.” In other words: developers smell promise, but they also smell edge cases.
Key Points
- •Slopo is a CLI tool that detects non-exact code duplication by comparing embeddings of code units rather than relying on exact matching.
- •The tool targets hard-to-spot similar code that is far apart in a codebase and supports Python, TypeScript, JavaScript, Java, Kotlin, C#, Go, and Rust.
- •Its output is clusters of similar code units ranked by similarity and codebase distance, intended for review and possible refactoring.
- •Embeddings are generated through external providers, with code-focused models such as Voyage AI recommended and LiteLLM-compatible providers supported.
- •The workflow includes configuration, indexing, embedding generation, analysis, incremental re-indexing, and optional use of `slopo.ignore.txt` to suppress already reviewed clusters.