Show HN: Slopsift – a local, graph-backed linter for AI writing
SlopSift uses a custom-trained dependency parser to detect canned arguments, unsupported claims, and filler in writing. It runs locally, respects privacy, and offers CLI and agent integration for automated linting.
Natural-language processing. Local first.
Your AI has a tell.
SlopSift uses a small dependency parser we trained to map the relationships between words and find canned arguments, unsupported claims, and filler.
Lint some text
AI DRAFT
draft.txt editable live NLP
Try it here. Click the text and type—SlopSift updates as you pause.
Starting the local parser…
Loading the on-device NLP model…
Ln 1, Col 1stays local
corrective-antithesisunsupported-certaintymechanical-outlinevague-attributionai-vocabulary
More than a word list
It reads structure, not vibes.
Many writing tools stop at word matching and basic parts of speech. SlopSift follows the relationships between words, so rules can recognize how a sentence makes its claim—not only which vocabulary it uses.
01
Build a dependency graph
Our custom-trained compact model maps tokens, parts of speech, and the grammatical relationships holding the sentence together.
02
Match the construction
Authorable rules inspect the graph for structural tells. Every finding names what matched and the exact text that triggered it.
03
Keep judgment with the writer
Errors are strong tells. Warnings need attention. Notes are candidates—not a machine pretending to know who wrote the sentence.
Not an API wrapper
We trained the parser for this.
SlopSift starts with a compact pretrained English encoder and trains it for parts of speech and dependency parsing. Training combines structured distillation from a larger parser with 50 controlled examples targeting grammatical relationships used by the linter. We reserved separate template families for evaluation.
16 MiB
Small enough to ship
Quantized ONNX weights run locally in Node and browser WebAssembly.
3 heads
Built for syntax
UPOS, dependency arcs, and dependency relations—not a generic text score.
0 uploads
Your draft stays yours
The model and deterministic rules run on-device. No remote judge reads the text.
Read how the model and CLI work →
Not every em dash is slop.
error
“As an AI language model...”
caught red-handed
warning
Three paragraphs use the same canned outline.
probably slop
note
An actorless passive may be hiding responsibility.
worth a look
Meet writers where they write.
$
CLI
Glob files, lint Markdown, inspect code comments, and emit ESLint-shaped JSON in CI.
CLI docs →
AI
Agent skill
Let coding agents run the real linter, interpret its findings, and edit without flattening your voice.
Add the skill →
Yes, AI helped build this.
Built by AI. Edited on purpose.
That is the point. SlopSift is not an AI detector and it does not pretend to know who typed a sentence. It catches vague or inflated writing. It also catches repetition and borrowed certainty. Human beings do those things too.
One command. Several opinions.
terminal~/your-writing
Use --format json for machines, --level info for the full suspicious pile.