agent-quality-controls

agent-quality-controls / slopless

Public

Deterministic textlint rules and CLI for catching prose slop in Markdown

www.npmjs.compackageslopless ai cli lint linter llm
34
13
89% credibility
Found May 18, 2026 at 90 stars 3x -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
HTML
AI Summary

Slopless is an open-source writing quality checker that scans Markdown documents for 50+ types of writing problems. It catches both AI-generated text patterns (like overused LLM vocabulary, hedge stacking, and formulaic openers) and human writing issues (like clichés, redundancy, and wordiness). The tool also includes readability metrics and academic-specific checks for "tortured phrases" used in paper spinning. Users install it as a development dependency and run it against their files to get a detailed JSON report of issues they can then fix.

How It Works

1
✍️ You write documentation or blog posts

You spend hours crafting clear, helpful writing for your readers and want to make sure it sounds genuinely human.

2
🔍 You discover your writing might sound robotic

You hear about slopless - a tool that catches both AI-style writing and human writing problems in one pass.

3
📦 You add it to your project

You install slopless as a development tool, just like a spell-checker for your writing workflow.

4
⚡ You run it on your Markdown files

With one simple command, slopless scans all your documentation and checks for 50+ different writing issues at once.

5
📋 You review the findings

The tool shows you exactly where your writing uses tired phrases, sounds too formal, or follows predictable patterns.

6
You decide how to fix the issues
✏️
Fix it yourself

You rewrite the flagged sentences with fresher, more specific language.

🤖
Let automation help

Other tools can use the JSON output to suggest or make corrections automatically.

✨ Your writing sounds genuinely human

Your documentation is clearer, avoids AI-sounding phrases, and your readers get helpful content instead of generic filler.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 90 to 34 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 slopless?

Slopless is an npm package and CLI tool that catches AI-generated or low-quality prose in your Markdown files. It uses 50+ deterministic textlint rules to flag clichés, weasel words, wordy phrases, passive voice, and LLM-specific writing patterns like "delve," "tapestry," or "leverage" without requiring an API call or external service. You run it with a simple command like `npx slopless "docs/**/*.md"` and get structured JSON output with findings, or it exits with clear codes (0 for clean, 1 for findings, 2 for failure). It's a local-only guardrail for maintaining prose quality in documentation.

Why is it gaining traction?

The deterministic approach is the key differentiator. Unlike AI-based writing analysis tools, slopless runs entirely offline, produces consistent results, and avoids the cost and latency of calling an LLM. The ruleset draws from established linting sources like proselint, write-good, and various style guides, giving it credibility through provenance. The CLI-first design makes it easy to integrate into CI pipelines or pre-commit hooks for automated prose quality gates.

Who should use this?

Documentation maintainers who want to enforce writing standards across team Markdown files will find this most useful. Technical writers vetting AI-assisted drafts can use it to catch overused LLM patterns. Open source maintainers who want to keep READMEs free from corporate-speak or clichés. It's less useful for creative writing where some of these patterns might be intentional.

Verdict

With a credibility score of 0.8999999761581421% and only 34 stars, slopless is early-stage but conceptually sound. The deterministic, offline approach fills a real gap for teams that want linting-style prose checks without AI dependencies. Start with a small test run on your docs folder to see if the rule set matches your style preferences before committing to full integration.

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.