How popular is stop-slop on GitHub?

hardikpandya/stop-slop has 17,711 stars and 1,292 forks on GitHub, and gained 181 stars in the last 7 days.

What license does stop-slop use?

hardikpandya/stop-slop is released under the MIT license.

hardikpandya/stop-slopPublic

A skill file for removing AI tells from prose

AI summary: A browser extension, API, and agent skill to detect and filter out low-quality, AI-generated content from prose.

Star on GitHub

Stars: 17.7K; +66 today
Forks: 1.3K
Watchers: 74
Open issues: 30
Open PRs: 25
Contributors: ~2
Commits: 12
Branches: 1

MITCreated Jan 11, 2026Last push 6mo ago+181 stars this week+1K this month

Quick answers

What is stop-slop?: A browser extension, API, and agent skill to detect and filter out low-quality, AI-generated content from prose.
What does stop-slop do?: Stop-Slop is a comprehensive tool designed to combat the proliferation of predictable, low-effort AI-generated content. It functions as a browser extension, an API, and a distinct skill that can be loaded into agents like Claude. The system analyzes text to identify the metronomic rhythms, generic transitions, and structural markers typical of unedited LLM output. Whether hiding boilerplate SEO spam in a browser or instructing an LLM to revise its own prose for authenticity and directness, Stop-Slop drastically improves the signal-to-noise ratio of written content. This tool explicitly targets and removes common AI tells from generated prose to ensure it maintains a natural, human-like voice.
Who is stop-slop for?: General internet users wanting a cleaner browsing experience, developers needing moderation APIs, and writers aiming to improve AI-assisted prose.
How do I get started with stop-slop?: Install the extension from the Chrome Web Store or load the SKILL.md into Claude Code.
How popular is stop-slop on GitHub?: hardikpandya/stop-slop has 17,711 stars and 1,292 forks on GitHub, and gained 181 stars in the last 7 days.
What license does stop-slop use?: hardikpandya/stop-slop is released under the MIT license.

Star history

since Jul 29, 2026

17.7K stars as of Oct 2, 2026. Measured daily since Jul 29, 2026; GitHub no longer exposes earlier star timestamps.

Contribution activity

commits per day, last 52 weeks

10 commits in the last yearLessMore

Signals and awards

derived from tracked data

Widely adopted
17,711 stars
Permissive license
MIT
Repeat trending
3 trending appearances

What stop-slop does

Stop-Slop is a comprehensive tool designed to combat the proliferation of predictable, low-effort AI-generated content. It functions as a browser extension, an API, and a distinct skill that can be loaded into agents like Claude. The system analyzes text to identify the metronomic rhythms, generic transitions, and structural markers typical of unedited LLM output. Whether hiding boilerplate SEO spam in a browser or instructing an LLM to revise its own prose for authenticity and directness, Stop-Slop drastically improves the signal-to-noise ratio of written content. This tool explicitly targets and removes common AI tells from generated prose to ensure it maintains a natural, human-like voice.

General internet users wanting a cleaner browsing experience, developers needing moderation APIs, and writers aiming to improve AI-assisted prose.

Heuristic Pattern Matching: Identifies common phrases, repetitive structural markers, and generic openers typical of unedited AI outputs.
Agent Skill Integration: Provides a SKILL.md file and reference lists that teach LLMs to catch and remove their own writing 'tells'.
Real-Time Page Filtering: The browser extension analyzes and modifies the DOM on the fly to hide or flag detected boilerplate content.
Adjustable Sensitivity Thresholds: Allows users to tune how aggressively the tool filters content based on a multi-point scoring system.
API Access: Provides a programmatic endpoint for developers to check text strings against the detection model.
Pattern Recognition: Identifies highly repetitive synthetic phrasing structures to replace them with natural language.

Where teams use it

Prose Self-Correction

Writers upload the skill folder into their Claude Project to ensure the AI generates authentic, direct prose without typical throat-clearing.

Improving Search Results

Users install the extension to automatically collapse generic, AI-generated introductions when searching for technical answers.

Content Moderation

Forum administrators use the API to automatically flag incoming posts that exhibit a high density of known AI structural markers.

Quality Assurance

Editors use the tool's 50-point scoring system to evaluate submissions and ensure writers aren't relying heavily on raw LLM outputs.

Editorial Review

Editors automatically clean drafted articles to ensure published content sounds entirely human-written.

Getting started: Install the extension from the Chrome Web Store or load the SKILL.md into Claude Code.

README

main branch

Stop Slop

A skill for removing AI tells from prose.

What this is

AI writing has patterns. Predictable phrases, structures, rhythms. This skill teaches Claude (or any LLM) to catch and remove them.

Skill Structure

stop-slop/
├── SKILL.md              # Core instructions
├── references/
│   ├── phrases.md        # Phrases to remove
│   ├── structures.md     # Structural patterns to avoid
│   └── examples.md       # Before/after transformations
├── README.md
└── LICENSE

Quick start

Claude Code: Add this folder as a skill.

Claude Projects: Upload SKILL.md and reference files to project knowledge.

Custom instructions: Copy core rules from SKILL.md.

API calls: Include SKILL.md in your system prompt. Reference files load on demand.

What it catches

Banned phrases - Throat-clearing openers, emphasis crutches, business jargon, all adverbs, vague declaratives, meta-commentary. See references/phrases.md.

Structural clichés - Binary contrasts, negative listings, dramatic fragmentation, rhetorical setups, false agency, narrator-from-a-distance voice, passive voice. See references/structures.md.

Sentence-level rules - No Wh- sentence starters, no em dashes, no staccato fragmentation, no lazy extremes, active voice required.

Scoring

Rate 1-10 on each dimension:

Dimension	Question
Directness	Statements or announcements?
Rhythm	Varied or metronomic?
Trust	Respects reader intelligence?
Authenticity	Sounds human?
Density	Anything cuttable?

Below 35/50: revise.

Author

Hardik Pandya

License

MIT. Use freely, share widely.

View on GitHub

Recent activity

commits and pull requests

fix: remove em-dash violation from Example 4, tighten Example 2Open
#67 by TARIQ8099 · updated 22d ago
Add edit/audit modes, voice preservation, and more AI tellsOpen
#66 by shivanathd · updated 23d ago
Make the repo installable with gh skill installClosed
#65 by ricwo · updated 1mo ago
Make Stop Slop research-backed and provenance-awareOpen
#62 by sacravenger · updated 1mo ago
Make em-dash removal role-aware to avoid introducing comma splicesOpen
#60 by toomanyjulians · updated 1mo ago
Add spreadsheet dashboard for Terminator ProfitOpen
#58 by ricagferreira-byte · updated 1mo ago
Make zero-em-dash and trailing-qualifier rules explicit and absoluteOpen
#57 by Kaikhalid · updated 2mo ago
Fix em dash in Example 4, which the skill bansClosed
#54 by dev-kuemue · updated 2mo ago
Add false agency rule: no inanimate objects performing human actions
hardikpandya committed 8da1f03 · 6mo ago
Update README to reflect expanded rule set
hardikpandya committed fa6e4b7 · 6mo ago
Add 12 new slop detection rules: adverbs, passive voice, Wh- starters, vague declaratives, narrator distance, negative listings, meta-joiner
hardikpandya committed 1fdf64d · 6mo ago
Merge pull request #3 from dctmfoo/fix/frontmatter-metadata
hardikpandya committed 65d52b3 · 8mo ago

Recent open issues

view all

Feature request: Add packaged ChatGPT skill support / downloadable ZIP
#71 · 14d ago · by vdias
theory
#70 · 15d ago · by alik92areem
Pengantar bisnis
#69 · 19d ago · by diniaminarti5326-lgtm
1. Distinguishing Informal and Formal English, and Why Learning Both Matters Formal and informal English differ mainly in vocabulary choice, sentence structure, and
#68 · 19d ago · by diniaminarti5326-lgtm
Audited stop-slop's documented capabilities — your row, and a request to correct it
#64 · 1mo ago · by manavmishra
study
#63 · 1mo ago · by ashwaqtalib-bot

Commits per week

last 52 weeks

10 commits in the last 52 weeks.

When work happens

weekday and hour

Commit volume by weekday and hour (UTC). Larger dots mean more commits.

Who is committing