Stop Slop

Stop Slop is an open-source skill file created by Hardik Pandya that teaches Claude (or any LLM) to identify and eliminate the predictable phrases, structures, and rhythms that mark AI-generated writing. It is hosted on GitHub under the MIT License and has accumulated over 5,000 stars and 400 forks according to the repository metadata.

What It Is

Stop Slop is a prompt-engineering skill — a structured set of instructions and reference files that can be loaded into Claude Code, Claude Projects, custom instructions, or any LLM API system prompt. Its core job is to act as a style editor: it flags and removes the linguistic fingerprints that make AI prose sound robotic, formulaic, or hollow. The skill is not a standalone application but a portable set of files that augment an existing LLM workflow.

What It Catches

The skill targets three categories of AI writing problems:

• Banned phrases — throat-clearing openers, emphasis crutches, business jargon, adverbs, vague declaratives, and meta-commentary (catalogued in references/phrases.md)
• Structural clichés — binary contrasts, negative listings, dramatic fragmentation, rhetorical setups, false agency, narrator-from-a-distance voice, and passive voice (catalogued in references/structures.md)
• Sentence-level rules — no Wh- sentence starters, no em dashes, no staccato fragmentation, no lazy extremes, active voice required

Scoring System

The skill includes a 1–10 scoring rubric across five dimensions: Directness (statements vs. announcements), Rhythm (varied vs. metronomic), Trust (reader intelligence respected), Authenticity (sounds human), and Density (anything cuttable). A combined score below 35 out of 50 triggers a revision pass.

How to Deploy It

The skill is designed to be dropped into multiple LLM workflows without modification:

• Claude Code — add the 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 the system prompt; reference files load on demand

Why It Got Attention

The repository reached over 5,000 GitHub stars and 402 forks, signaling strong community interest in practical, reusable prompt-engineering resources for prose quality. The project addresses a widely felt problem — AI writing that is technically correct but tonally flat — with a concrete, file-based solution rather than a vague style guide.

Update: Active Development

The repository was created in January 2026 and last pushed in March 2026, with the last metadata update in May 2026. With 13 open issues and an active fork count, the project shows ongoing community engagement. The MIT license allows free use, modification, and redistribution.