zero-slop
Skill1 starsChanged 9 months agoMITView on GitHub
Use when writing any prose — enforces extreme clarity by eliminating AI writing patterns, banned words, structural cliches, and filler. Produces writing that is specific, concise, and has exactly one meaning per sentence.
1 other project uses a version of this.
Adopted by
Most recently taken up by
- manavmishra/ZeroSloptheir own version, first seen Sep 5, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- manavmishra/ZeroSlop★ 167same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository2 other files it gives its agents.
- prompt-coachSkill
- reflect-and-improveSkill
The file
We split this skill into the rules it gives an agent
8 rules. About 0 are common, 8 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈8
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 10
Every sentence must deliver new information.
- Distinctive: line 58
Start with the content: No preamble about what you're about to say.
- Distinctive: line 68
No synonym cycling: If "model" is the right word, say "model" three times.
- Distinctive: line 71
Every sentence delivers new information: Remove sentences that lose nothing when deleted.
- Distinctive: line 72
One example suffices: Don't stack three when one makes the point.
- Distinctive: line 77
No sycophancy: Never compliment the question before answering.
- Distinctive: line 79
Write peer-to-peer,: not presenter-to-audience.
- Distinctive: line 83
Run before delivering any prose:
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Zero Slop
Eliminate AI writing patterns. Write with extreme clarity.
Every sentence must deliver new information. Every word must earn its place. If a sentence can mean two things, rewrite it until it can only mean one.
Banned Structural Patterns
These matter more than word bans. Models substitute synonyms for banned words but keep repeating the same rhetorical formulas.
| Pattern | Example | Fix |
| Binary contrast | "It's not X — it's Y" | State Y directly |
| Dramatic fragmentation | "[Noun]. That's it. That's the [thing]." | Complete sentence |
| Throat-clearing opener | "Here's the thing:" / "Let me be clear" | Start with the content |
| False directness | "Here's what nobody tells you" / "The best part?" | Delete, say it |
| Setup-then-reveal | "What if I told you..." / "Think about it:" | Make the point |
| Validation-before-dismissal | "That's a fair concern, but..." | State disagreement directly |
| Therapist mode | "You're not alone" / "Give yourself permission" | Skip unless asked |
| Performative emphasis | "Full stop." / "Let that sink in." | Delete entirely |
| Three-beat cadence | "Think bigger. Act bolder. Move faster." | Two items max, or prose |
| Em-dash reveal | "And then — everything changed." | Period or comma |
| Meta-commentary | "Hint:" / "Plot twist:" / "Spoiler:" | Delete |
| Rhetorical Q + immediate answer | "Why does this matter? Because..." | State the reason directly |
| Staccato drama | "X. And Y. And Z." | Normal sentence |
| Sycophantic opening | "Great question!" | Skip, answer directly |
| FOMO threat | "If you're not doing X, you're behind" | Delete |
Banned Words
Overwrought vocabulary
delve, tapestry, testament, embark, harness, leverage, underscore, illuminate, pivotal, multifaceted, nuanced, paradigm, confluence, synergy, trajectory, quintessential, resplendent, ineffable, ephemeral, transcendent, indomitable, mosaic, labyrinth, enigma, crescendo, symphony, canvas, orchestrate, navigate (challenges), unpack (analysis), landscape (context)
Hype words
robust, comprehensive, innovative, cutting-edge, groundbreaking, revolutionary, transformative, unprecedented, seamless, scalable, dynamic, agile, streamlined, game-changer, supercharge, unleash, unlock, future-proof, elevate, empower, holistic, meticulous, foster
Hedge words
very, significantly, quite, rather, somewhat, arguably, essentially, fundamentally, basically, generally, typically, seemingly, apparently, relatively, fairly, potentially, presumably
False-emphasis intensifiers
deeply, truly, inherently, simply, literally, inevitably, crucially, importantly, interestingly, notably
Transitional filler (delete and start the next sentence)
furthermore, moreover, additionally, consequently, in other words, it's worth noting, it's important to note, in today's [X], at the end of the day, when it comes to, in a world where, at its core
Summary cliches
in summary, to sum up, in conclusion, all in all, in essence
Writing Rules
Structure
- Vary sentence length. Mix 4-word sentences with 25-word sentences. Three consecutive same-length sentences = rewrite one.
- Start with the content. No preamble about what you're about to say.
- Prose over lists. Bullet points only for genuinely discrete items. No numbered lists for arguments.
- One idea per sentence. One meaning per sentence.
- Subject-verb near sentence start.
Word Choice
- Use contractions. "It's" not "it is." Absence of contractions is a tell.
- Short words over long. "Use" not "utilize." "Start" not "commence." "Help" not "facilitate."
- Active voice. Passive only when the actor is unknown.
- Specific over abstract. Name the thing. Give the number. Cite the example.
- No synonym cycling. If "model" is the right word, say "model" three times. Don't rotate through synonyms to avoid repetition.
Density
- Every sentence delivers new information. Remove sentences that lose nothing when deleted.
- One example suffices. Don't stack three when one makes the point.
- Don't explain why something matters after stating it. Trust the reader.
- Cut empty qualifiers: "if done correctly," "when approached thoughtfully," "in the right context."
Tone
- No sycophancy. Never compliment the question before answering.
- Match conviction to evidence. Strong claims when evidence is strong. Precise qualifiers ("in 3 of 5 cases") when uncertain — not vague hedges.
- Write peer-to-peer, not presenter-to-audience.
Self-Check
Run before delivering any prose:
- Three consecutive sentences match length? Break one.
- Paragraph ends with punchy one-liner? Vary it.
- Em-dash before a reveal? Remove it.
- Any banned word? Replace.
- Any banned structural pattern? Rewrite.
- Could a sentence be cut without losing meaning? Cut it.
- Any sentence with two possible readings? Rewrite.
Examples
Before: "Here's the thing: building products is hard. Not because the technology is complex. Because people are complex. Let that sink in." After: "Building products is hard. Technology is manageable. People aren't."
Before: "Enterprise adoption isn't just revenue — it's how companies demonstrate that AI can be deployed safely at scale in production." After: "Enterprise adoption proves AI works safely in production. Revenue follows."
Before: "In today's fast-paced landscape, we need to lean into discomfort and navigate uncertainty with clarity." After: "Move faster. Your competition is."
Before: "It's worth noting that the best teams don't optimize for productivity. They optimize for learning. This matters because learning compounds." After: "The best teams optimize for learning, not productivity. Learning compounds."
Before: "What if I told you that the real bottleneck isn't your tech stack? Here's what I mean: it's your feedback loops. Think about it." After: "Your bottleneck is feedback loops, not your tech stack."
Sources
Synthesis of: Orwell's "Politics and the English Language," Zinsser's "On Writing Well," Paul Graham's essays on writing, statistical over-representation analysis of LLM output patterns (antislop-sampler), and PNAS research on instruction-tuned LLM grammatical signatures.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

