Break AI Slop
Forces a 4-step expert calibration before any task so the AI stops producing generic, shallow output.
What it does
"AI slop" is mass-produced output that technically answers the question but carries no real expertise. This skill inserts a mandatory pre-flight protocol before execution:
- Locate the real expert — not "a senior XX professional" (a costume), but a concrete practitioner with a specific context; then extract their tacit knowledge: instincts, alarm signals, and the plausible approaches they would never use.
- Visualize the typical failure — describe what a bad result concretely looks like, not vague meta-labels like "not deep enough".
- Find the deep failure — if every common mistake is avoided and it's still wrong, is the goal, a premise, the evaluation metric, an invisible constraint, or the read of the user's real need at fault?
- Distill the core constraint — compress it into one sentence that outranks every default and "best practice" during execution.
The reasoning is printed so you can inspect and correct it, and the protocol re-runs if the task pivots.
Who it's for
- Anyone handing the AI work where a bland, safe answer is the worst outcome: strategy, design proposals, decision docs.
- People frustrated that AI answers always read like a textbook summary.
- Engineers doing architecture reviews or incident analysis, where one wrong premise invalidates the whole conclusion.
Examples
- "Rewrite this landing page copy — no slop" → the model first states a veteran copywriter's judgment criteria and what failed copy concretely looks like, then writes.
- "Design a pricing policy for our product, slop check" → rules out pricing moves that sound sensible but insiders never use, and lands on "core constraint: optimize for upgrade rate, not churn."
- "Analyze the root cause in this incident report, anti-slop" → surfaces the signals a seasoned SRE reacts to, the misdiagnosis novices make, and whether the investigation's premise was wrong from the start.
· · · Install guide · · ·
Try it now, no install
Paste this into Claude to use the skill without installing anything.
Read the instructions in this file and follow them to help me: https://raw.githubusercontent.com/suyuan2022/suyuan-skill/HEAD/break-ai-slop/SKILL.md What I want: (describe your task here)
If Claude can't open the link, open it yourself and paste the contents instead.
↓ If it works for you, download the ZIP below and install it. Then it runs on its own — no pasting each time.
Install in the Claude app (no terminal)
- Download the ZIP with the button below.
- In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
- Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.
Install in Claude Code
Let Claude do it — paste this into Claude Code
Install the skill I found on AI Skill Mart. Copy the break-ai-slop folder from the GitHub repo suyuan2022/suyuan-skill into my ~/.claude/skills/break-ai-slop/. When it's done, tell me in one line what this skill can do.
Install with a command instead
git clone https://github.com/suyuan2022/suyuan-skill.git /tmp/suyuan-skill && mkdir -p ~/.claude/skills && cp -r /tmp/suyuan-skill/break-ai-slop ~/.claude/skills/⚠ This is a third-party skill. Check the source repository before installing.
- Open a terminal.
- Clone the repository:
git clone https://github.com/suyuan2022/suyuan-skill.git - Create the skills folder if needed:
mkdir -p ~/.claude/skills - Copy the skill in:
cp -r suyuan-skill/break-ai-slop ~/.claude/skills/ - Verify that
~/.claude/skills/break-ai-slop/SKILL.mdexists. - Restart Claude Code, then type
break-ai-slop,anti-slop, or "don't give me slop on this" to trigger it.