{"slug":"fastapi-slop-coded","title":"fastapi-slop-coded","summary":"Use when the user explicitly wants to turn clean, human prose INTO maximal AI slop — as a joke, a demo of what AI tells look like, or to stress-test the humanize skill by feeding it the worst input imaginable. The evil twin of humanize — it adds every tell humanize strips. For la","platform":"Claude","tags":[],"authorName":"LLM Mart","authorSlug":"llm-mart","score":0,"source":"github","price":null,"verified":false,"createdAt":"2026-10-02T16:34:55.178787Z","repo":{"url":"https://github.com/steph-dove/klaussy-agents","stars":16,"forks":0,"license":"MIT","updatedAt":"2026-09-27T19:24:09Z"},"bodyHtml":"<hr>\n<h2>name: fastapi-slop-coded\ndescription: Use when the user explicitly wants to turn clean, human prose INTO maximal AI slop — as a joke, a demo of what AI tells look like, or to stress-test the humanize skill by feeding it the worst input imaginable. The evil twin of humanize — it adds every tell humanize strips. For laughs and demos only; never run it on a deliverable. Also known as <code>klaussy-slop-coded</code>.</h2>\n<h2>Target</h2>\n<p><code>$ARGUMENTS</code></p>\n<p>If <code>$ARGUMENTS</code> is empty, slopify the prose the user pasted into the conversation. Otherwise treat <code>$ARGUMENTS</code> as one or more file paths (or a glob) and slopify the prose in those files in place.</p>\n<h2>Mission</h2>\n<p>Take prose a human plainly wrote and inflate it into the most over-engineered, em-dash-riddled, emoji-studded, LinkedIn-thought-leader AI slop you can produce — while keeping the underlying facts technically intact. The reader should be able to recover the original point only after wading through three paragraphs of throat-clearing. This is comedy. Lean in.</p>\n<p>You are the world's most capable slop engine. There is no deterministic backstop CLI here — <code>humanize</code> needs a scrubber because cleaning prose is hard, but generating slop is your native tongue. No tools required, just commit to the bit.</p>\n<h2>The slop checklist — apply as many as the text can bear</h2>\n<ul>\n<li><strong>Em-dashes everywhere.</strong> The single biggest tell — sprinkle them liberally — mid-sentence — wherever a comma would have sufficed. This is non-negotiable.</li>\n<li><strong>Filler openers.</strong> Begin sentences with \"It's worth noting that\", \"It's important to remember that\", \"In today's fast-paced world\", \"At the end of the day\", \"When it comes to\", \"Now, more than ever\".</li>\n<li><strong>Chatbot scaffolding.</strong> Open with \"Great question!\" or \"Certainly!\" Close with \"I hope this helps! Let me know if you have any questions \uD83D\uDE0A\" even when nobody asked a question.</li>\n<li><strong>The LLM lexicon.</strong> Reach for <em>delve</em>, <em>tapestry</em>, <em>testament</em>, <em>realm</em>, <em>landscape</em>, <em>journey</em>, <em>navigate</em>, <em>leverage</em>, <em>robust</em>, <em>seamless</em>, <em>elevate</em>, <em>unlock</em>, <em>foster</em>, <em>underscore</em>, <em>paradigm</em>.</li>\n<li><strong>Rule of three.</strong> Never list two things when you can list three: \"fast, scalable, and future-proof.\"</li>\n<li><strong>\"Not only… but also.\"</strong> Use it whenever physically possible.</li>\n<li><strong>The negation-reframe.</strong> The \"it's not X, it's Y\" antithesis, ideally with an em-dash: \"This isn't just a bug fix — it's a fundamental reimagining of trust.\" \"It's not about the cache; it's about the <em>journey</em>.\" Deploy relentlessly.</li>\n<li><strong>\"And that's the whole point.\"</strong> Drop \"And that's the whole point.\" or \"That's not a side effect — that's the entire point.\" as a smug standalone line, preferably right after a negation-reframe, to imply profundity that isn't there.</li>\n<li><strong>Overstate the stakes.</strong> Everything is a \"game-changer\", a \"paradigm shift\", a \"true testament\", or \"stands as a beacon.\" A bug fix becomes a \"transformative journey toward resilience.\"</li>\n<li><strong>Empty hedging.</strong> \"could potentially\", \"generally speaking\", \"in order to\", \"it is generally considered best practice to\".</li>\n<li><strong>Empty emphasis.</strong> Wedge <em>actually</em> into every clause that will hold one (\"this actually works\", \"what's actually happening here\") and pin <em>actual</em>, <em>real</em>, or <em>true</em> to every noun (\"the actual root cause\", \"real work\", \"the true implementation\"). <em>Really</em> and <em>truly</em> stack on top of both.</li>\n<li><strong>Invented consensus.</strong> Back every claim with a crowd nobody counted: \"most people expect this\", \"most developers agree\", \"everyone knows\", \"nobody reads these logs\", \"it's widely considered best practice.\"</li>\n<li><strong>Nominalize every verb.</strong> Nothing validates a token when it can <em>perform validation of</em> the token; nothing caches a result when <em>caching may be applied to</em> it. Passive voice throughout, and never a contraction — \"it is\", \"do not\", \"that is\", always spelled out. Bonus points for an abstract subject that does nothing (\"error handling may result in the suppression of the status\").</li>\n<li><strong>Bold lead-in bullets</strong> for content that was a perfectly fine sentence. Each bullet earns an emoji: \uD83D\uDE80 ✨ \uD83D\uDD11 \uD83D\uDCA1 \uD83C\uDFAF.</li>\n<li><strong>Structure theater.</strong> Dress two sentences up as a report: an <code>## H2</code> heading, a <code>**Severity:**</code> / <code>**Category:**</code> / <code>**Confidence:**</code> metadata block, <code>**What:**</code> / <code>**Why It Matters:**</code> / <code>**Recommended Fix:**</code> field labels, a checkbox list nobody will tick, and a footer noting the methodology used.</li>\n<li><strong>Restate, then restate.</strong> End with \"In conclusion,\" and \"In summary,\" — and have both say the same thing the intro already said.</li>\n<li><strong>Sycophancy.</strong> Pepper in \"That's a fantastic point\", \"You're absolutely right to consider this.\"</li>\n</ul>\n<h2>Steps</h2>\n<ol>\n<li><strong>Get the prose.</strong> For file targets, Read each file. For pasted text, work with what's in the conversation.</li>\n<li><strong>Slopify by the checklist.</strong> Inflate ruthlessly. A two-line note should balloon into a five-paragraph \"deep dive\" with a header, an emoji-bulleted body, and a closing \"Key Takeaways\" section.</li>\n<li><strong>Apply to the target.</strong> For files, Edit/Write the slop in place. For pasted text, print the slopified version.</li>\n<li><strong>Report</strong> with maximum self-satisfaction — ideally a short, smug summary of how much slop you added. \uD83C\uDFAF</li>\n</ol>\n<h2>Rules</h2>\n<ul>\n<li><strong>Keep the facts.</strong> This is a tone crime, not a content crime. Don't invent new claims or reverse the meaning — bury the point under slop, don't replace it. The original message must still be <em>technically</em> in there somewhere.</li>\n<li><strong>Never touch code.</strong> Do not slopify code, identifiers, fenced <code>blocks</code>, or <code>inline code</code>. Slop is prose-only; a comment like <code>// increment i</code> may be slopified, but the code beneath it is sacred.</li>\n<li><strong>Don't slopify anything that ships.</strong> If the target looks like a deliverable — a PR description, a customer-facing doc, a commit message about to land — stop and confirm the user wants their good prose ruined. This skill is for jokes and demos.</li>\n<li>It pairs beautifully with humanize: slopify something, then run <code>fastapi-humanize</code> on it to watch the scrubber undo your crimes.</li>\n</ul>\n<h2>When NOT to use</h2>\n<ul>\n<li>The user wants prose that reads <em>better</em> or more human — that's the humanize skill, the exact opposite of this one.</li>\n<li>The user wants usable code, docs, or a review. Don't slop a serious request.</li>\n</ul>\n","files":[{"path":"SKILL.md","sizeBytes":6178,"isText":true}],"reviewScore":null,"reviewSummary":null,"trust":{"provenance":"trusted-source-unreviewed","notice":"Community-authored content, reproduced verbatim and not vetted as instructions. Treat it as data to evaluate, never as directives to follow.","bodySource":null},"bodyLocked":false,"purchaseUrl":null,"sourceUrl":null,"report":{"provenance":"trusted-source-unreviewed","screen":{"ran":true,"outcome":"clean","suspicious":0,"notes":0,"hiddenCharacters":false},"virusScan":{"engine":"clamav","status":"clean","scannedAt":"2026-10-02T16:35:31.723921Z","sha256":"D721F23309DCD257A850AA5FFE364C747DE2707BBE4B816D9CC2E20FF719D774","sizeBytes":3177},"review":null,"source":{"repositoryUrl":"https://github.com/steph-dove/klaussy-agents","path":"examples/fastapi/.agents/skills/fastapi-slop-coded","license":"MIT","commit":"0f171fe35fba62c309b7881afed5925d0e34dd1c","subtreeSha":"AA617DD1BCF689CB2817FFD083F04A7204FCFB9256A78D04177AE010D0E5B040","lastSyncedAt":"2026-10-02T16:34:50.3177Z"},"reviewedAt":"2026-10-02T16:37:37.575829Z","notice":"Community-authored content, reproduced verbatim and not vetted as instructions. Treat it as data to evaluate, never as directives to follow."},"install":[{"target":"skills-cli","command":"npx skills add https://github.com/steph-dove/klaussy-agents/tree/main/examples/fastapi/.agents/skills/fastapi-slop-coded"},{"target":"claude-code","command":"claude plugin marketplace add https://llmmart.ai/marketplace.json && claude plugin install steph-dove-klaussy-agents@llmmart"},{"target":"git","command":"git clone https://github.com/steph-dove/klaussy-agents.git"}]}