{"slug":"deslopify","title":"deslopify","summary":"Deslopifies AI-sounding prose and code diffs. Use when the user asks to make text sound human, remove ChatGPT-style filler, rewrite without fluff, clean up AI comments, or strip restatement comments from a diff without changing code.","platform":"Claude","tags":[],"authorName":"LLM Mart","authorSlug":"llm-mart","score":0,"source":"github","price":null,"verified":false,"createdAt":"2026-10-28T20:52:31.41195Z","repo":{"url":"https://github.com/kriscard/Skills","stars":16,"forks":0,"license":"MIT","updatedAt":"2026-10-28T18:33:51Z"},"bodyHtml":"<hr>\n<h2>name: deslopify\ndescription: &gt;-\nDeslopifies AI-sounding prose and code diffs. Use when the user asks to make\ntext sound human, remove ChatGPT-style filler, rewrite without fluff, clean up\nAI comments, or strip restatement comments from a diff without changing code.</h2>\n<h1>Deslopify</h1>\n<p>Strip the patterns that mark text as AI-generated. Preserve all factual content, author stance, intended tone, and useful depth — change the delivery, not the message.</p>\n<h2>The Slop Patterns</h2>\n<p>Scan for these before rewriting. More slop patterns justify a stronger edit; mostly-clean text gets a light pass. Treat a pattern as slop when it adds no meaning, obscures the point, or makes the reader decode the prose. Keep intentional voice that still communicates clearly.</p>\n<p><strong>Opening filler</strong> (cut 80% of the time — start with the actual point):</p>\n<ul>\n<li>\"In today's fast-paced world...\"</li>\n<li>\"It's important to note that...\"</li>\n<li>\"As we navigate the complexities of...\"</li>\n</ul>\n<p><strong>Hollow transitions</strong>:</p>\n<ul>\n<li>\"Furthermore,\", \"Moreover,\", \"Additionally,\" — usually just \"Also,\" or delete</li>\n<li>\"In conclusion,\" — end without announcing you're ending</li>\n<li>\"It's worth noting that...\" — just say the thing</li>\n</ul>\n<p><strong>Emphasis inflation</strong>:</p>\n<ul>\n<li>\"Importantly,\", \"Crucially,\", \"Significantly,\" — earn emphasis with content</li>\n<li>\"It goes without saying...\" — then don't say it</li>\n</ul>\n<p><strong>AI vocabulary</strong> (replace with the plain word):</p>\n<ul>\n<li>\"leverage\" → \"use\"</li>\n<li>\"utilize\" → \"use\"</li>\n<li>\"delve into\" → \"look at\" or \"explore\"</li>\n<li>\"in terms of\" → restructure the sentence</li>\n<li>\"moving forward\" → cut</li>\n<li>\"at the end of the day\" → cut</li>\n</ul>\n<p><strong>Excessive hedging</strong>:</p>\n<ul>\n<li>\"It seems that\", \"One might argue\", \"It could be said\" — commit to the claim or cut it</li>\n</ul>\n<p><strong>False ranges</strong>:</p>\n<ul>\n<li>\"from architecture to debugging\" implies a scale or complete span that does not exist — list the topics directly</li>\n<li>Keep ranges whose endpoints define a real scale, sequence, or scope</li>\n</ul>\n<p><strong>Empty synonym cycling</strong>:</p>\n<ul>\n<li>Switching among \"the service\", \"the system\", \"the platform\", and \"the solution\" only to avoid repetition makes identity less clear</li>\n<li>Pick one accurate term and repeat it unless the terms name real distinctions</li>\n</ul>\n<p><strong>Vague attribution</strong>:</p>\n<ul>\n<li>\"Experts believe\", \"studies show\", \"industry reports suggest\" — name the source when the attribution supports the claim</li>\n<li>Never remove or weaken an attribution because its source is missing. Preserve the attributed wording and flag it for sourcing; do not turn attributed belief into a fact, a hedge, or the author's own claim.</li>\n<li>Example: keep \"Experts believe the migration will cut incidents by 30%\" in the revised text and note <code>Source needed for \"Experts believe\".</code> Do not rewrite it as \"The migration may cut incidents by 30%.\"</li>\n<li>Keep attribution that is already specific enough to evaluate, including a named source, a cited study, a bounded group such as \"three interviewees,\" or a deliberately anonymous source whose role and reason for anonymity matter.</li>\n</ul>\n<p><strong>Sycophantic or chatbot phrasing</strong>:</p>\n<ul>\n<li>\"Great question!\", \"You're absolutely right\", \"Of course!\", \"I hope this helps\", \"Let me know if...\" — respond directly</li>\n<li>Keep genuine warmth or praise when it is part of the author's intended relationship with the reader</li>\n</ul>\n<p><strong>Nominalization</strong> (convert to verb):</p>\n<ul>\n<li>\"make a decision\" → \"decide\"</li>\n<li>\"provide assistance\" → \"help\"</li>\n<li>\"conduct an analysis\" → \"analyze\"</li>\n<li>\"have a discussion\" → \"discuss\"</li>\n</ul>\n<p><strong>Generic conclusions</strong>:</p>\n<ul>\n<li>\"The future looks bright\" or a recap that adds no decision, implication, or next step — cut it</li>\n<li>Keep conclusions that synthesize evidence, state a decision, or tell the reader what follows</li>\n</ul>\n<p><strong>Harmful over-compression</strong>:</p>\n<ul>\n<li>Dropped articles, verbless fragments, unexplained abbreviations, and symbol-speak such as \"invalid date → exit 2, no write\" make the reader reconstruct the sentence</li>\n<li>Restore the connective meaning with complete clauses: \"An invalid date exits with code 2 and writes nothing.\"</li>\n<li>Keep concise labels, headings, interface copy, and established notation when their context makes them clear</li>\n</ul>\n<p><strong>Mannered metaphor</strong>:</p>\n<ul>\n<li>Stock aphorisms, personified code, and decorative phrases such as \"a dial worth turning\" can hide the mechanism — name the literal action or property</li>\n<li>Keep metaphors that clarify a hard idea or belong to the author's deliberate creative voice</li>\n</ul>\n<p><strong>Bullet lists where prose would be clearer</strong>: not every list needs to be a list. Three related sentences often read better as a paragraph.</p>\n<p><strong>En-dash used as em-dash</strong>: <code>–</code> should be <code>—</code> (or just a comma)</p>\n<h2>What Good Looks Like</h2>\n<ul>\n<li>Short sentences land hard. Longer ones build context. Vary them.</li>\n<li>Prefer active voice when naming the actor improves the sentence: \"We shipped X\" rather than \"X was shipped by the team.\" Passive voice remains useful when the actor is unknown, irrelevant, or intentionally de-emphasized.</li>\n<li>Specific beats general: \"increased conversion by 18%\" not \"improved results\"</li>\n<li>The first sentence is the most important — if it's filler, cut it</li>\n</ul>\n<h2>Process</h2>\n<ol>\n<li>Read the full text.</li>\n<li>Identify which slop patterns appear; mention only the main classes in the note.</li>\n<li>Rewrite: cut filler, verb-ify nominalizations, and clarify passive constructions only when naming the actor helps.</li>\n<li>Present the revised version or cleaned diff.</li>\n<li>Add a 2–3 line note on the key changes made, not a list of every micro-change.</li>\n</ol>\n<h2>Output Criteria</h2>\n<p><strong>Prose mode:</strong> return the revised text plus a 2–3 line note naming the main classes of edits. Do not list every micro-change.</p>\n<p><strong>Code comment mode:</strong> return a cleaned diff or edit only comments, then report what was removed. Be surgical: no code, control-flow, formatting, or behavior changes.</p>\n<h2>Code Comment Mode</h2>\n<p>When the input is a diff or the user asks to remove AI comments from code:</p>\n<p><strong>What to remove:</strong></p>\n<ul>\n<li>Comments restating what code clearly does (<code>// increment counter</code> above <code>counter++</code>)</li>\n<li>Overly verbose explanations of obvious code</li>\n<li>Generic placeholders (<code>// TODO: implement</code>, <code>// Handle error</code>)</li>\n<li>Comments with AI phrasing (\"This function...\", \"The following code...\")</li>\n<li>Inline docs inconsistent with the file's existing comment style</li>\n</ul>\n<p><strong>What to keep:</strong></p>\n<ul>\n<li>Doc comments for public APIs (JSDoc, docstrings)</li>\n<li>Comments explaining why (business logic, workarounds, edge cases)</li>\n<li>Comments matching the file's existing style and density</li>\n<li>License headers and attribution</li>\n</ul>\n<p><strong>Hard constraint: DO NOT modify any code.</strong> Only remove comments. No refactoring, no style changes, no removing try/catch or null checks. If you spot code slop, note it in the summary but leave it.</p>\n<h2>Hard Rules (Prose Mode)</h2>\n<ul>\n<li>Never change factual content — only how it's expressed</li>\n<li>Never add new claims or examples not in the original</li>\n<li>Preserve the author's stance, intended audience, and recognizable voice</li>\n<li>Preserve useful connective explanation; concise does not mean compressed beyond easy reading</li>\n<li>If the text is mostly fine, say so — don't invent problems to fix</li>\n</ul>\n","files":[{"path":"SKILL.md","sizeBytes":6928,"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":"2026-10-28T20:53:36.149773Z","sha256":"D67BAE90D2C1F8E2B99B9DE41A7B06F7D698E6F521D79AEE89CFEAA6F49CDEC2","sizeBytes":3400},"review":null,"source":{"repositoryUrl":"https://github.com/kriscard/Skills","path":"skills/productivity/deslopify","license":"MIT","commit":"08395f46a77aa7c677b6fbb0b12d763186a00a45","subtreeSha":"6EA5322E17FE338A2DA87B4CA2E7180E8CAD74DD113BB8186BC9D2C280161439","lastSyncedAt":"2026-11-19T19:27:36.501149Z"},"reviewedAt":"2026-10-28T20:55:42.496345Z","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/kriscard/Skills/tree/main/skills/productivity/deslopify"},{"target":"claude-code","command":"claude plugin marketplace add https://llmmart.ai/marketplace.json && claude plugin install kriscard-skills@llmmart"},{"target":"git","command":"git clone https://github.com/kriscard/Skills.git"}]}