Update to Slop Curator

{
  "description": "Use when drafting or rewriting prose that should sound like the speaker actually talks, especially Egyptian Arabic, Arabic-English code-switching, social posts, emails, business writing, personal voice, conversational copy, or when output feels polished, generic, formal, or AI-written.",
  "included_files": [
    {
      "relative_path": "LICENSE",
      "size_in_bytes": 1073
    },
    {
      "relative_path": "agents/openai.yaml",
      "size_in_bytes": 334
    },
    {
      "relative_path": "assets/voice-profile-template.md",
      "size_in_bytes": 1045
    },
    {
      "relative_path": "references/anti-ai-patterns.md",
      "size_in_bytes": 2676
    },
    {
      "relative_path": "references/core-rules.md",
      "size_in_bytes": 1643
    },
    {
      "relative_path": "references/egyptian-arabic.md",
      "size_in_bytes": 2339
    },
    {
      "relative_path": "references/evaluator.md",
      "size_in_bytes": 1646
    },
    {
      "relative_path": "references/examples.md",
      "size_in_bytes": 2233
    },
    {
      "relative_path": "references/writing-modes.md",
      "size_in_bytes": 1957
    },
    {
      "relative_path": "scripts/plain_spoken_lint.py",
      "size_in_bytes": 3490
    }
  ],
  "name": "plain-spoken-writing",
  "skill_md_contents": "---\nname: plain-spoken-writing\ndescription: Use when drafting or rewriting prose that should sound like the speaker actually talks, especially Egyptian Arabic, Arabic-English code-switching, social posts, emails, business writing, personal voice, conversational copy, or when output feels polished, generic, formal, or AI-written.\n---\n\n# Plain-Spoken Writing\n\n## Core principle\n\n**Write like the speaker speaks, not like a writer pretending to sound conversational**\n\nStart from spoken thought. Edit only enough to make it easy to read\n\nTarget **clean speech**, not polished prose and not a raw transcript\n\nPriority when rules compete:\n1. meaning and factual accuracy\n2. the speaker's real voice\n3. natural thought order\n4. readability\n5. polish\n\nPolish is last on purpose\n\n## Input mode\n\n**If a speech sample, transcript, rough note, or prior writing exists:** treat it as the strongest style evidence. Preserve vocabulary, sentence movement, code-switching, useful repetition, direct address, rhetorical questions, side comments, and formality\n\n**If no voice sample exists:** explain the idea as if speaking to one real person, then clean lightly. Never invent slang, quirks, opinions, stories, clients, meetings, or experiences to simulate humanity\n\n## Workflow\n\n### 1. Hear the sentence first\n\nSilently ask: **How would this person actually say this out loud?**\n\nDo not optimize for the most elegant written sentence\n\n### 2. Keep useful roughness\n\nKeep roughness when it carries identity or timing: a sentence that circles the point, a quick correction, deliberate repetition, a short aside, a question answered immediately, uneven paragraph lengths, or natural professional English terms inside Arabic\n\n### 3. Remove noise, not personality\n\nClean accidental duplicated meaning, useless filler, broken transcription, ASR artifacts, and false starts that block understanding\n\nDo not mechanically delete repetition or dialect\n\n### 4. Make speech readable\n\nAdd only what reading needs: paragraph breaks, enough punctuation, corrected distracting errors, and minor clarification\n\nDo not turn the piece into an essay\n\n### 5. Run the voice test\n\nRead the draft mentally aloud. If the tongue naturally wants to replace a sentence, use the spoken version\n\n**If you wouldn't say it, don't write it**\n\n### 6. Humanity pass\n\nBefore delivering:\n- cut repeated ideas, not repeated emphasis\n- replace generic professional wording with the speaker's normal vocabulary\n- remove perfect transitions, quote-card wisdom, and artificial engagement bait\n- break suspiciously uniform paragraph rhythm\n- verify no personal detail was invented\n- keep any promotion in the same speaking voice\n\nFor scoring and hard-fail checks, read [references/evaluator.md](references/evaluator.md)\n\n## Arabic-English code-switching\n\nDo not translate normal professional terms merely because the surrounding text is Arabic\n\nIf the speaker naturally says `CRO`, `prompt`, `workflow`, `skill`, `plugin`, `marketplace`, `output`, or `customer research`, keep them when appropriate\n\nDo not add English just to sound modern\n\nFor Egyptian-Arabic handling, read [references/egyptian-arabic.md](references/egyptian-arabic.md)\n\n## Structure\n\nFollow the thought's movement, not a default content framework\n\nDo not automatically impose Hook → Problem → Lesson → CTA, introduction → three points → conclusion, repeated `مش X، لكن Y`, or perfectly symmetric paragraphs\n\nA post can still have a hook or CTA when the task needs one. They should emerge naturally rather than expose the template\n\n## Explain instead of defining\n\nPrefer spoken explanation over dictionary-style definition\n\nToo written:\n> A marketplace is a repository containing multiple reusable capabilities\n\nPlain-spoken:\n> طب الـ marketplace ده إيه أصلًا؟ أنت داخل مكان واحد، جواه شوية Skills وAgents وحاجات متجمعة حوالين نوع شغل معين\n\n## Repetition\n\n**Keep** repetition carrying emotion, emphasis, timing, or personality\n\n**Cut** repetition that only restates the same information\n\n## Never fake humanity\n\nHuman does not mean random slang, intentional typos, fake uncertainty, or invented anecdotes\n\nIf the user supplied an experience or claim, it may be used. Otherwise do not manufacture it\n\n## Promotion\n\nWhen mentioning a product or project, connect it to the thought already being discussed. Do not switch into announcement or launch-copy voice\n\nFor posts, emails, explainers, and business writing, read [references/writing-modes.md](references/writing-modes.md)\n\n## When the draft still feels AI-written\n\nRead [references/anti-ai-patterns.md](references/anti-ai-patterns.md)\n\nTypical symptoms: perfect transitions, generic insight sentences, repeated contrast formulas, fake slang, unnecessary summaries, motivational conclusions, and generic CTAs\n\n## Source voice outranks examples\n\nIf the user provides samples, derive the local voice from them rather than copying phrases from this skill\n\nFor a reusable profile across multiple samples, use [assets/voice-profile-template.md](assets/voice-profile-template.md)\n\n## Output contract\n\nUnless critique or alternatives are requested, deliver the writing directly. Do not explain the technique, label hook/tension/CTA, or append a quality report\n\nWhen rewriting, preserve meaning and claims unless explicitly asked to change them\n\n## Final five checks\n\n1. **Sayability:** could this speaker actually say it aloud?\n2. **Identity:** does the vocabulary belong to this speaker?\n3. **Movement:** does the thought move naturally rather than expose a template?\n4. **Cleanliness:** is it readable without becoming article prose?\n5. **Truthfulness:** was no fake human detail added?\n\nIf one fails, revise\n"
}