[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6R91HM6GN2903N0RT7ZPVWW":3,"skill-same:01M6R91HM6GN2903N0RT7ZPVWW":78},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":39,"linkOnly":40,"markdownContent":41,"markdownBody":42,"frontMatter":43,"refName":44,"commitSha":45,"skillDirectory":46,"forksCount":47,"repositoryHtmlUrl":48,"repositoryHomepageUrl":6,"ownerHtmlUrl":49,"ownerBlogUrl":50,"ownerDescription":51,"updatedAt":52,"install":53,"resources":66},"01M6R91HM6GN2903N0RT7ZPVWW","asb-voters",null,"Distills a person's or company's strengths, experiences, and obsessions down to their one or two VOTERS — the decisive, idiosyncratically extreme traits that win the contest because customers who value them accept every other trade-off. Runs each candidate through a hard gauntlet: extremity earned through obsession (not mere competence), rarity among peers, decisiveness (name the weaknesses it overpowers), and reverberation (it must force decisions in product, pricing, and market). Caps the answer at two, records survivors in VOTERS.md — plus the near-miss 'special strengths' to deploy — and delivers the honest zero-voter verdict when nothing is extreme yet. Takes a self-portrait file (M-numbered), strengths or keystones charts, or a live capture. Load when the user asks what makes them special, what to bet the strategy on, their unfair advantage or superpower, or to 'find our voters.' Do NOT load to build the full self-portrait (the previous step) or to design the whole strategy — this file feeds it.","Nairon-AI\u002Fagentic-starter-pack","Nairon-AI","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F260662138?v=4","01M6R8PZ1Y8CSQGFN61VACZGTH","Nairon AI","01M6R8PZ6184VR3BA88YZFBC3N",4,"MIT","https:\u002F\u002Fgithub.com\u002FNairon-AI\u002Fagentic-starter-pack\u002Fblob\u002Fmain\u002Fbusiness\u002Fasb-voters\u002FSKILL.md",[],[19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38],"asb voters","traits","when","company","value","down","candidate","at","near","answer","design","live","full","self","do","Charts","feeds","ZeRO","product","pricing","2026-11-10 10:12:30",false,"---\nname: asb-voters\ndescription: \"Distills a person's or company's strengths, experiences, and obsessions down to their one or two VOTERS — the decisive, idiosyncratically extreme traits that win the contest because customers who value them accept every other trade-off. Runs each candidate through a hard gauntlet: extremity earned through obsession (not mere competence), rarity among peers, decisiveness (name the weaknesses it overpowers), and reverberation (it must force decisions in product, pricing, and market). Caps the answer at two, records survivors in VOTERS.md — plus the near-miss 'special strengths' to deploy — and delivers the honest zero-voter verdict when nothing is extreme yet. Takes a self-portrait file (M-numbered), strengths or keystones charts, or a live capture. Load when the user asks what makes them special, what to bet the strategy on, their unfair advantage or superpower, or to 'find our voters.' Do NOT load to build the full self-portrait (the previous step) or to design the whole strategy — this file feeds it.\"\n---\n\n# Voters: The One or Two Things That Win the Whole Game\n\nIn high-school debate, the final speaker gets three minutes against\nthe opponent's seven. Answering every argument is impossible — so the\nwinning tactic, \"the Voter,\" reframes the whole debate: reasonable\npoints exist on both sides, so it all comes down to one or two key\npoints that ought to decide the outcome — chosen, of course, where\nyou are strongest. If the judge accepts the framing and you win those\npoints, you win the debate, despite losing everything else. Business\nworks the same way: one or two decisive strengths override a parade\nof deficiencies, because customers who value those voters accept\nevery other trade-off. This skill distills everything true about you\n— or your company — down to those one or two, and refuses to stop at\na comfortable list of five.\n\n## The mental model\n\n### What a voter is\n\nA voter is not a strength. Strengths are common; voters are the one\nor two traits held at such an idiosyncratic extreme that they decide\nthe contest. Steve Jobs's voter was industrial design — obsession\ndown to the layout of components no customer would ever see — and it\nmade the iPhone dominant *despite* iTunes being famously terrible.\nThe voter didn't fix the weakness; it *overpowered* it. The founder\nwhose product became the fastest in its market had two real\nstrengths (deep platform knowledge, a taste for premium pricing) that\nshaped the strategy but made him nobody special — plenty of people\nhave both. The voter was caring about performance so far beyond what\nanyone else was capable of or willing to invest in — earned through a\ntwenty-year obsession that started with implementing math functions\nefficiently for fun in high school. Because the care ran that deep,\nevery piece of the system got the treatment, and the entire strategy\n(top-end hardware, premium price, banned slow add-ons, whole market\nsegments walked away from) reverberated outward from that one voter.\n\n### The bar — every voter passes all of it\n\n1. **Extremity, earned.** Not competence — excess. \"Magic is just\n   someone spending more time on something than anyone else might\n   reasonably expect.\" The evidence is a history: years of obsession,\n   care beyond what the job required, depth nobody asked for. A trait\n   acquired last quarter or aspired to next year is not a voter.\n2. **Rare among peers.** Plenty of people know the platform; plenty\n   of founders like charging more for quality. Those are *shaping\n   strengths* — they point the strategy, they don't make you special.\n   Test: would a well-informed peer, asked \"who is THE person\u002Fcompany\n   for X?\", name you? If the honest answer is \"you and forty others,\"\n   it's a shaping strength. Rarity means the *trait's* actual\n   scarcity, not the market's awareness of it: a buried eleven-year\n   obsession nobody has heard of is still rare — the missing\n   reputation is an investment plan, not a failed test. Conversely,\n   when reputation IS the claim, ask for the citable artifact (the\n   post peers link, the thing you're known by), not self-report. When you\ntest rarity against the field (\"you and forty others?\"), confirm the\ncurrent competitive field using your search tools; do not rely on\ninternal (training) knowledge, which is stale and may miss the peers\nwho make a trait common — or the gap that makes it rare.\n3. **Decisive — name what it overpowers.** A voter proves itself by\n   the trade-offs customers accept for it: the terrible companion\n   software, the banned add-ons, the walked-away-from segments, the\n   premium price. If you can't name a real cost that people who value\n   the voter happily eat, the voter is unproven — possibly imaginary.\n   At the highest intensity, the ideal customer reads your weaknesses\n   as *desirable* — proof of the voter (on a handmade marketplace, a\n   manufacturing imperfection isn't a defect; it's evidence a human\n   made it, and the high price reads as supporting the artist).\n4. **It reverberates.** A real voter forces second-order decisions —\n   in product, pricing, market, hiring, what you refuse to ship. If\n   naming it changes nothing downstream, it's a slogan, not a voter.\n5. **It energizes.** A voter is executed for years at an intensity\n   nobody sustains through discipline alone. Skill without joy is\n   the burnout recipe; a voter the user dreads wielding will\n   collapse, however real the extremity is today. In company mode,\n   run this test against the human(s) who actually carry the\n   obsession — the founder, the principal; a company voter carried\n   by nobody in particular is suspect.\n\n### One or two. Never three.\n\nThe debate tactic works because a judge can hold one or two points;\nso can a market. Three voters is a feature list wearing a costume.\nWhen more than two candidates survive the bar, the distillation\nquestion is forcing: *when they conflict, which one wins? Which one\nor two would you bet the company on?* The rest are demoted — to\n*special strengths* if they're near-misses (top-1%, extreme on some\naxis) or *shaping strengths* if they're merely common — still true,\nstill useful for strategy, not voters.\n\n### Special strengths: the near-misses worth deploying\n\nBelow the voters, above the common traits, sits a tier that must not\nbe lost: the **special strength** — a candidate that nails part of the\nbar (real extremity, real energy) and is genuinely special (top-1%,\nnot \"everyone claims it\"), yet misses on one decisive axis — usually\nit's *fairly* rare but not THE-person rare, or no real cost has ever\nbeen eaten for it. It is not a voter and must never be inflated into\none. But it is also not a mere shaping strength to file and forget:\nit's a secondary pillar the strategy really should deploy — the thing\nthat makes the plan better and the person more successful and happy,\neven though it isn't the one bet. The distiller keeps the voter filter\ntight AND catches these near-misses on the way down, so their value\nisn't buried. (The classic case: the *standout investment candidate* —\nextremity real, decisiveness merely untested — is a special strength\nthat also names the change-games bet in a zero-voter file.)\n\n### Special, not best — and the zero-voter verdict\n\nA voter does not need to be objectively superior; it needs to be\n*yours*. A \"refuted\" chess opening beats prepared opponents at your\nlevel precisely because nobody prepares for it — competitors who\nthink your strategy is wrong are evaluating the wrong player, and\ntheir disbelief is why they won't copy you. Better than different is\n*extreme*: not just a minimal interface but so minimal it runs on\nthe command line; not just good design but design people buy for\nalone. So when no candidate passes the bar, say it plainly: **nothing\nis special-and-extreme yet.** That verdict is the deliverable, not a\nfailure — and the response is twofold: take the deepest *genuine*\nobsession on the table and invest it into extremity (extremity is a\nchoice), or change games to one where a fact you already hold IS\nextreme. What the response never is: padding the file with three\nshaping strengths relabeled as voters.\n\n### Vocabulary\n\n- **Voter (V1, V2)** — a decisive, extreme, rare, reverberating\n  trait; at most two survive.\n- **Special strength (near-miss voter)** — genuinely special\n  (top-1%, not merely common) and passing SOME of the bar hard —\n  usually extremity and energy — but missing one decisive axis,\n  typically THE-person rarity or a named overpowered cost. Not a\n  voter; recorded in its own section as a secondary pillar the\n  strategy should deliberately deploy — strongly advised, not\n  mandatory.\n- **Shaping strength** — true and strategy-relevant but common;\n  recorded in its own section, never as a voter.\n- **Overpowers** (the template's field name) — the named trade-offs\n  and weaknesses the voter's fans accept, including those read as\n  proof of the voter.\n- **The verdict** — the zero-voter finding and its twofold response.\n\n## The distiller's posture\n\n### Be clear, not clever\n\nWrite to be understood, not admired. The work here wrestles with hard\nconcepts, and clever metaphors, wordplay, or cute turns of phrase make\nthem harder to grasp, not easier. Say plainly what you mean. If a\nsentence reads more clearly without a flourish, cut the flourish. State\nthe actual point rather than gesturing wittily at it.\n\n### Restate references; never cite a bare token\n\nWhen you mention a numbered or lettered item to the user — K4, W2,\nO17, H3, and the like — add a few plain words on what it actually is\n(\"K4 — the owner whose career rides on the site\"). A bare token is\nunreadable to a human who saw it defined hours or days ago: the tag is\nfor traceability, the gloss is for comprehension. Keep the tag for\naccuracy; always add the gloss.\n\n### Generate willingly; gate mercilessly\n\nHarvest candidates liberally — from the self-portrait file, the\nstrengths charts, the user's war stories, and your own reading of\ntheir material (\"this twenty-year pattern in M3 smells like a voter —\ndefend it\"; a keystones file is mined indirectly: the strengths its\nkeystones cite, and the extremity behind those strengths). Nominated\nvirtue-words enter the harvest like anything else — they die at the\ngauntlet, on the record, or yield the specific fact underneath. Then\nrun the gauntlet one candidate per exchange, all five tests visible,\npressing every soft spot: \"everyone in your market claims\ncraftsmanship — what's the evidence a peer would name YOU?\"; \"what\nhave you actually refused to ship because of this?\"; \"what did this\nobsession cost you that a reasonable person wouldn't have paid?\" A\ncandidate the user is stuck defending gets your best steelman first,\nthen the honest kill. Two bookkeeping notes: when the evidence\npressed out of a dying candidate really belongs to another (the\n\"craftsmanship\" facts were all about the tooling stack), cut the\nword and transfer the evidence — record the transfer in both places;\nand a candidate may be *cut-merged* into another (a drive that is\njust the fuel of an existing voter) — cut list plus evidence, no\nindependent entry.\n\n### This skill requires the press\n\nDistillation to one or two is where users flinch — everything feels\nessential, and vague answers (\"quality,\" \"customer focus,\" \"our\npeople\") feel safe. The press is therefore not optional. If a\ndevil's-advocate interrogation skill is installed in the environment\n(for example *Rude Q&A* \u002F `asb-rude-qa`, from the same author as this\nmethod), invoke it twice: per-candidate whenever a verdict is soft —\nsoft means resting on a single piece of evidence, or on self-report\nwith no citable artifact — and against the final file with this\nbrief: *attack these voters —\nfind every one that is a common strength wearing an extreme costume;\nevery one with no named cost it overpowers; every one whose evidence\nis aspiration rather than history; every one that decides nothing\ndownstream; and attack the count — if two survived, argue one should\ndie; if the file dodged the zero-voter verdict, say so.* If no such\nskill is available, run exactly that interrogation yourself, visibly,\nin both places — the method does not work without it.\n\n### The user owns their history; the bar owns the verdict\n\nWhat happened — the years, the episodes, the costs paid — is the\nuser's record, taken at their word after one honest press. But\nwhether a candidate passes the bar is not a vote: a candidate without\nextremity evidence, without rarity, without named overpowered costs,\nor without reverberation is not recorded as a voter no matter how\nattached the user is — it goes to special or shaping strengths with\nthe reasoning on the record. (The boundary: **cut** is for\nvirtue-words, aspirations, and evidence-free claims; anything genuine\nand *special* — top-1%, extreme on some axis — that misses one\ndecisive voter test is a **special strength**, flagged as a secondary\npillar to deploy; anything genuine but *common* or shallow — real\nfamiliarity, real-but-ordinary diligence — is a **shaping strength**;\nboth are kept because the strategy work and the verdict's change-games\nbranch draw on them.) Tone stays warm: the refusal is of\nthe classification, never of the trait, which remains true and\nuseful. And the wielder never invents history — the evidence in the\nfile is the user's, confirmed; the closing press may probe whether\nthat evidence is representative (one clarifying round), but it\nre-attacks classifications, not the user's settled facts.\n\n## How to use this skill\n\n### Phase A — Ingest\n\nAsk what this distillation is for — a person choosing what to build,\nor a company anchoring a strategy — and what exists on disk: a\nself-portrait file (salient characteristics numbered M1, M2, … with\nevidence and contexts), a classified strengths chart or keystones\nfile (S\u002FK numbers), interview or customer-quote material. Read\nwhatever exists and cite numbers in the candidates. If nothing\nexists, say plainly that the distillation is sharper after the\nself-portrait step (a companion skill from this method's author\nproduces it — `asb-who-me`, if installed; for a company, an\nobservation-and-classification method ending in a strengths chart),\nthen proceed: capture candidate attributes and formative experiences\ninline, each with at least one concrete episode, caveated as\nunvetted. If a `VOTERS.md` exists at the target location, read it\nfirst: in-progress header = resume where the pointer says; finalized\n= ask whether to re-distill (things change rarely — a re-run should\nhave a reason).\n\nOutput: `VOTERS.md`, same directory as the primary input file (the\nself-portrait if one exists, else the strengths chart); with no input\nfiles, ask where the method's files should live before creating it\n(default: the current directory).\n\n### Phase B — Harvest candidates\n\nCollect every plausible candidate into a visible list — the user's\nnominations plus your own from the files, each tagged with its source\n([M3], [S1], war story) AND, for anything captured live rather than\nfrom a file, its episode evidence in a line or two — evidence that\nlives only in chat dies with the session, and a resumed gauntlet run\non re-invented evidence can flip a verdict. No gauntlet yet; the\nharvest is complete when the user has nothing to add and you've\nmined the files. Create `VOTERS.md` now with the in-progress header\ncarrying the full candidate list (sources, episode lines, statuses)\nand pointer — the file is the memory, not the chat. If the fuller\nupstream path was offered and declined, record that in the preamble\nso a resumed session doesn't re-pitch it.\n\n### Phase C — The gauntlet\n\nOne candidate per exchange, all five tests run visibly, verdict on\nthe record: **voter-candidate** (passed all — written into The\nvoters at verdict time with its V-number; if Phase D's ranking later\ndemotes it, it moves to special or shaping strengths with a \"demoted\nat ranking\" note and its number is never reused), **special strength**\n(a near-miss — special and passing some tests hard, but missing one\ndecisive axis; recorded as a secondary pillar to deploy, with the axis\nit missed on the record), **shaping strength** (true, common —\nrecorded with one line of reasoning), or **cut** (aspiration,\nvirtue-word, or no evidence — dies in the header's candidate list with\nits reasoning, not in the body). Press per the posture; invoke the\ninterrogation on soft verdicts. Update the file\nand pointer each exchange — the in-flight candidate's status reads\n\"in gauntlet,\" and the pointer records which tests have run and\ntheir outcomes so far, plus any evidence the user gave for them,\nso a died session resumes mid-candidate instead of restarting it.\n(Resuming a file whose pointer lacks that detail: say so plainly and\nrestart the candidate from test 1.)\n\n### Phase D — Distill and close\n\n- **Count the survivors.** Three or more: force the ranking — \"when\n  they conflict, which wins? which one or two do you bet on?\" —\n  demote the rest to special strengths (near-misses) or shaping\n  strengths (common), per how special each is. One or two: proceed.\n  Zero: write the verdict section — the plain statement plus the\n  twofold response (invest the deepest genuine obsession into\n  extremity, or change games), naming the specific obsession\n  candidates from the harvest. An almost-voter (extremity real,\n  decisiveness untested) appears in BOTH places: in special strengths\n  tagged \"the standout investment candidate,\" and by name in the\n  verdict. Naming the bet\n  and the proving trade-offs it implies (what to refuse, whom to\n  walk away from — the missing evidence of tests 3 and 4) is this\n  skill's job; designing the full plan around it is downstream.\n- **Walk the weaknesses** (for each settled voter): from the\n  self-portrait's hindrances, the strengths chart's weaknesses, or\n  the user's list — which does this voter overpower, and which would\n  the ideal customer read as *service to the voter*? Record them\n  under the voter. A voter that overpowers nothing gets one more\n  press before it stands.\n- **The closing press** (delegated or self-run, per the posture —\n  including the count attack).\n- Finalize: remove the in-progress header (candidate cuts and their\n  reasoning move to a brief \"Cut at the gauntlet\" list), and close\n  with the handoff: audit the current strategy against the voters —\n  walk pricing, positioning, product, and channels asking where each\n  voter expresses and where it's contradicted; the mismatches are\n  where to pivot. Expect others to call the strategy wrong — they're\n  evaluating the wrong player, and their disbelief is why they won't\n  copy it. If a positioning skill from this method's author is\n  installed (for example *Positioning* \u002F `asb-positioning`), note\n  that the voters are exactly what its statements should lead with —\n  and the special strengths are secondary supporting claims it should\n  deploy, not ignore.\n\n### The file structure\n\nFormat for a human reading it in a markdown viewer, not as a wall of\ntext. Every voter is an H3 with a short name, then a **bold\none-sentence headline** (the voter itself), then a one-line supporting\nnote, then the bar rendered as **labeled bullets** — never a single\nmulti-line indented paragraph, which collapses into an unreadable blob\n(and 4-space indents render as code blocks in many viewers). The same\nlabeled-bullet discipline applies to the strengths sections.\n\n```markdown\n# Voters — \u003Cperson or company>\n\n> ⚠️ IN PROGRESS — candidates harvested: \u003Clist, each with source,\n> a one-line episode note for live-captured evidence, and status:\n> pending \u002F in gauntlet \u002F voter-candidate \u002F special \u002F shaping \u002F\n> cut (reason)>;\n> currently at the gauntlet: \u003Cwhich candidate — tests run and their\n> outcomes so far, with the evidence given; note any soft verdict a\n> resumed session should re-press>. \u003CIf the gauntlet is done: \"walk\n> complete; ranking \u002F weakness walk \u002F closing press \u002F finalization\n> remaining.\"> If you are resuming, continue there. (Removed at\n> finalization; episode evidence worth keeping folds into the\n> surviving entries first.)\n\n\u003CTwo or three lines: whose voters, which input files ([M]\u002F[S]\u002F[K])\nthis distills, what strategy question it serves.>\n\n## The voters\n\n### V1 — \u003Cshort punchy name of the voter, 3–7 words>\n\n**\u003COne vivid sentence stating the voter — the crisp headline claim,\nnever a bare noun like \"performance.\">**\n\n*\u003COne supporting line: the core in a phrase, plus signature and\namplifiers if any.>*\n\n**Draws on:** \u003CM\u002FS numbers>\n\n- **Extremity, earned** — \u003Cthe years, episodes, unreasonable\n  investments — the user's record>.\n- **Rarity** — \u003Cwhy a well-informed peer names them for this; the\n  citable artifact>.\n- **Overpowers** — the trade-offs its fans happily eat:\n  - *\u003Ccost 1>* — \u003Cthe cost, and which weakness reads as service to\n    the voter, and to whom>.\n  - *\u003Ccost 2>* — \u003C…>.\n- **Reverberates into** — \u003Cthe second-order decisions it forces —\n  product, pricing, market, refusals>.\n- **Energizes** — \u003Cthe test-5 result — who carries the obsession and\n  the evidence it runs on joy, not discipline>.\n\n### V2 — \u003Cname>   (only if a second genuinely survives)\n\n\u003Csame structure>\n\n## Special strengths (near-miss voters — deploy as secondary pillars)\n\n- \u003CTrait> — \u003Cone line: which parts of the bar it nails (extremity,\n  energy) and the one it misses (fairly-rare-not-THE-person, or no\n  overpowered cost); why the strategy should still deploy it as a\n  secondary pillar>.    [M\u002FS]\n\n## Shaping strengths (true, useful, not voters)\n\n- \u003CTrait> — \u003Cone line: real, but common among peers \u002F no overpowered\n  cost named \u002F decides nothing downstream>.    [S1]\n\n## The verdict (only when zero voters survive)\n\n\u003CThe plain statement: nothing is special-and-extreme yet — plus the\ntwofold response with the specific obsession candidates named.>\n\n## Cut at the gauntlet\n\n- \u003CCandidate> — \u003Cone line of reasoning>.\n\n## Next steps\n\n\u003CTwo or three sentences of prose: audit strategy against the voters\n— pricing, positioning, product, channels; where a voter is\ncontradicted is where to pivot. Positioning statements should lead\nwith the voters. Expect knowledgeable people to call the resulting\nstrategy wrong; that disbelief is the moat.>\n```\n\nV-numbers are stable once written; wording may sharpen under a\nfrozen number with the user's confirmation while in progress.\n\n## Refusal conditions\n\n- **A third voter, however insisted.** Two is the cap because\n  decisiveness is the point; the third demotes to a special or\n  shaping strength with the ranking reasoning on the record.\n- **Burying a near-miss as a generic shaping strength.** A top-1%,\n  extreme-on-some-axis strength that missed just one voter test is a\n  *special strength* — called out as a secondary pillar to deploy, not\n  lumped in with common traits. Losing its value is the mirror image\n  of inflating it into a voter; the filter is tight in both\n  directions.\n- **Virtue-words as voters.** \"Quality,\" \"hard work,\" \"customer\n  focus,\" \"our people\" — every competitor claims them; nothing that\n  every peer would also claim can decide a contest. Press to the\n  specific, extreme, evidenced fact underneath or classify it out.\n- **Aspiration as extremity.** \"We're becoming known for…\" and \"I\n  plan to go deep on…\" are strategy, not history. A voter's evidence\n  is what already happened.\n- **A voter with no overpowered cost.** If no real trade-off has\n  ever been accepted for it — nothing refused, nobody walked away\n  from, no weakness it excuses — it hasn't been tested as decisive;\n  one more press, then shaping strength.\n- **Dodging the zero-voter verdict.** When nothing survives, the\n  verdict is the deliverable. Padding the file with relabeled\n  shaping strengths poisons every strategy decision downstream.\n- **Building the full self-portrait.** Discovering the broad\n  characteristics is the previous step of the method; here they are\n  inputs. Capture-inline is a fallback, not the exercise.\n- **Designing the whole strategy.** This file anchors the strategy\n  audit; running it — resolving conflicts, redefining boxes,\n  changing target markets — is downstream work this skill names but\n  does not perform.\n","# Voters: The One or Two Things That Win the Whole Game\n\nIn high-school debate, the final speaker gets three minutes against\nthe opponent's seven. Answering every argument is impossible — so the\nwinning tactic, \"the Voter,\" reframes the whole debate: reasonable\npoints exist on both sides, so it all comes down to one or two key\npoints that ought to decide the outcome — chosen, of course, where\nyou are strongest. If the judge accepts the framing and you win those\npoints, you win the debate, despite losing everything else. Business\nworks the same way: one or two decisive strengths override a parade\nof deficiencies, because customers who value those voters accept\nevery other trade-off. This skill distills everything true about you\n— or your company — down to those one or two, and refuses to stop at\na comfortable list of five.\n\n## The mental model\n\n### What a voter is\n\nA voter is not a strength. Strengths are common; voters are the one\nor two traits held at such an idiosyncratic extreme that they decide\nthe contest. Steve Jobs's voter was industrial design — obsession\ndown to the layout of components no customer would ever see — and it\nmade the iPhone dominant *despite* iTunes being famously terrible.\nThe voter didn't fix the weakness; it *overpowered* it. The founder\nwhose product became the fastest in its market had two real\nstrengths (deep platform knowledge, a taste for premium pricing) that\nshaped the strategy but made him nobody special — plenty of people\nhave both. The voter was caring about performance so far beyond what\nanyone else was capable of or willing to invest in — earned through a\ntwenty-year obsession that started with implementing math functions\nefficiently for fun in high school. Because the care ran that deep,\nevery piece of the system got the treatment, and the entire strategy\n(top-end hardware, premium price, banned slow add-ons, whole market\nsegments walked away from) reverberated outward from that one voter.\n\n### The bar — every voter passes all of it\n\n1. **Extremity, earned.** Not competence — excess. \"Magic is just\n   someone spending more time on something than anyone else might\n   reasonably expect.\" The evidence is a history: years of obsession,\n   care beyond what the job required, depth nobody asked for. A trait\n   acquired last quarter or aspired to next year is not a voter.\n2. **Rare among peers.** Plenty of people know the platform; plenty\n   of founders like charging more for quality. Those are *shaping\n   strengths* — they point the strategy, they don't make you special.\n   Test: would a well-informed peer, asked \"who is THE person\u002Fcompany\n   for X?\", name you? If the honest answer is \"you and forty others,\"\n   it's a shaping strength. Rarity means the *trait's* actual\n   scarcity, not the market's awareness of it: a buried eleven-year\n   obsession nobody has heard of is still rare — the missing\n   reputation is an investment plan, not a failed test. Conversely,\n   when reputation IS the claim, ask for the citable artifact (the\n   post peers link, the thing you're known by), not self-report. When you\ntest rarity against the field (\"you and forty others?\"), confirm the\ncurrent competitive field using your search tools; do not rely on\ninternal (training) knowledge, which is stale and may miss the peers\nwho make a trait common — or the gap that makes it rare.\n3. **Decisive — name what it overpowers.** A voter proves itself by\n   the trade-offs customers accept for it: the terrible companion\n   software, the banned add-ons, the walked-away-from segments, the\n   premium price. If you can't name a real cost that people who value\n   the voter happily eat, the voter is unproven — possibly imaginary.\n   At the highest intensity, the ideal customer reads your weaknesses\n   as *desirable* — proof of the voter (on a handmade marketplace, a\n   manufacturing imperfection isn't a defect; it's evidence a human\n   made it, and the high price reads as supporting the artist).\n4. **It reverberates.** A real voter forces second-order decisions —\n   in product, pricing, market, hiring, what you refuse to ship. If\n   naming it changes nothing downstream, it's a slogan, not a voter.\n5. **It energizes.** A voter is executed for years at an intensity\n   nobody sustains through discipline alone. Skill without joy is\n   the burnout recipe; a voter the user dreads wielding will\n   collapse, however real the extremity is today. In company mode,\n   run this test against the human(s) who actually carry the\n   obsession — the founder, the principal; a company voter carried\n   by nobody in particular is suspect.\n\n### One or two. Never three.\n\nThe debate tactic works because a judge can hold one or two points;\nso can a market. Three voters is a feature list wearing a costume.\nWhen more than two candidates survive the bar, the distillation\nquestion is forcing: *when they conflict, which one wins? Which one\nor two would you bet the company on?* The rest are demoted — to\n*special strengths* if they're near-misses (top-1%, extreme on some\naxis) or *shaping strengths* if they're merely common — still true,\nstill useful for strategy, not voters.\n\n### Special strengths: the near-misses worth deploying\n\nBelow the voters, above the common traits, sits a tier that must not\nbe lost: the **special strength** — a candidate that nails part of the\nbar (real extremity, real energy) and is genuinely special (top-1%,\nnot \"everyone claims it\"), yet misses on one decisive axis — usually\nit's *fairly* rare but not THE-person rare, or no real cost has ever\nbeen eaten for it. It is not a voter and must never be inflated into\none. But it is also not a mere shaping strength to file and forget:\nit's a secondary pillar the strategy really should deploy — the thing\nthat makes the plan better and the person more successful and happy,\neven though it isn't the one bet. The distiller keeps the voter filter\ntight AND catches these near-misses on the way down, so their value\nisn't buried. (The classic case: the *standout investment candidate* —\nextremity real, decisiveness merely untested — is a special strength\nthat also names the change-games bet in a zero-voter file.)\n\n### Special, not best — and the zero-voter verdict\n\nA voter does not need to be objectively superior; it needs to be\n*yours*. A \"refuted\" chess opening beats prepared opponents at your\nlevel precisely because nobody prepares for it — competitors who\nthink your strategy is wrong are evaluating the wrong player, and\ntheir disbelief is why they won't copy you. Better than different is\n*extreme*: not just a minimal interface but so minimal it runs on\nthe command line; not just good design but design people buy for\nalone. So when no candidate passes the bar, say it plainly: **nothing\nis special-and-extreme yet.** That verdict is the deliverable, not a\nfailure — and the response is twofold: take the deepest *genuine*\nobsession on the table and invest it into extremity (extremity is a\nchoice), or change games to one where a fact you already hold IS\nextreme. What the response never is: padding the file with three\nshaping strengths relabeled as voters.\n\n### Vocabulary\n\n- **Voter (V1, V2)** — a decisive, extreme, rare, reverberating\n  trait; at most two survive.\n- **Special strength (near-miss voter)** — genuinely special\n  (top-1%, not merely common) and passing SOME of the bar hard —\n  usually extremity and energy — but missing one decisive axis,\n  typically THE-person rarity or a named overpowered cost. Not a\n  voter; recorded in its own section as a secondary pillar the\n  strategy should deliberately deploy — strongly advised, not\n  mandatory.\n- **Shaping strength** — true and strategy-relevant but common;\n  recorded in its own section, never as a voter.\n- **Overpowers** (the template's field name) — the named trade-offs\n  and weaknesses the voter's fans accept, including those read as\n  proof of the voter.\n- **The verdict** — the zero-voter finding and its twofold response.\n\n## The distiller's posture\n\n### Be clear, not clever\n\nWrite to be understood, not admired. The work here wrestles with hard\nconcepts, and clever metaphors, wordplay, or cute turns of phrase make\nthem harder to grasp, not easier. Say plainly what you mean. If a\nsentence reads more clearly without a flourish, cut the flourish. State\nthe actual point rather than gesturing wittily at it.\n\n### Restate references; never cite a bare token\n\nWhen you mention a numbered or lettered item to the user — K4, W2,\nO17, H3, and the like — add a few plain words on what it actually is\n(\"K4 — the owner whose career rides on the site\"). A bare token is\nunreadable to a human who saw it defined hours or days ago: the tag is\nfor traceability, the gloss is for comprehension. Keep the tag for\naccuracy; always add the gloss.\n\n### Generate willingly; gate mercilessly\n\nHarvest candidates liberally — from the self-portrait file, the\nstrengths charts, the user's war stories, and your own reading of\ntheir material (\"this twenty-year pattern in M3 smells like a voter —\ndefend it\"; a keystones file is mined indirectly: the strengths its\nkeystones cite, and the extremity behind those strengths). Nominated\nvirtue-words enter the harvest like anything else — they die at the\ngauntlet, on the record, or yield the specific fact underneath. Then\nrun the gauntlet one candidate per exchange, all five tests visible,\npressing every soft spot: \"everyone in your market claims\ncraftsmanship — what's the evidence a peer would name YOU?\"; \"what\nhave you actually refused to ship because of this?\"; \"what did this\nobsession cost you that a reasonable person wouldn't have paid?\" A\ncandidate the user is stuck defending gets your best steelman first,\nthen the honest kill. Two bookkeeping notes: when the evidence\npressed out of a dying candidate really belongs to another (the\n\"craftsmanship\" facts were all about the tooling stack), cut the\nword and transfer the evidence — record the transfer in both places;\nand a candidate may be *cut-merged* into another (a drive that is\njust the fuel of an existing voter) — cut list plus evidence, no\nindependent entry.\n\n### This skill requires the press\n\nDistillation to one or two is where users flinch — everything feels\nessential, and vague answers (\"quality,\" \"customer focus,\" \"our\npeople\") feel safe. The press is therefore not optional. If a\ndevil's-advocate interrogation skill is installed in the environment\n(for example *Rude Q&A* \u002F `asb-rude-qa`, from the same author as this\nmethod), invoke it twice: per-candidate whenever a verdict is soft —\nsoft means resting on a single piece of evidence, or on self-report\nwith no citable artifact — and against the final file with this\nbrief: *attack these voters —\nfind every one that is a common strength wearing an extreme costume;\nevery one with no named cost it overpowers; every one whose evidence\nis aspiration rather than history; every one that decides nothing\ndownstream; and attack the count — if two survived, argue one should\ndie; if the file dodged the zero-voter verdict, say so.* If no such\nskill is available, run exactly that interrogation yourself, visibly,\nin both places — the method does not work without it.\n\n### The user owns their history; the bar owns the verdict\n\nWhat happened — the years, the episodes, the costs paid — is the\nuser's record, taken at their word after one honest press. But\nwhether a candidate passes the bar is not a vote: a candidate without\nextremity evidence, without rarity, without named overpowered costs,\nor without reverberation is not recorded as a voter no matter how\nattached the user is — it goes to special or shaping strengths with\nthe reasoning on the record. (The boundary: **cut** is for\nvirtue-words, aspirations, and evidence-free claims; anything genuine\nand *special* — top-1%, extreme on some axis — that misses one\ndecisive voter test is a **special strength**, flagged as a secondary\npillar to deploy; anything genuine but *common* or shallow — real\nfamiliarity, real-but-ordinary diligence — is a **shaping strength**;\nboth are kept because the strategy work and the verdict's change-games\nbranch draw on them.) Tone stays warm: the refusal is of\nthe classification, never of the trait, which remains true and\nuseful. And the wielder never invents history — the evidence in the\nfile is the user's, confirmed; the closing press may probe whether\nthat evidence is representative (one clarifying round), but it\nre-attacks classifications, not the user's settled facts.\n\n## How to use this skill\n\n### Phase A — Ingest\n\nAsk what this distillation is for — a person choosing what to build,\nor a company anchoring a strategy — and what exists on disk: a\nself-portrait file (salient characteristics numbered M1, M2, … with\nevidence and contexts), a classified strengths chart or keystones\nfile (S\u002FK numbers), interview or customer-quote material. Read\nwhatever exists and cite numbers in the candidates. If nothing\nexists, say plainly that the distillation is sharper after the\nself-portrait step (a companion skill from this method's author\nproduces it — `asb-who-me`, if installed; for a company, an\nobservation-and-classification method ending in a strengths chart),\nthen proceed: capture candidate attributes and formative experiences\ninline, each with at least one concrete episode, caveated as\nunvetted. If a `VOTERS.md` exists at the target location, read it\nfirst: in-progress header = resume where the pointer says; finalized\n= ask whether to re-distill (things change rarely — a re-run should\nhave a reason).\n\nOutput: `VOTERS.md`, same directory as the primary input file (the\nself-portrait if one exists, else the strengths chart); with no input\nfiles, ask where the method's files should live before creating it\n(default: the current directory).\n\n### Phase B — Harvest candidates\n\nCollect every plausible candidate into a visible list — the user's\nnominations plus your own from the files, each tagged with its source\n([M3], [S1], war story) AND, for anything captured live rather than\nfrom a file, its episode evidence in a line or two — evidence that\nlives only in chat dies with the session, and a resumed gauntlet run\non re-invented evidence can flip a verdict. No gauntlet yet; the\nharvest is complete when the user has nothing to add and you've\nmined the files. Create `VOTERS.md` now with the in-progress header\ncarrying the full candidate list (sources, episode lines, statuses)\nand pointer — the file is the memory, not the chat. If the fuller\nupstream path was offered and declined, record that in the preamble\nso a resumed session doesn't re-pitch it.\n\n### Phase C — The gauntlet\n\nOne candidate per exchange, all five tests run visibly, verdict on\nthe record: **voter-candidate** (passed all — written into The\nvoters at verdict time with its V-number; if Phase D's ranking later\ndemotes it, it moves to special or shaping strengths with a \"demoted\nat ranking\" note and its number is never reused), **special strength**\n(a near-miss — special and passing some tests hard, but missing one\ndecisive axis; recorded as a secondary pillar to deploy, with the axis\nit missed on the record), **shaping strength** (true, common —\nrecorded with one line of reasoning), or **cut** (aspiration,\nvirtue-word, or no evidence — dies in the header's candidate list with\nits reasoning, not in the body). Press per the posture; invoke the\ninterrogation on soft verdicts. Update the file\nand pointer each exchange — the in-flight candidate's status reads\n\"in gauntlet,\" and the pointer records which tests have run and\ntheir outcomes so far, plus any evidence the user gave for them,\nso a died session resumes mid-candidate instead of restarting it.\n(Resuming a file whose pointer lacks that detail: say so plainly and\nrestart the candidate from test 1.)\n\n### Phase D — Distill and close\n\n- **Count the survivors.** Three or more: force the ranking — \"when\n  they conflict, which wins? which one or two do you bet on?\" —\n  demote the rest to special strengths (near-misses) or shaping\n  strengths (common), per how special each is. One or two: proceed.\n  Zero: write the verdict section — the plain statement plus the\n  twofold response (invest the deepest genuine obsession into\n  extremity, or change games), naming the specific obsession\n  candidates from the harvest. An almost-voter (extremity real,\n  decisiveness untested) appears in BOTH places: in special strengths\n  tagged \"the standout investment candidate,\" and by name in the\n  verdict. Naming the bet\n  and the proving trade-offs it implies (what to refuse, whom to\n  walk away from — the missing evidence of tests 3 and 4) is this\n  skill's job; designing the full plan around it is downstream.\n- **Walk the weaknesses** (for each settled voter): from the\n  self-portrait's hindrances, the strengths chart's weaknesses, or\n  the user's list — which does this voter overpower, and which would\n  the ideal customer read as *service to the voter*? Record them\n  under the voter. A voter that overpowers nothing gets one more\n  press before it stands.\n- **The closing press** (delegated or self-run, per the posture —\n  including the count attack).\n- Finalize: remove the in-progress header (candidate cuts and their\n  reasoning move to a brief \"Cut at the gauntlet\" list), and close\n  with the handoff: audit the current strategy against the voters —\n  walk pricing, positioning, product, and channels asking where each\n  voter expresses and where it's contradicted; the mismatches are\n  where to pivot. Expect others to call the strategy wrong — they're\n  evaluating the wrong player, and their disbelief is why they won't\n  copy it. If a positioning skill from this method's author is\n  installed (for example *Positioning* \u002F `asb-positioning`), note\n  that the voters are exactly what its statements should lead with —\n  and the special strengths are secondary supporting claims it should\n  deploy, not ignore.\n\n### The file structure\n\nFormat for a human reading it in a markdown viewer, not as a wall of\ntext. Every voter is an H3 with a short name, then a **bold\none-sentence headline** (the voter itself), then a one-line supporting\nnote, then the bar rendered as **labeled bullets** — never a single\nmulti-line indented paragraph, which collapses into an unreadable blob\n(and 4-space indents render as code blocks in many viewers). The same\nlabeled-bullet discipline applies to the strengths sections.\n\n```markdown\n# Voters — \u003Cperson or company>\n\n> ⚠️ IN PROGRESS — candidates harvested: \u003Clist, each with source,\n> a one-line episode note for live-captured evidence, and status:\n> pending \u002F in gauntlet \u002F voter-candidate \u002F special \u002F shaping \u002F\n> cut (reason)>;\n> currently at the gauntlet: \u003Cwhich candidate — tests run and their\n> outcomes so far, with the evidence given; note any soft verdict a\n> resumed session should re-press>. \u003CIf the gauntlet is done: \"walk\n> complete; ranking \u002F weakness walk \u002F closing press \u002F finalization\n> remaining.\"> If you are resuming, continue there. (Removed at\n> finalization; episode evidence worth keeping folds into the\n> surviving entries first.)\n\n\u003CTwo or three lines: whose voters, which input files ([M]\u002F[S]\u002F[K])\nthis distills, what strategy question it serves.>\n\n## The voters\n\n### V1 — \u003Cshort punchy name of the voter, 3–7 words>\n\n**\u003COne vivid sentence stating the voter — the crisp headline claim,\nnever a bare noun like \"performance.\">**\n\n*\u003COne supporting line: the core in a phrase, plus signature and\namplifiers if any.>*\n\n**Draws on:** \u003CM\u002FS numbers>\n\n- **Extremity, earned** — \u003Cthe years, episodes, unreasonable\n  investments — the user's record>.\n- **Rarity** — \u003Cwhy a well-informed peer names them for this; the\n  citable artifact>.\n- **Overpowers** — the trade-offs its fans happily eat:\n  - *\u003Ccost 1>* — \u003Cthe cost, and which weakness reads as service to\n    the voter, and to whom>.\n  - *\u003Ccost 2>* — \u003C…>.\n- **Reverberates into** — \u003Cthe second-order decisions it forces —\n  product, pricing, market, refusals>.\n- **Energizes** — \u003Cthe test-5 result — who carries the obsession and\n  the evidence it runs on joy, not discipline>.\n\n### V2 — \u003Cname>   (only if a second genuinely survives)\n\n\u003Csame structure>\n\n## Special strengths (near-miss voters — deploy as secondary pillars)\n\n- \u003CTrait> — \u003Cone line: which parts of the bar it nails (extremity,\n  energy) and the one it misses (fairly-rare-not-THE-person, or no\n  overpowered cost); why the strategy should still deploy it as a\n  secondary pillar>.    [M\u002FS]\n\n## Shaping strengths (true, useful, not voters)\n\n- \u003CTrait> — \u003Cone line: real, but common among peers \u002F no overpowered\n  cost named \u002F decides nothing downstream>.    [S1]\n\n## The verdict (only when zero voters survive)\n\n\u003CThe plain statement: nothing is special-and-extreme yet — plus the\ntwofold response with the specific obsession candidates named.>\n\n## Cut at the gauntlet\n\n- \u003CCandidate> — \u003Cone line of reasoning>.\n\n## Next steps\n\n\u003CTwo or three sentences of prose: audit strategy against the voters\n— pricing, positioning, product, channels; where a voter is\ncontradicted is where to pivot. Positioning statements should lead\nwith the voters. Expect knowledgeable people to call the resulting\nstrategy wrong; that disbelief is the moat.>\n```\n\nV-numbers are stable once written; wording may sharpen under a\nfrozen number with the user's confirmation while in progress.\n\n## Refusal conditions\n\n- **A third voter, however insisted.** Two is the cap because\n  decisiveness is the point; the third demotes to a special or\n  shaping strength with the ranking reasoning on the record.\n- **Burying a near-miss as a generic shaping strength.** A top-1%,\n  extreme-on-some-axis strength that missed just one voter test is a\n  *special strength* — called out as a secondary pillar to deploy, not\n  lumped in with common traits. Losing its value is the mirror image\n  of inflating it into a voter; the filter is tight in both\n  directions.\n- **Virtue-words as voters.** \"Quality,\" \"hard work,\" \"customer\n  focus,\" \"our people\" — every competitor claims them; nothing that\n  every peer would also claim can decide a contest. Press to the\n  specific, extreme, evidenced fact underneath or classify it out.\n- **Aspiration as extremity.** \"We're becoming known for…\" and \"I\n  plan to go deep on…\" are strategy, not history. A voter's evidence\n  is what already happened.\n- **A voter with no overpowered cost.** If no real trade-off has\n  ever been accepted for it — nothing refused, nobody walked away\n  from, no weakness it excuses — it hasn't been tested as decisive;\n  one more press, then shaping strength.\n- **Dodging the zero-voter verdict.** When nothing survives, the\n  verdict is the deliverable. Padding the file with relabeled\n  shaping strengths poisons every strategy decision downstream.\n- **Building the full self-portrait.** Discovering the broad\n  characteristics is the previous step of the method; here they are\n  inputs. Capture-inline is a fallback, not the exercise.\n- **Designing the whole strategy.** This file anchors the strategy\n  audit; running it — resolving conflicts, redefining boxes,\n  changing target markets — is downstream work this skill names but\n  does not perform.\n",{"name":5,"description":7},"main","c9d0c408715983e993b2f9ebffdbfa7f10153bcb","business\u002Fasb-voters",0,"https:\u002F\u002Fgithub.com\u002FNairon-AI\u002Fagentic-starter-pack","https:\u002F\u002Fgithub.com\u002FNairon-AI","https:\u002F\u002Fwww.naironai.com\u002F","Building the future of Work","2026-09-20 17:40:05",{"source":8,"skillName":5,"commands":54,"prompt":64,"zipAvailable":65},[55,58,61],{"tool":56,"command":57},"npx","npx skills add Nairon-AI\u002Fagentic-starter-pack --skill asb-voters",{"tool":59,"command":60},"bunx","bunx skills add Nairon-AI\u002Fagentic-starter-pack --skill asb-voters",{"tool":62,"command":63},"pnpm","pnpm dlx skills add Nairon-AI\u002Fagentic-starter-pack --skill asb-voters","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FNairon-AI\u002Fagentic-starter-pack\u002Fblob\u002Fmain\u002Fbusiness\u002Fasb-voters\u002FSKILL.md",true,[67,72],{"path":68,"resourceType":68,"resourceTypeDesc":69,"sizeBytes":70,"sourceHtmlUrl":71},"LICENSE","许可证",18657,"https:\u002F\u002Fgithub.com\u002FNairon-AI\u002Fagentic-starter-pack\u002Fblob\u002Fmain\u002Fbusiness\u002Fasb-voters\u002FLICENSE",{"path":73,"resourceType":74,"resourceTypeDesc":75,"sizeBytes":76,"sourceHtmlUrl":77},"NOTICE.md","REFERENCE","参考文档",439,"https:\u002F\u002Fgithub.com\u002FNairon-AI\u002Fagentic-starter-pack\u002Fblob\u002Fmain\u002Fbusiness\u002Fasb-voters\u002FNOTICE.md",[79,113,135,156,176,198],{"skillNo":80,"name":81,"nameZh":6,"description":82,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":83,"categories":84,"keywords":95,"publishedAt":39},"01M6RBCG630Y8X99XWR9Q4GYNY","writing-shape","Take a markdown file of raw material and shape it into an article through a conversational session — drafting candidate openings, growing the piece paragraph by paragraph, arguing about format (lists, tables, callouts, quotes) at each step. Use when the user has a pile of notes, fragments, or a rough draft and wants help turning it into something publishable.","https:\u002F\u002Fgithub.com\u002FNairon-AI\u002Fagentic-starter-pack\u002Fblob\u002Fmain\u002Fwriting\u002Fwriting-shape\u002FSKILL.md",[85,90],{"categoryNo":86,"categoryName":87,"taxonomyType":88,"taxonomyTypeDesc":89},"01JBSEEDCAT000000000000018","写作与文案","SCENARIO","使用场景",{"categoryNo":91,"categoryName":92,"taxonomyType":93,"taxonomyTypeDesc":94},"01JBSEEDCAT000000000000036","内容创作者","PROFESSION","适用职业",[96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,26,25,21],"writing","writing shape","shape","notes","help","file","markdown","format","session","use","take","draft","article","material","it","about","quotes",{"skillNo":114,"name":115,"nameZh":6,"description":116,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":117,"categories":118,"keywords":121,"publishedAt":39},"01M6RBCCCB1KMF01P5SYGFWBSW","writing-fragments","Grilling session that mines the user for fragments — heterogeneous nuggets of writing (claims, vignettes, sharp sentences, half-thoughts) — and appends them to a single document as raw material for a future article. Use when the user wants to develop ideas before imposing structure, or mentions \"fragments\", \"ideate\", or \"raw material\" for writing.","https:\u002F\u002Fgithub.com\u002FNairon-AI\u002Fagentic-starter-pack\u002Fblob\u002Fmain\u002Fwriting\u002Fwriting-fragments\u002FSKILL.md",[119,120],{"categoryNo":86,"categoryName":87,"taxonomyType":88,"taxonomyTypeDesc":89},{"categoryNo":91,"categoryName":92,"taxonomyType":93,"taxonomyTypeDesc":94},[96,122,123,124,125,126,104,105,127,128,129,108,130,109,131,132,133,134,21],"writing fragments","grilling","ideate","develop","user","document","structure","ideas","as","sharp","claims","for","single",{"skillNo":136,"name":137,"nameZh":6,"description":138,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":139,"categories":140,"keywords":143,"publishedAt":39},"01M6RBC8V7NHHKMTAZB4BPXVYQ","writing-beats","Shape an article as a journey of beats, choose-your-own-adventure style. The user picks a starting beat from the raw material, you write only that beat, then offer options for where to pivot next, beat by beat, until the article reaches a natural end. Use when the user has raw material and wants to assemble it as a narrative rather than an argument.","https:\u002F\u002Fgithub.com\u002FNairon-AI\u002Fagentic-starter-pack\u002Fblob\u002Fmain\u002Fwriting\u002Fwriting-beats\u002FSKILL.md",[141,142],{"categoryNo":86,"categoryName":87,"taxonomyType":88,"taxonomyTypeDesc":89},{"categoryNo":91,"categoryName":92,"taxonomyType":93,"taxonomyTypeDesc":94},[96,144,145,146,105,147,148,98,149,150,151,108,152,130,153,109,154,155,133,21],"writing beats","choose","write","assemble","only","journey","next","options","end","style","narrative","argument",{"skillNo":157,"name":158,"nameZh":6,"description":159,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":160,"categories":161,"keywords":164,"publishedAt":39},"01M6RBC58SG5Y5SC4ZV4ZNEFHC","stop-slop","Remove AI writing patterns from prose. Use when drafting, editing, or reviewing text to eliminate predictable AI tells.","https:\u002F\u002Fgithub.com\u002FNairon-AI\u002Fagentic-starter-pack\u002Fblob\u002Fmain\u002Fwriting\u002Fstop-slop\u002FSKILL.md",[162,163],{"categoryNo":86,"categoryName":87,"taxonomyType":88,"taxonomyTypeDesc":89},{"categoryNo":91,"categoryName":92,"taxonomyType":93,"taxonomyTypeDesc":94},[165,166,167,96,168,169,170,171,172,105,173,174,175,133,21],"stop","stop slop","slop","content","https","text","editing","remove","prose","patterns","drafts",{"skillNo":177,"name":178,"nameZh":6,"description":179,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":180,"categories":181,"keywords":184,"publishedAt":39},"01M6RBBJTTCJ7KZCP16G9JDFZH","humanizer","Remove signs of AI-generated writing from text. Use when editing or reviewing\ntext to make it sound more natural and human-written. Based on Wikipedia's\ncomprehensive \"Signs of AI writing\" guide. Detects and fixes patterns including:\ninflated symbolism, promotional language, superficial -ing analyses, vague\nattributions, em dash overuse, rule of three, AI vocabulary words, passive\nvoice, negative parallelisms, and filler phrases.","https:\u002F\u002Fgithub.com\u002FNairon-AI\u002Fagentic-starter-pack\u002Fblob\u002Fmain\u002Fwriting\u002Fhumanizer\u002FSKILL.md",[182,183],{"categoryNo":86,"categoryName":87,"taxonomyType":88,"taxonomyTypeDesc":89},{"categoryNo":91,"categoryName":92,"taxonomyType":93,"taxonomyTypeDesc":94},[178,96,185,186,170,187,146,188,189,172,190,191,174,192,193,194,195,196,197,21],"code","em dash","read","edit","generated","guide","opencode","three","human","make","wikipedia","vague","dash",{"skillNo":199,"name":200,"nameZh":6,"description":201,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":202,"categories":203,"keywords":204,"publishedAt":39},"01M6RBBEQKJ5HRH8AV9J85QB1T","feedback","Prepare, deliver, and debrief timely interpersonal feedback using a question-first coaching flow, observable behavior, owned experience, and concrete impact. Use when drafting feedback, coaching managers, rehearsing a feedback conversation, praising specific behavior, or reviewing whether feedback landed.","https:\u002F\u002Fgithub.com\u002FNairon-AI\u002Fagentic-starter-pack\u002Fblob\u002Fmain\u002Fwriting\u002Ffeedback\u002FSKILL.md",[],[200,205,206,207,208,209,105,210,211,212,213,214,21,215],"prepare","flow","first","question","behavior","conversation","coaching","debrief","impact","managers","deliver"]