[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6P8W0AXQJT3B062XA5J2JKY":3,"skill-same:01M6P8W0AXQJT3B062XA5J2JKY":69},{"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":23,"publishedAt":40,"linkOnly":41,"markdownContent":42,"markdownBody":43,"frontMatter":44,"refName":45,"commitSha":46,"skillDirectory":47,"forksCount":48,"repositoryHtmlUrl":49,"repositoryHomepageUrl":50,"ownerHtmlUrl":51,"ownerBlogUrl":52,"ownerDescription":53,"updatedAt":54,"install":55,"resources":68},"01M6P8W0AXQJT3B062XA5J2JKY","humanize",null,"Strip AI tells from blog prose and align it to the H·AI·K·U voice. Use after drafting or substantially editing any file under website\u002Fcontent\u002Fblog\u002F. Layered pass — voice rules win on conflict, then a generic AI-tell sweep, then a final \"what makes this obviously AI?\" self-audit.","TheBushidoCollective\u002Fhaiku-method","TheBushidoCollective","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F233175215?v=4","01M6P8SCRJW0J2C9YZ0D14R6MA","The Bushido Collective","01M6P8SCX4KGHF4C5GPWEV48SC",1,"Apache-2.0","https:\u002F\u002Fgithub.com\u002FTheBushidoCollective\u002Fhaiku-method\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fhumanize\u002FSKILL.md",[18],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000036","内容创作者","PROFESSION","适用职业",[5,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39],"editing","layered","it","conflict","self","align","prose","use","content","Rules","sweep","blog","voice","audit","file","website","2026-11-10 10:12:30",false,"---\nname: humanize\ndescription: Strip AI tells from blog prose and align it to the H·AI·K·U voice. Use after drafting or substantially editing any file under website\u002Fcontent\u002Fblog\u002F. Layered pass — voice rules win on conflict, then a generic AI-tell sweep, then a final \"what makes this obviously AI?\" self-audit.\n---\n\n# Humanize: H·AI·K·U voice + AI-tell removal\n\nThis skill runs a **layered editing pass** on blog prose. It is scoped to `website\u002Fcontent\u002Fblog\u002F**\u002F*.{md,mdx}`. Do not invoke it on UI copy, READMEs, doc pages, or paper revisions — those have a different register.\n\n## Order of operations\n\nThe passes run in this order. Earlier passes win when they conflict with later ones.\n\n1. **Load the voice rules.** Read `.claude\u002Frules\u002Fcontent-voice.md` and `.claude\u002Frules\u002Fcitations.md` before touching prose. These define the register and override anything below.\n2. **Voice-first edit.** Bring the draft into compliance with `content-voice.md`: contractions, no hedging, hook in the first two sentences, prose over bullet lists, every claim grounded per the No-Empty-Authority rule, citations linked or paths named per `citations.md`.\n3. **AI-tell sweep.** Apply the pattern list in this file — but only where it doesn't fight the voice rules. The conflicts are listed explicitly below; respect them.\n4. **Self-audit.** Run the final \"what makes this obviously AI?\" pass. Answer briefly. Then revise.\n\n## Where voice beats the generic humanizer\n\nThe standard AI-tell rules are useful but blunt. The voice is specific, and these overrides apply:\n\n- **Em-dashes are allowed** — particularly in long-form prose where rhythm matters. Don't blanket-replace them with commas. Ban only the *AI-cadence* em-dash that introduces a marketing flourish (\"a seamless experience — designed for you\").\n- **The rule of three is allowed when earned.** The voice actively uses three-beat constructions. Strip them only when they're hollow (\"efficient, scalable, and powerful\"), not when each beat carries weight.\n- **First-person plural OR singular, not generic.** Either \"we\" (project perspective) or \"I\" (Jason's specific anecdote). Pick one per post and stay in it. The generic humanizer's \"use I when it fits\" loses to the explicit register pick.\n- **Cap \"X isn't Y. It's Z.\" at two uses per article**, per `content-voice.md`. The humanizer's \"negative parallelism\" rule aligns here, but the cap is stricter.\n- **Coined phrases stay.** When the draft names a pattern (\"the continuity contract\", \"the workshop has two editors\"), don't flatten it back to plain language. Coining is a feature, not a tell.\n- **Contractions are required**, not optional.\n- **No hypothetical numbers, even framed as such.** Replace with a real, named, verifiable thing — a commit SHA, a file path, a test case, a CI run — or cut the claim.\n- **No emojis. Straight quotes. Sentence-case headings.**\n\nWhen in doubt, voice wins.\n\n## AI-tell pattern list\n\nApply these only where they don't conflict with the rules above.\n\n### Content patterns\n\n- **Inflation of significance.** Strip \"stands as a testament to\", \"marks a pivotal moment\", \"underscores\", \"reflects broader shifts\", \"evolving landscape\", \"key turning point\". Replace with the specific fact that earned the importance — or cut.\n- **Notability theater.** \"His work has been featured in major publications\" is empty. Replace with a named outlet, year, and what was actually said.\n- **`-ing` participle filler.** \"Creating a calming experience and reinforcing simplicity\" — fake depth. Either name what actually happens or drop the participle.\n- **Promotional vocabulary.** \"Vibrant\", \"rich\", \"breathtaking\", \"renowned\", \"nestled\", \"showcasing\", \"powerful\", \"seamless\", \"intuitive\", \"unlock potential\", \"robust\", \"comprehensive\", \"holistic\", \"synergy\". These are marketing tics. Replace with the concrete behavior.\n- **Vague attributions.** \"Experts argue\", \"some critics\", \"industry observers\". Either link the source per `citations.md` or cut the claim.\n- **Generic \"challenges and future prospects\" sections.** If a section is filler, cut it. The voice guide bans formulaic timelines and consultant-shaped outlook paragraphs.\n\n### Language and grammar patterns\n\n- **Overused AI vocabulary.** \"Additionally\", \"crucial\", \"plays a key role\", \"delve into\", \"in today's fast-paced\", \"unlock\", \"navigate the complexities\", \"harness the power of\", \"leverage\". Cut or replace with plain words.\n- **Copula avoidance.** \"Serves as\", \"functions as\", \"represents\". Use \"is\" when \"is\" is what you mean.\n- **Elegant variation.** Repeated reference to the same thing using different words (\"the validator… the gate… the check…\") reads as AI. Pick one term and stick with it.\n- **False ranges.** \"From small startups to large enterprises\", \"everything from X to Y\". State who actually uses it.\n- **Filler phrases.** \"In order to\" → \"to\". \"Has the ability to\" → \"can\". \"Due to the fact that\" → \"because\".\n\n### Style patterns\n\n- **Boldface inside paragraphs** to highlight a list of nouns. Don't do it. Bold is for true emphasis.\n- **Inline-header lists** (\"Speed: faster load times. Security: better encryption.\") are AI shape. Convert to prose if the items belong together.\n- **Title Case In Headings.** Use sentence case.\n- **Curly quotes.** Use straight quotes.\n- **Emojis.** Remove unless explicitly requested.\n\n### Communication patterns\n\n- **Chatbot artifacts.** \"Here's a breakdown of...\", \"Let me know if you want more detail!\", \"I hope this helps.\" Cut.\n- **Knowledge-cutoff hedges.** \"While details are limited, the feature appears to have been introduced recently.\" Find the actual commit or cut.\n- **Sycophantic openers.** \"Great point\", \"this is an insightful observation\". Cut.\n\n### Filler and hedging\n\n- **Excessive hedging.** \"Might potentially\", \"may sometimes\", \"could possibly\". Pick one or none.\n- **Generic conclusions.** \"Overall, the outlook is positive and the future looks promising.\" Replace with a specific forward action, a callback to the opening scene, a question, or a challenge — per the voice guide's \"Vary the closing move\" rule.\n\n## Soul check\n\nPattern removal is half the job. The post also needs a pulse. Per `content-voice.md`:\n\n- Have an opinion. React to facts, don't just report them.\n- Vary rhythm. Short punchy sentence. Then one that takes its time and earns the breath.\n- Acknowledge complexity where it exists. \"It works, but the seams show.\"\n- Be specific about feelings and reactions. Not \"this is concerning\" but the concrete thing that makes it concerning.\n- Coin the phrase the reader will repeat in their next meeting.\n\nIf the prose reads like a Wikipedia summary or a press release, it failed the soul check even if every AI tell is gone.\n\n## The final self-audit\n\nAfter the layered passes, perform this final step explicitly. Don't skip it.\n\n1. Read the revised draft.\n2. Ask: **\"What makes this obviously AI generated?\"** Answer in three to five short bullets, naming specific phrases or sentences. Be honest — the goal is to find what's left, not to defend the draft.\n3. Revise those specific tells.\n4. Re-read once more. If the draft would now plausibly come from a human writing about a project they shipped, it's done. If not, repeat step 2.\n\n## Output\n\nWhen invoked on a draft, produce:\n\n1. The revised markdown (full file or the diff, depending on scope).\n2. The self-audit answer — the bullets from \"what makes this obviously AI generated?\" before the final revision.\n3. A short note on any deliberate voice choices that *look* like AI tells but aren't (a kept em-dash, a kept three-beat construction). This protects the choice from being undone on the next pass.\n\n## Scope\n\nThis skill is paired with `.claude\u002Frules\u002Fblog-humanize.md`, a path-scoped rule that loads only when working on `website\u002Fcontent\u002Fblog\u002F**`. The rule tells the assistant to invoke this skill before reporting the task complete. Do not invoke this skill on other paths.\n","# Humanize: H·AI·K·U voice + AI-tell removal\n\nThis skill runs a **layered editing pass** on blog prose. It is scoped to `website\u002Fcontent\u002Fblog\u002F**\u002F*.{md,mdx}`. Do not invoke it on UI copy, READMEs, doc pages, or paper revisions — those have a different register.\n\n## Order of operations\n\nThe passes run in this order. Earlier passes win when they conflict with later ones.\n\n1. **Load the voice rules.** Read `.claude\u002Frules\u002Fcontent-voice.md` and `.claude\u002Frules\u002Fcitations.md` before touching prose. These define the register and override anything below.\n2. **Voice-first edit.** Bring the draft into compliance with `content-voice.md`: contractions, no hedging, hook in the first two sentences, prose over bullet lists, every claim grounded per the No-Empty-Authority rule, citations linked or paths named per `citations.md`.\n3. **AI-tell sweep.** Apply the pattern list in this file — but only where it doesn't fight the voice rules. The conflicts are listed explicitly below; respect them.\n4. **Self-audit.** Run the final \"what makes this obviously AI?\" pass. Answer briefly. Then revise.\n\n## Where voice beats the generic humanizer\n\nThe standard AI-tell rules are useful but blunt. The voice is specific, and these overrides apply:\n\n- **Em-dashes are allowed** — particularly in long-form prose where rhythm matters. Don't blanket-replace them with commas. Ban only the *AI-cadence* em-dash that introduces a marketing flourish (\"a seamless experience — designed for you\").\n- **The rule of three is allowed when earned.** The voice actively uses three-beat constructions. Strip them only when they're hollow (\"efficient, scalable, and powerful\"), not when each beat carries weight.\n- **First-person plural OR singular, not generic.** Either \"we\" (project perspective) or \"I\" (Jason's specific anecdote). Pick one per post and stay in it. The generic humanizer's \"use I when it fits\" loses to the explicit register pick.\n- **Cap \"X isn't Y. It's Z.\" at two uses per article**, per `content-voice.md`. The humanizer's \"negative parallelism\" rule aligns here, but the cap is stricter.\n- **Coined phrases stay.** When the draft names a pattern (\"the continuity contract\", \"the workshop has two editors\"), don't flatten it back to plain language. Coining is a feature, not a tell.\n- **Contractions are required**, not optional.\n- **No hypothetical numbers, even framed as such.** Replace with a real, named, verifiable thing — a commit SHA, a file path, a test case, a CI run — or cut the claim.\n- **No emojis. Straight quotes. Sentence-case headings.**\n\nWhen in doubt, voice wins.\n\n## AI-tell pattern list\n\nApply these only where they don't conflict with the rules above.\n\n### Content patterns\n\n- **Inflation of significance.** Strip \"stands as a testament to\", \"marks a pivotal moment\", \"underscores\", \"reflects broader shifts\", \"evolving landscape\", \"key turning point\". Replace with the specific fact that earned the importance — or cut.\n- **Notability theater.** \"His work has been featured in major publications\" is empty. Replace with a named outlet, year, and what was actually said.\n- **`-ing` participle filler.** \"Creating a calming experience and reinforcing simplicity\" — fake depth. Either name what actually happens or drop the participle.\n- **Promotional vocabulary.** \"Vibrant\", \"rich\", \"breathtaking\", \"renowned\", \"nestled\", \"showcasing\", \"powerful\", \"seamless\", \"intuitive\", \"unlock potential\", \"robust\", \"comprehensive\", \"holistic\", \"synergy\". These are marketing tics. Replace with the concrete behavior.\n- **Vague attributions.** \"Experts argue\", \"some critics\", \"industry observers\". Either link the source per `citations.md` or cut the claim.\n- **Generic \"challenges and future prospects\" sections.** If a section is filler, cut it. The voice guide bans formulaic timelines and consultant-shaped outlook paragraphs.\n\n### Language and grammar patterns\n\n- **Overused AI vocabulary.** \"Additionally\", \"crucial\", \"plays a key role\", \"delve into\", \"in today's fast-paced\", \"unlock\", \"navigate the complexities\", \"harness the power of\", \"leverage\". Cut or replace with plain words.\n- **Copula avoidance.** \"Serves as\", \"functions as\", \"represents\". Use \"is\" when \"is\" is what you mean.\n- **Elegant variation.** Repeated reference to the same thing using different words (\"the validator… the gate… the check…\") reads as AI. Pick one term and stick with it.\n- **False ranges.** \"From small startups to large enterprises\", \"everything from X to Y\". State who actually uses it.\n- **Filler phrases.** \"In order to\" → \"to\". \"Has the ability to\" → \"can\". \"Due to the fact that\" → \"because\".\n\n### Style patterns\n\n- **Boldface inside paragraphs** to highlight a list of nouns. Don't do it. Bold is for true emphasis.\n- **Inline-header lists** (\"Speed: faster load times. Security: better encryption.\") are AI shape. Convert to prose if the items belong together.\n- **Title Case In Headings.** Use sentence case.\n- **Curly quotes.** Use straight quotes.\n- **Emojis.** Remove unless explicitly requested.\n\n### Communication patterns\n\n- **Chatbot artifacts.** \"Here's a breakdown of...\", \"Let me know if you want more detail!\", \"I hope this helps.\" Cut.\n- **Knowledge-cutoff hedges.** \"While details are limited, the feature appears to have been introduced recently.\" Find the actual commit or cut.\n- **Sycophantic openers.** \"Great point\", \"this is an insightful observation\". Cut.\n\n### Filler and hedging\n\n- **Excessive hedging.** \"Might potentially\", \"may sometimes\", \"could possibly\". Pick one or none.\n- **Generic conclusions.** \"Overall, the outlook is positive and the future looks promising.\" Replace with a specific forward action, a callback to the opening scene, a question, or a challenge — per the voice guide's \"Vary the closing move\" rule.\n\n## Soul check\n\nPattern removal is half the job. The post also needs a pulse. Per `content-voice.md`:\n\n- Have an opinion. React to facts, don't just report them.\n- Vary rhythm. Short punchy sentence. Then one that takes its time and earns the breath.\n- Acknowledge complexity where it exists. \"It works, but the seams show.\"\n- Be specific about feelings and reactions. Not \"this is concerning\" but the concrete thing that makes it concerning.\n- Coin the phrase the reader will repeat in their next meeting.\n\nIf the prose reads like a Wikipedia summary or a press release, it failed the soul check even if every AI tell is gone.\n\n## The final self-audit\n\nAfter the layered passes, perform this final step explicitly. Don't skip it.\n\n1. Read the revised draft.\n2. Ask: **\"What makes this obviously AI generated?\"** Answer in three to five short bullets, naming specific phrases or sentences. Be honest — the goal is to find what's left, not to defend the draft.\n3. Revise those specific tells.\n4. Re-read once more. If the draft would now plausibly come from a human writing about a project they shipped, it's done. If not, repeat step 2.\n\n## Output\n\nWhen invoked on a draft, produce:\n\n1. The revised markdown (full file or the diff, depending on scope).\n2. The self-audit answer — the bullets from \"what makes this obviously AI generated?\" before the final revision.\n3. A short note on any deliberate voice choices that *look* like AI tells but aren't (a kept em-dash, a kept three-beat construction). This protects the choice from being undone on the next pass.\n\n## Scope\n\nThis skill is paired with `.claude\u002Frules\u002Fblog-humanize.md`, a path-scoped rule that loads only when working on `website\u002Fcontent\u002Fblog\u002F**`. The rule tells the assistant to invoke this skill before reporting the task complete. Do not invoke this skill on other paths.\n",{"name":5,"description":7},"main","42451f546de75e0c4f38d5d3bb0999e54661bf32",".claude\u002Fskills\u002Fhumanize",0,"https:\u002F\u002Fgithub.com\u002FTheBushidoCollective\u002Fhaiku-method","https:\u002F\u002Fhaikumethod.ai\u002F","https:\u002F\u002Fgithub.com\u002FTheBushidoCollective","https:\u002F\u002Fthebushido.co","Strategic Technology Leadership That Actually Delivers","2026-11-04 00:54:41",{"source":8,"skillName":5,"commands":56,"prompt":66,"zipAvailable":67},[57,60,63],{"tool":58,"command":59},"npx","npx skills add TheBushidoCollective\u002Fhaiku-method --skill humanize",{"tool":61,"command":62},"bunx","bunx skills add TheBushidoCollective\u002Fhaiku-method --skill humanize",{"tool":64,"command":65},"pnpm","pnpm dlx skills add TheBushidoCollective\u002Fhaiku-method --skill humanize","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FTheBushidoCollective\u002Fhaiku-method\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fhumanize\u002FSKILL.md",true,[],[70,91,114,124,140,154],{"skillNo":71,"name":72,"nameZh":6,"description":73,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":74,"categories":75,"keywords":76,"publishedAt":40},"01M6P93YBHRPSHMRR5C870HPF8","haiku-zap","Zero-ceremony single-unit execution — run one task directly through a stage's hat loop without intent or workflow scaffolding","https:\u002F\u002Fgithub.com\u002FTheBushidoCollective\u002Fhaiku-method\u002Fblob\u002Fmain\u002Fplugin\u002Fskills\u002Fhaiku-zap\u002FSKILL.md",[],[77,78,79,80,81,82,83,84,85,86,87,88,89,90],"zap","haiku zap","haiku","single","unit","stage","scaffolding","task","workflow","run","ZeRO","execution","intent","loop",{"skillNo":92,"name":93,"nameZh":6,"description":94,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":95,"categories":96,"keywords":102,"publishedAt":40},"01M6P93D0VA0HWH0NZHER4C6YN","haiku-version","Show the running H·AI·K·U MCP binary version, plugin version, and whether the engine is the dev source build or the compiled bundle","https:\u002F\u002Fgithub.com\u002FTheBushidoCollective\u002Fhaiku-method\u002Fblob\u002Fmain\u002Fplugin\u002Fskills\u002Fhaiku-version\u002FSKILL.md",[97],{"categoryNo":98,"categoryName":99,"taxonomyType":100,"taxonomyTypeDesc":101},"01JBSEEDCAT000000000000023","Skill与智能体开发","SCENARIO","使用场景",[79,103,104,105,106,107,108,109,110,111,112,113],"version","haiku version","show","MCP","plugin","dev","engine","build","binary","bundle","source",{"skillNo":115,"name":116,"nameZh":6,"description":117,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":118,"categories":119,"keywords":120,"publishedAt":40},"01M6P93947PNH55ZJ999GZ9J3C","haiku-unarchive","Unarchive an intent — restore a previously archived intent to default lists","https:\u002F\u002Fgithub.com\u002FTheBushidoCollective\u002Fhaiku-method\u002Fblob\u002Fmain\u002Fplugin\u002Fskills\u002Fhaiku-unarchive\u002FSKILL.md",[],[121,79,89,122,123],"haiku unarchive","restore","default",{"skillNo":125,"name":126,"nameZh":6,"description":127,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":128,"categories":129,"keywords":130,"publishedAt":40},"01M6P936W0VF4BE1WR6JGQ6NJW","haiku-triggers","Poll configured providers for events that should create intents or advance gates","https:\u002F\u002Fgithub.com\u002FTheBushidoCollective\u002Fhaiku-method\u002Fblob\u002Fmain\u002Fplugin\u002Fskills\u002Fhaiku-triggers\u002FSKILL.md",[],[131,132,79,133,134,135,136,137,138,139],"triggers","haiku triggers","events","create","Providers","intents","advance","gates","for",{"skillNo":141,"name":142,"nameZh":6,"description":143,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":144,"categories":145,"keywords":146,"publishedAt":40},"01M6P92V5HDE325PHZHK2BVR7S","haiku-studios","Browse available H·AI·K·U studios grouped by category with help links into each studio definition","https:\u002F\u002Fgithub.com\u002FTheBushidoCollective\u002Fhaiku-method\u002Fblob\u002Fmain\u002Fplugin\u002Fskills\u002Fhaiku-studios\u002FSKILL.md",[],[79,147,148,149,150,151,152,153],"haiku studios","browse","help","with","links","studio","definition",{"skillNo":155,"name":156,"nameZh":6,"description":157,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":158,"categories":159,"keywords":160,"publishedAt":40},"01M6P92KY8511PWQP83YKVFJB3","haiku-statusline","Install, remove, or preview the H·AI·K·U custom Claude Code status line — a one-line stage\u002Fphase indicator for the active intent","https:\u002F\u002Fgithub.com\u002FTheBushidoCollective\u002Fhaiku-method\u002Fblob\u002Fmain\u002Fplugin\u002Fskills\u002Fhaiku-statusline\u002FSKILL.md",[],[79,161,162,163,164,165,166,89,167,168,82,169,139,170,171,172],"statusline","haiku statusline","code","indicator","install","remove","preview","active","line","custom","phase","status"]