[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M78FPKEYQHMMHR8EHDDB81NA":3,"skill-same:01M78FPKEYQHMMHR8EHDDB81NA":45},{"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":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":24,"commitSha":25,"skillDirectory":26,"forksCount":27,"repositoryHtmlUrl":28,"repositoryHomepageUrl":6,"ownerHtmlUrl":29,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":30,"install":31,"resources":44},"01M78FPKEYQHMMHR8EHDDB81NA","demonstrate",null,"Use at a notable increment with a taste-surface, before a human Go\u002FNo-Go — the kit's own increment-demonstration craft (a kit-original; superpowers has no demonstrate craft). Proactively surface a working, observable increment for the human's taste\u002Fjudgment, frame the specific judgment call, and record the verdict in a UAT sign-off. Distinct from review (agent→agent correctness) and verification (agent→self evidence).","SeaBrad72\u002Fsparkwright","SeaBrad72","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F189294229?v=4","01M78FP39B4DE6NKHX1AJKSPFX","Bradley James","01M78FP3A2X3TW6T6GHBPB7PT3",3,"Apache-2.0","https:\u002F\u002Fgithub.com\u002FSeaBrad72\u002Fsparkwright\u002Fblob\u002Fmain\u002Fskills\u002Fdemonstrate\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: demonstrate\ndescription: Use at a notable increment with a taste-surface, before a human Go\u002FNo-Go — the kit's own increment-demonstration craft (a kit-original; superpowers has no demonstrate craft). Proactively surface a working, observable increment for the human's taste\u002Fjudgment, frame the specific judgment call, and record the verdict in a UAT sign-off. Distinct from review (agent→agent correctness) and verification (agent→self evidence).\n---\n\n# Demonstrate — surface a working increment for human taste\u002Fjudgment (the informed half of Go\u002FNo-Go)\n\nThe kit's own increment-demonstration craft: at a notable increment (slice \u002F sub-slice \u002F epic) that carries a **taste-surface** — a place where *correct ≠ good* — proactively produce a working, observable thing for the human to *react to*, frame the exact judgment call, and record the verdict. This is the **\"informed\" half** of the promotion Go\u002FNo-Go: the human cannot render a good GO on UX\u002Fflow\u002Fworking-functionality they never saw. An existing seat (Engineer \u002F Reviewer \u002F Orchestrator) invokes this craft — it is a **skill, not a new seat** (agents-vs-skills: the kit adds a discipline, not a standing role).\n\n**Distinct discipline** (why it earns a skill slot): different *audience* and *purpose* than its neighbours. `review` is agent→agent, judging **correctness**. `verification` is agent→self, producing **evidence** before a done-claim. `demonstrate` is **agent→human**, surfacing **taste\u002FUX judgment** — the class of question a green test can never answer (\"does this flow *feel* right?\").\n\n\u003C!-- The frontmatter and the discipline phrases below are conformance-load-bearing:\n     conformance\u002Forchestrator-loop-wired.sh greps this file for these exact kit-distinctive\n     markers (each quoted — preserve them verbatim, none may contain a pipe char or a TAB):\n       \"name: demonstrate\"  \"## When to use\"  \"taste-surface\"\n       \"demonstrable increment\"  \"frame the specific judgment call\"\n       \"visually-observable-first\"  \"verdict recorded in a UAT sign-off\"\n     Edits that drop or rename any of them can turn the skill-spine lock RED. -->\n\n## When to use\nAt **each notable increment** (a slice, a sub-slice, or an epic) that has a **taste-surface** — UI, UX, a user flow, working functionality, a rendered data table: anything where *correct ≠ good* — **before the human Go\u002FNo-Go**. If the increment is pure plumbing with no taste-surface (a refactor behind a stable interface, a conformance-lock edit), there is nothing to demonstrate — skip it honestly. The trigger is the taste-surface, not the calendar.\n\n## The craft — the flow\n1. **Detect the taste-surface.** Scan the increment for the place where *correct ≠ good*: a screen, a flow, a table, a piece of working functionality a person would form an opinion about. No taste-surface → nothing to demonstrate.\n2. **Produce the observable — a thing to react to, not a description.** Spin up an instance, capture a screenshot, render the table, run the flow end to end. The output is a **demonstrable increment** the human can *look at and judge*, never a paragraph claiming it works. A description is what `verification` produces for correctness; `demonstrate` produces the artifact itself.\n3. **Frame the specific judgment call.** Ask the real question — \"does this checkout flow feel right?\", \"is this table scannable?\", \"would you ship this empty state?\" — never \"LGTM?\". A blind nod is the K16 hole wearing a UX hat; **frame the specific judgment call** so the human engages the actual taste question, not a rubber stamp.\n4. **Record the verdict.** The **verdict recorded in a UAT sign-off** — the existing `templates\u002FUAT-SIGNOFF-TEMPLATE.md` (accept \u002F reject + the demonstrated evidence + signer + date). Reuse it; do **not** invent a parallel artifact.\n5. **Ordering rule — visually-observable-first.** When task sequencing is otherwise indifferent, build the **visually-observable-first**: put the thing a human can react to earliest, so there is something to engage with while the rest is still forming. Early observability turns a late surprise into an early course-correction.\n\n## Autonomy coupling (KW20) — density scales with human distance\nAutonomy modulates *demonstrate density*. When you **fan out N agents** with the human out of the loop, a demonstrable increment is the human's **only** window onto taste — so demonstrable increments are **mandatory and frequent**; the observable substitutes for the absent human glance. On a **small, human-proximate build** where the human is already watching the work, the touch is **lighter** — the human's live attention already covers much of what a formal demonstration would surface. Same rigor, different composition: the further the human is from the keyboard, the more the taste-check must be paid in demonstrated artifacts rather than over-the-shoulder glances.\n\n## Honest ceiling\nWhether the human **actually looks** is un-gateable — no CI check proves a person engaged their taste. The provable **FLOOR** is *increment demonstrated + judgment call framed + verdict recorded in a UAT sign-off*; that is the harness-neutral core this skill guarantees. A NATIVE harness that auto-launches the instance or embeds the screenshot is a **bonus**, never the FLOOR. This slice ships the discipline **advisory**: the promotion-readiness surfacing raises the obligation at every RC promotion, and `build` \u002F `review` reference it — but a *hard* \"taste-surface touched + no record → fail CI\" gate (a bespoke taste-surface detector) is deliberately **not** built here; it rides the KW7 generic obligation-trigger mechanism as a co-follow, so the trigger is designed once, not twice.\n\n## Rationalizations to refuse\n| Rationalization | Why it fails |\n|---|---|\n| \"The tests are green, so it's good.\" | Green proves *correct*, not *good*. A taste-surface is exactly where *correct ≠ good* — demonstrate it and let a human judge. |\n| \"I'll describe what it does instead of showing it.\" | Produce the **demonstrable increment** — a thing to react to. A description is `verification` evidence; taste needs the artifact itself. |\n| \"I'll just ask 'LGTM?'.\" | **Frame the specific judgment call** — \"does this flow feel right?\", not a blind nod. A rubber stamp is the K16 hole in a UX hat. |\n| \"The reviewer already approved it.\" | `review` is agent→agent correctness; it does not surface taste to the human. Different audience, different purpose — both are needed. |\n| \"I'll note it in the PR body; no need for a sign-off.\" | The **verdict recorded in a UAT sign-off** is the auditable record — reuse `templates\u002FUAT-SIGNOFF-TEMPLATE.md`; PR prose is not the record. |\n| \"We're fanning out ten agents; I'll demo at the very end.\" | Autonomy raises demonstrate density — fan-out means demonstrable increments are **mandatory and frequent**, the human's only window. Demo per increment, not once at the end. |\n| \"I'll build the observable thing last, after the plumbing.\" | **visually-observable-first** when ordering is indifferent — early observability turns a late surprise into an early course-correction. |\n\n## Terminal state\nFor each notable increment carrying a taste-surface: the taste-surface was detected, a **demonstrable increment** was produced (an observable thing to react to, not a description), the **specific judgment call** was framed for the human, and the **verdict recorded in a UAT sign-off** (`templates\u002FUAT-SIGNOFF-TEMPLATE.md`) — with the **visually-observable-first** ordering honoured where sequencing was indifferent, and demonstrate density scaled to the autonomy level. The human's Go\u002FNo-Go is now *informed* by taste, not just by correctness. Honest ceiling: the discipline is *provided + surfaced + roster-protected*; whether the human actually looked is un-gateable.\n","# Demonstrate — surface a working increment for human taste\u002Fjudgment (the informed half of Go\u002FNo-Go)\n\nThe kit's own increment-demonstration craft: at a notable increment (slice \u002F sub-slice \u002F epic) that carries a **taste-surface** — a place where *correct ≠ good* — proactively produce a working, observable thing for the human to *react to*, frame the exact judgment call, and record the verdict. This is the **\"informed\" half** of the promotion Go\u002FNo-Go: the human cannot render a good GO on UX\u002Fflow\u002Fworking-functionality they never saw. An existing seat (Engineer \u002F Reviewer \u002F Orchestrator) invokes this craft — it is a **skill, not a new seat** (agents-vs-skills: the kit adds a discipline, not a standing role).\n\n**Distinct discipline** (why it earns a skill slot): different *audience* and *purpose* than its neighbours. `review` is agent→agent, judging **correctness**. `verification` is agent→self, producing **evidence** before a done-claim. `demonstrate` is **agent→human**, surfacing **taste\u002FUX judgment** — the class of question a green test can never answer (\"does this flow *feel* right?\").\n\n\u003C!-- The frontmatter and the discipline phrases below are conformance-load-bearing:\n     conformance\u002Forchestrator-loop-wired.sh greps this file for these exact kit-distinctive\n     markers (each quoted — preserve them verbatim, none may contain a pipe char or a TAB):\n       \"name: demonstrate\"  \"## When to use\"  \"taste-surface\"\n       \"demonstrable increment\"  \"frame the specific judgment call\"\n       \"visually-observable-first\"  \"verdict recorded in a UAT sign-off\"\n     Edits that drop or rename any of them can turn the skill-spine lock RED. -->\n\n## When to use\nAt **each notable increment** (a slice, a sub-slice, or an epic) that has a **taste-surface** — UI, UX, a user flow, working functionality, a rendered data table: anything where *correct ≠ good* — **before the human Go\u002FNo-Go**. If the increment is pure plumbing with no taste-surface (a refactor behind a stable interface, a conformance-lock edit), there is nothing to demonstrate — skip it honestly. The trigger is the taste-surface, not the calendar.\n\n## The craft — the flow\n1. **Detect the taste-surface.** Scan the increment for the place where *correct ≠ good*: a screen, a flow, a table, a piece of working functionality a person would form an opinion about. No taste-surface → nothing to demonstrate.\n2. **Produce the observable — a thing to react to, not a description.** Spin up an instance, capture a screenshot, render the table, run the flow end to end. The output is a **demonstrable increment** the human can *look at and judge*, never a paragraph claiming it works. A description is what `verification` produces for correctness; `demonstrate` produces the artifact itself.\n3. **Frame the specific judgment call.** Ask the real question — \"does this checkout flow feel right?\", \"is this table scannable?\", \"would you ship this empty state?\" — never \"LGTM?\". A blind nod is the K16 hole wearing a UX hat; **frame the specific judgment call** so the human engages the actual taste question, not a rubber stamp.\n4. **Record the verdict.** The **verdict recorded in a UAT sign-off** — the existing `templates\u002FUAT-SIGNOFF-TEMPLATE.md` (accept \u002F reject + the demonstrated evidence + signer + date). Reuse it; do **not** invent a parallel artifact.\n5. **Ordering rule — visually-observable-first.** When task sequencing is otherwise indifferent, build the **visually-observable-first**: put the thing a human can react to earliest, so there is something to engage with while the rest is still forming. Early observability turns a late surprise into an early course-correction.\n\n## Autonomy coupling (KW20) — density scales with human distance\nAutonomy modulates *demonstrate density*. When you **fan out N agents** with the human out of the loop, a demonstrable increment is the human's **only** window onto taste — so demonstrable increments are **mandatory and frequent**; the observable substitutes for the absent human glance. On a **small, human-proximate build** where the human is already watching the work, the touch is **lighter** — the human's live attention already covers much of what a formal demonstration would surface. Same rigor, different composition: the further the human is from the keyboard, the more the taste-check must be paid in demonstrated artifacts rather than over-the-shoulder glances.\n\n## Honest ceiling\nWhether the human **actually looks** is un-gateable — no CI check proves a person engaged their taste. The provable **FLOOR** is *increment demonstrated + judgment call framed + verdict recorded in a UAT sign-off*; that is the harness-neutral core this skill guarantees. A NATIVE harness that auto-launches the instance or embeds the screenshot is a **bonus**, never the FLOOR. This slice ships the discipline **advisory**: the promotion-readiness surfacing raises the obligation at every RC promotion, and `build` \u002F `review` reference it — but a *hard* \"taste-surface touched + no record → fail CI\" gate (a bespoke taste-surface detector) is deliberately **not** built here; it rides the KW7 generic obligation-trigger mechanism as a co-follow, so the trigger is designed once, not twice.\n\n## Rationalizations to refuse\n| Rationalization | Why it fails |\n|---|---|\n| \"The tests are green, so it's good.\" | Green proves *correct*, not *good*. A taste-surface is exactly where *correct ≠ good* — demonstrate it and let a human judge. |\n| \"I'll describe what it does instead of showing it.\" | Produce the **demonstrable increment** — a thing to react to. A description is `verification` evidence; taste needs the artifact itself. |\n| \"I'll just ask 'LGTM?'.\" | **Frame the specific judgment call** — \"does this flow feel right?\", not a blind nod. A rubber stamp is the K16 hole in a UX hat. |\n| \"The reviewer already approved it.\" | `review` is agent→agent correctness; it does not surface taste to the human. Different audience, different purpose — both are needed. |\n| \"I'll note it in the PR body; no need for a sign-off.\" | The **verdict recorded in a UAT sign-off** is the auditable record — reuse `templates\u002FUAT-SIGNOFF-TEMPLATE.md`; PR prose is not the record. |\n| \"We're fanning out ten agents; I'll demo at the very end.\" | Autonomy raises demonstrate density — fan-out means demonstrable increments are **mandatory and frequent**, the human's only window. Demo per increment, not once at the end. |\n| \"I'll build the observable thing last, after the plumbing.\" | **visually-observable-first** when ordering is indifferent — early observability turns a late surprise into an early course-correction. |\n\n## Terminal state\nFor each notable increment carrying a taste-surface: the taste-surface was detected, a **demonstrable increment** was produced (an observable thing to react to, not a description), the **specific judgment call** was framed for the human, and the **verdict recorded in a UAT sign-off** (`templates\u002FUAT-SIGNOFF-TEMPLATE.md`) — with the **visually-observable-first** ordering honoured where sequencing was indifferent, and demonstrate density scaled to the autonomy level. The human's Go\u002FNo-Go is now *informed* by taste, not just by correctness. Honest ceiling: the discipline is *provided + surfaced + roster-protected*; whether the human actually looked is un-gateable.\n",{"name":5,"description":7},"main","46ed464a0bfe79d18251e771dd0fc9e1e89fe43f","skills\u002Fdemonstrate",0,"https:\u002F\u002Fgithub.com\u002FSeaBrad72\u002Fsparkwright","https:\u002F\u002Fgithub.com\u002FSeaBrad72","2026-11-15 04:00:12",{"source":8,"skillName":5,"commands":32,"prompt":42,"zipAvailable":43},[33,36,39],{"tool":34,"command":35},"npx","npx skills add SeaBrad72\u002Fsparkwright --skill demonstrate",{"tool":37,"command":38},"bunx","bunx skills add SeaBrad72\u002Fsparkwright --skill demonstrate",{"tool":40,"command":41},"pnpm","pnpm dlx skills add SeaBrad72\u002Fsparkwright --skill demonstrate","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FSeaBrad72\u002Fsparkwright\u002Fblob\u002Fmain\u002Fskills\u002Fdemonstrate\u002FSKILL.md",true,[],[46,53,60,66,73,80],{"skillNo":47,"name":48,"nameZh":6,"description":49,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":50,"categories":51,"keywords":52,"publishedAt":19},"01M7DZCWVTK18C484JQPMN4NAZ","inception-from-code","Use when adopting the kit into an EXISTING repository (brownfield), to draft the Inception artifacts from the code instead of from blank templates. The agent reads the repository and writes five drafts, only under docs\u002Finception-draft\u002F, each tagged by how each fact is known; a person reviews them and moves what is right into place. Agents propose, humans ratify.","https:\u002F\u002Fgithub.com\u002FSeaBrad72\u002Fsparkwright\u002Fblob\u002Fmain\u002Fskills\u002Finception-from-code\u002FSKILL.md",[],[],{"skillNo":54,"name":55,"nameZh":6,"description":56,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":57,"categories":58,"keywords":59,"publishedAt":19},"01M78FQSZRGQH2D4RCKCH9594N","worktrees","Use BEFORE fanning out parallel Engineers or starting isolated feature work — the kit's own isolation skill (replaces, does not depend on, superpowers using-git-worktrees). The isolation craft, wired to the Orchestrator seat — detect-existing-first, native tools first, the kit's disjoint-set parallel-safety rule, and conflict-safe integration.","https:\u002F\u002Fgithub.com\u002FSeaBrad72\u002Fsparkwright\u002Fblob\u002Fmain\u002Fskills\u002Fworktrees\u002FSKILL.md",[],[],{"skillNo":61,"name":62,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":63,"categories":64,"keywords":65,"publishedAt":19},"01M78FQQXX2CYGF8V0GVG35TTB","verification","https:\u002F\u002Fgithub.com\u002FSeaBrad72\u002Fsparkwright\u002Fblob\u002Fmain\u002Fskills\u002Fverification\u002FSKILL.md",[],[],{"skillNo":67,"name":68,"nameZh":6,"description":69,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":70,"categories":71,"keywords":72,"publishedAt":19},"01M78FQK7Y0JVS2V77S8SEBD29","using-skills","Use at the START of any task, before any response or action — the kit's own discovery keystone (replaces, does not depend on, superpowers using-superpowers). How the kit's skills are found and invoked (invoke by reading), the discipline of checking for a relevant skill before acting, instruction priority, and the index of the kit's own spine skills.","https:\u002F\u002Fgithub.com\u002FSeaBrad72\u002Fsparkwright\u002Fblob\u002Fmain\u002Fskills\u002Fusing-skills\u002FSKILL.md",[],[],{"skillNo":74,"name":75,"nameZh":6,"description":76,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":77,"categories":78,"keywords":79,"publishedAt":19},"01M78FQEA5ATPZDYN780F4VGQD","tdd","Use when implementing any feature, bug fix, refactor, or behaviour change — BEFORE writing production code. The kit's own test-driven-development skill (replaces, does not depend on, superpowers test-driven-development). Red-green-refactor framed as the kit's non-vacuity law, plus coverage, testing-pyramid, and AI-eval disciplines.","https:\u002F\u002Fgithub.com\u002FSeaBrad72\u002Fsparkwright\u002Fblob\u002Fmain\u002Fskills\u002Ftdd\u002FSKILL.md",[],[],{"skillNo":81,"name":82,"nameZh":6,"description":6,"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":85,"publishedAt":19},"01M78FQADNX9RZQ63C7WG7HPZH","review","https:\u002F\u002Fgithub.com\u002FSeaBrad72\u002Fsparkwright\u002Fblob\u002Fmain\u002Fskills\u002Freview\u002FSKILL.md",[],[]]