[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M78FP9QDFCRM62WZP16P316N":3,"skill-same:01M78FP9QDFCRM62WZP16P316N":44},{"skillNo":4,"name":5,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":11,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":15,"categories":16,"keywords":17,"publishedAt":18,"linkOnly":19,"markdownContent":20,"markdownBody":21,"frontMatter":22,"refName":23,"commitSha":24,"skillDirectory":25,"forksCount":26,"repositoryHtmlUrl":27,"repositoryHomepageUrl":6,"ownerHtmlUrl":28,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":29,"install":30,"resources":43},"01M78FP9QDFCRM62WZP16P316N","continuous-discovery",null,"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\u002Fcontinuous-discovery\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: continuous-discovery\ndescription: Use BEFORE design shapes a solution — at the front of the loop, when a problem or outcome is assumed rather than validated, or a slice carries an untested assumption. The kit's own problem-space product-discovery craft (a kit-original; superpowers has no equivalent). The human↔AI discovery partner: structure the opportunity space, surface and test the riskiest assumption, keep work honest to outcomes over output — before any solution is designed.\n---\n\n# Continuous discovery — interrogate the problem before designing the solution\n\nThe kit's own problem-space discovery craft: take a felt need → a *validated* problem, a measurable outcome, and the riskiest assumption tested — before `design` shapes any solution. This is the **front of the loop** (Discover), the partner to `skills\u002Fdesign\u002FSKILL.md`'s solution-space. Continuous discovery is Teresa Torres' practice made workable for a human↔AI pair; it is a kit-original (superpowers has no discovery skill — its `brainstorming` is solution-space, the kit's `design`).\n\n> **Not to be confused with the discovery *keystone*.** `skills\u002Fusing-skills\u002FSKILL.md` is *skill*-discovery (\"which skill applies?\"). This skill is *product* discovery (\"which problem, and is it the right one?\"). Different axis entirely.\n\n\u003C!-- The frontmatter and the discipline markers below are conformance-load-bearing:\n     conformance\u002Forchestrator-loop-wired.sh greps this file for kit-distinctive markers\n     (name: continuous-discovery, discovery partner, outcome over output,\n     opportunity solution tree, riskiest assumption, small bet).\n     Edits that drop or rename them can turn the skill-spine lock RED. -->\n\n## When to use\nAt the **front of the loop**, before `design`: when the problem is taken for granted, when a slice's value rests on an assumption no one has tested, or when \"what we'll build\" is clear but \"what outcome it changes, and for whom\" is not. Every non-trivial slice deserves a discovery pass — a small one for a well-understood problem, a deliberate one when the problem is fuzzy or the bet is large.\n\n## The discovery partner — not the decider\nThe **discovery partner** is the kit's framing of the human↔AI relationship here, and it is the load-bearing constraint of this whole skill: **the human is the PO.** The agent does *not* decide what to build or declare a problem worth solving. The agent makes continuous discovery *accessible and honest*: it structures the space, asks the sharp question, names the assumption everyone is gliding past, and designs the cheap test — then hands the judgment back to the human. A partner that quietly starts deciding has stopped being a partner; that is the failure mode this skill exists to prevent. There is no Product *seat* for the same reason — the human owns the decision; the Orchestrator merely *wears the Product hat* to invoke this craft.\n\n## The flow (the proven spine)\n1. **Anchor on the outcome, not the feature — `outcome over output`.** Before anything, ask what customer or business *outcome* this work would change, and how we'd measure it. Frame the work by the outcome it moves, never by the feature shipped. A slice with no outcome hypothesis is output theatre — surface that out loud. (This is the Definition-of-Ready \"success metric \u002F hypothesis\" item, made a craft.)\n2. **Map the `opportunity solution tree`.** Build the tree *with* the human: the outcome at the root → the **opportunities** beneath it (the problems, needs, and desires that, if addressed, move the outcome) → candidate solution ideas → the experiments that would test them. The point is to make the chosen opportunity *explicit and compared*, not the first idea that arrived. Breadth before depth.\n3. **Surface and rank the `riskiest assumption`.** Every chosen opportunity and solution bets on assumptions — desirability (do they want it?), viability (does it work for the business?), feasibility (can we build it?), usability (can they use it?). Name them, then rank the **riskiest assumption**: the one that, if false, wastes the most work. That is what gets tested first — not the easiest thing to test.\n4. **Design the `small bet`.** Design the cheapest experiment that would prove or kill the riskiest assumption *before* a slice is built — a `small bet`, not a big-bang study. Continuous, not one-shot: small, frequent tests on a cadence beat a single upfront discovery phase. Define in advance what result would change the decision; a test whose outcome changes nothing is theatre.\n5. **Hand off to `design`.** Discovery's terminal state is a validated problem + a measurable outcome + the riskiest assumption tested — handed to `skills\u002Fdesign\u002FSKILL.md` to shape the solution. This skill never starts a solution; designing is `design`'s job, building is the Engineer's.\n\n## What this craft points at (it does not duplicate the infra)\n- Capture the validated problem + outcome in `templates\u002FFEATURE-REQUEST-TEMPLATE.md` (incl. its success-metric \u002F hypothesis fields).\n- The Definition-of-Ready \"success metric \u002F hypothesis\" item is the outcome bar this craft feeds; `conformance\u002Fdiscovery-complete.sh` is the gate that checks the artifact exists.\n- Prove any claim made during discovery with `skills\u002Fverification\u002FSKILL.md` (evidence before assertions — a test result is a claim until you've seen it).\n- Hand the validated problem to `skills\u002Fdesign\u002FSKILL.md`; the loop continues plan → build → review from there.\n\n## Honest ceiling\nDiscovery *quality* is un-gateable — there is no CI check that an opportunity tree is good or an assumption well-chosen, just as there is none for design quality. The structural proof is only that this craft is provided, kit-distinctive, and wired to the conductor; the honest enforcement of *outcome rigor* is the Definition-of-Ready success-metric item and `discovery-complete.sh`, which this craft feeds. Name that ceiling; never let a green check imply the discovery was actually good.\n\n## Rationalizations to refuse\n| Rationalization | Why it fails |\n|---|---|\n| \"The problem is obvious, just build it.\" | Obvious problems hide the `riskiest assumption`. Name it and design a `small bet` before building — that is the cheap insurance. |\n| \"We agreed on the feature, discovery is done.\" | A feature is an output. Until you can state the `outcome over output` and how you'd measure it, discovery is not done. |\n| \"I'll run discovery and tell the human what to build.\" | You are the **discovery partner**, not the decider. The human is the PO; structure and surface, then hand back the judgment. |\n| \"Let's just pick the first solution.\" | An `opportunity solution tree` exists so the chosen opportunity is compared, not defaulted. Breadth before depth. |\n| \"We'll validate it after we ship.\" | Continuous discovery tests the riskiest assumption with a `small bet` *before* the build — post-hoc validation is the waste it prevents. |\n\n## Terminal state\nA validated problem, a measurable outcome (`outcome over output`), an explicit `opportunity solution tree`, and the `riskiest assumption` tested with a `small bet` — captured in the feature request and handed to `skills\u002Fdesign\u002FSKILL.md`. The human made every call; the discovery partner kept it honest. This skill never starts implementation.\n","# Continuous discovery — interrogate the problem before designing the solution\n\nThe kit's own problem-space discovery craft: take a felt need → a *validated* problem, a measurable outcome, and the riskiest assumption tested — before `design` shapes any solution. This is the **front of the loop** (Discover), the partner to `skills\u002Fdesign\u002FSKILL.md`'s solution-space. Continuous discovery is Teresa Torres' practice made workable for a human↔AI pair; it is a kit-original (superpowers has no discovery skill — its `brainstorming` is solution-space, the kit's `design`).\n\n> **Not to be confused with the discovery *keystone*.** `skills\u002Fusing-skills\u002FSKILL.md` is *skill*-discovery (\"which skill applies?\"). This skill is *product* discovery (\"which problem, and is it the right one?\"). Different axis entirely.\n\n\u003C!-- The frontmatter and the discipline markers below are conformance-load-bearing:\n     conformance\u002Forchestrator-loop-wired.sh greps this file for kit-distinctive markers\n     (name: continuous-discovery, discovery partner, outcome over output,\n     opportunity solution tree, riskiest assumption, small bet).\n     Edits that drop or rename them can turn the skill-spine lock RED. -->\n\n## When to use\nAt the **front of the loop**, before `design`: when the problem is taken for granted, when a slice's value rests on an assumption no one has tested, or when \"what we'll build\" is clear but \"what outcome it changes, and for whom\" is not. Every non-trivial slice deserves a discovery pass — a small one for a well-understood problem, a deliberate one when the problem is fuzzy or the bet is large.\n\n## The discovery partner — not the decider\nThe **discovery partner** is the kit's framing of the human↔AI relationship here, and it is the load-bearing constraint of this whole skill: **the human is the PO.** The agent does *not* decide what to build or declare a problem worth solving. The agent makes continuous discovery *accessible and honest*: it structures the space, asks the sharp question, names the assumption everyone is gliding past, and designs the cheap test — then hands the judgment back to the human. A partner that quietly starts deciding has stopped being a partner; that is the failure mode this skill exists to prevent. There is no Product *seat* for the same reason — the human owns the decision; the Orchestrator merely *wears the Product hat* to invoke this craft.\n\n## The flow (the proven spine)\n1. **Anchor on the outcome, not the feature — `outcome over output`.** Before anything, ask what customer or business *outcome* this work would change, and how we'd measure it. Frame the work by the outcome it moves, never by the feature shipped. A slice with no outcome hypothesis is output theatre — surface that out loud. (This is the Definition-of-Ready \"success metric \u002F hypothesis\" item, made a craft.)\n2. **Map the `opportunity solution tree`.** Build the tree *with* the human: the outcome at the root → the **opportunities** beneath it (the problems, needs, and desires that, if addressed, move the outcome) → candidate solution ideas → the experiments that would test them. The point is to make the chosen opportunity *explicit and compared*, not the first idea that arrived. Breadth before depth.\n3. **Surface and rank the `riskiest assumption`.** Every chosen opportunity and solution bets on assumptions — desirability (do they want it?), viability (does it work for the business?), feasibility (can we build it?), usability (can they use it?). Name them, then rank the **riskiest assumption**: the one that, if false, wastes the most work. That is what gets tested first — not the easiest thing to test.\n4. **Design the `small bet`.** Design the cheapest experiment that would prove or kill the riskiest assumption *before* a slice is built — a `small bet`, not a big-bang study. Continuous, not one-shot: small, frequent tests on a cadence beat a single upfront discovery phase. Define in advance what result would change the decision; a test whose outcome changes nothing is theatre.\n5. **Hand off to `design`.** Discovery's terminal state is a validated problem + a measurable outcome + the riskiest assumption tested — handed to `skills\u002Fdesign\u002FSKILL.md` to shape the solution. This skill never starts a solution; designing is `design`'s job, building is the Engineer's.\n\n## What this craft points at (it does not duplicate the infra)\n- Capture the validated problem + outcome in `templates\u002FFEATURE-REQUEST-TEMPLATE.md` (incl. its success-metric \u002F hypothesis fields).\n- The Definition-of-Ready \"success metric \u002F hypothesis\" item is the outcome bar this craft feeds; `conformance\u002Fdiscovery-complete.sh` is the gate that checks the artifact exists.\n- Prove any claim made during discovery with `skills\u002Fverification\u002FSKILL.md` (evidence before assertions — a test result is a claim until you've seen it).\n- Hand the validated problem to `skills\u002Fdesign\u002FSKILL.md`; the loop continues plan → build → review from there.\n\n## Honest ceiling\nDiscovery *quality* is un-gateable — there is no CI check that an opportunity tree is good or an assumption well-chosen, just as there is none for design quality. The structural proof is only that this craft is provided, kit-distinctive, and wired to the conductor; the honest enforcement of *outcome rigor* is the Definition-of-Ready success-metric item and `discovery-complete.sh`, which this craft feeds. Name that ceiling; never let a green check imply the discovery was actually good.\n\n## Rationalizations to refuse\n| Rationalization | Why it fails |\n|---|---|\n| \"The problem is obvious, just build it.\" | Obvious problems hide the `riskiest assumption`. Name it and design a `small bet` before building — that is the cheap insurance. |\n| \"We agreed on the feature, discovery is done.\" | A feature is an output. Until you can state the `outcome over output` and how you'd measure it, discovery is not done. |\n| \"I'll run discovery and tell the human what to build.\" | You are the **discovery partner**, not the decider. The human is the PO; structure and surface, then hand back the judgment. |\n| \"Let's just pick the first solution.\" | An `opportunity solution tree` exists so the chosen opportunity is compared, not defaulted. Breadth before depth. |\n| \"We'll validate it after we ship.\" | Continuous discovery tests the riskiest assumption with a `small bet` *before* the build — post-hoc validation is the waste it prevents. |\n\n## Terminal state\nA validated problem, a measurable outcome (`outcome over output`), an explicit `opportunity solution tree`, and the `riskiest assumption` tested with a `small bet` — captured in the feature request and handed to `skills\u002Fdesign\u002FSKILL.md`. The human made every call; the discovery partner kept it honest. This skill never starts implementation.\n",{},"main","46ed464a0bfe79d18251e771dd0fc9e1e89fe43f","skills\u002Fcontinuous-discovery",0,"https:\u002F\u002Fgithub.com\u002FSeaBrad72\u002Fsparkwright","https:\u002F\u002Fgithub.com\u002FSeaBrad72","2026-11-15 04:00:12",{"source":7,"skillName":5,"commands":31,"prompt":41,"zipAvailable":42},[32,35,38],{"tool":33,"command":34},"npx","npx skills add SeaBrad72\u002Fsparkwright --skill continuous-discovery",{"tool":36,"command":37},"bunx","bunx skills add SeaBrad72\u002Fsparkwright --skill continuous-discovery",{"tool":39,"command":40},"pnpm","pnpm dlx skills add SeaBrad72\u002Fsparkwright --skill continuous-discovery","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FSeaBrad72\u002Fsparkwright\u002Fblob\u002Fmain\u002Fskills\u002Fcontinuous-discovery\u002FSKILL.md",true,[],[45,52,59,65,72,79],{"skillNo":46,"name":47,"nameZh":6,"description":48,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":11,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":49,"categories":50,"keywords":51,"publishedAt":18},"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":53,"name":54,"nameZh":6,"description":55,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":11,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":56,"categories":57,"keywords":58,"publishedAt":18},"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":60,"name":61,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":11,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":62,"categories":63,"keywords":64,"publishedAt":18},"01M78FQQXX2CYGF8V0GVG35TTB","verification","https:\u002F\u002Fgithub.com\u002FSeaBrad72\u002Fsparkwright\u002Fblob\u002Fmain\u002Fskills\u002Fverification\u002FSKILL.md",[],[],{"skillNo":66,"name":67,"nameZh":6,"description":68,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":11,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":69,"categories":70,"keywords":71,"publishedAt":18},"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":73,"name":74,"nameZh":6,"description":75,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":11,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":76,"categories":77,"keywords":78,"publishedAt":18},"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":80,"name":81,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":11,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":82,"categories":83,"keywords":84,"publishedAt":18},"01M78FQADNX9RZQ63C7WG7HPZH","review","https:\u002F\u002Fgithub.com\u002FSeaBrad72\u002Fsparkwright\u002Fblob\u002Fmain\u002Fskills\u002Freview\u002FSKILL.md",[],[]]