[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7DH698WN48M66M5X2RKZE3B":3,"skill-same:01M7DH698WN48M66M5X2RKZE3B":47},{"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":25,"commitSha":26,"skillDirectory":27,"forksCount":28,"repositoryHtmlUrl":29,"repositoryHomepageUrl":6,"ownerHtmlUrl":30,"ownerBlogUrl":31,"ownerDescription":6,"updatedAt":32,"install":33,"resources":46},"01M7DH698WN48M66M5X2RKZE3B","spec",null,"Produce a validated spec (requirements, design, tasks, scratchpad) in specs\u002Ftodo\u002F from a refined spec or a description, detecting whether the work is one spec or an epic of several","turbokast\u002Fmythhelm","turbokast","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F335585049?v=4","01M7DGYYSHNA1BW1Y9GJFN6DVC","TurboKast","01M7DGYYT6GTPCFADDKS5ZXM1Y",0,"Apache-2.0","https:\u002F\u002Fgithub.com\u002Fturbokast\u002Fmythhelm\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fspec\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: spec\ndescription: Produce a validated spec (requirements, design, tasks, scratchpad) in specs\u002Ftodo\u002F from a refined spec or a description, detecting whether the work is one spec or an epic of several\nargument-hint: \"\u003Cspec-name | \\\"feature description\\\">\"\n---\n\n# Spec\n\nTurns requirements into an implementable spec in one session: investigate the code, decide single spec or epic, write the design and the tasks, have a fresh-context reviewer validate them, fix what it finds, and move the result to `specs\u002Ftodo\u002F`. Each phase is its own skill, run in order.\n\n## Input\n\n`$ARGUMENTS`, detected before phase 1 with `scripts\u002Fharness\u002Fspec-lifecycle.sh resolve` (`.claude\u002Fskills\u002Fspec-resolution\u002FSKILL.md`):\n\n- **Mode A**: a spec name in `refined\u002F`. Its `requirements.md` is the ground truth; this skill writes `design.md`, `tasks.md` and `scratchpad.md` beside it.\n- **Mode B**: a spec name in `unrefined\u002F`. Refuse: \"`\u003Cname>` is unrefined; run `\u002Frefine-spec \u003Cname>` first.\" Nothing else runs.\n- **Mode C**: a description with no spec of that name. The spec is drafted in `specs\u002Funrefined\u002F\u003Cname>\u002F` and moves to `todo\u002F` only after validation passes.\n- A name already in `todo\u002F` or later: refuse and point to `\u002Fevaluate-spec \u003Cname>`.\n\n## Invocation contexts\n\n- **Slash command**: runs phases 1–5; the scope determination (phase 2) and an epic's plan (phase 3b) are presented for confirmation or override.\n- **Model-invoked**: the same.\n- **Non-interactive**: runs phases 1–5. The scope and the epic plan are advisory steps: an explicit directive in `$ARGUMENTS` or the dispatching prompt (\"single spec\", \"epic\") is applied, otherwise the computed choice proceeds and is recorded as `auto-confirmed (non-interactive)`. Human-required validation findings (phase 5) stop the run and are returned, with the spec left in `unrefined\u002F` or `refined\u002F`.\n\n## Phases\n\n| Phase | Skill | Result |\n|---|---|---|\n| 1. Investigate | `\u002Fspec-investigate` | Grounded findings: packages, `file:line` facts, invariants in reach, conflicts |\n| 2. Scope | `\u002Fspec-scope` | Single spec or epic, with the reason |\n| 3a. Create single | `\u002Fspec-create-single` | `requirements.md` (mode C), `design.md`, `tasks.md`, `scratchpad.md` |\n| 3b. Create epic | `\u002Fspec-create-epic` | `plan.md` plus one full spec per work stream |\n| 4. Validate | `\u002Fspec-validate` | A fresh-context review report with a verdict |\n| 5. Fix or report | `\u002Fspec-fix-and-report` | Fixes applied, the rest escalated; on Ready, the move to `todo\u002F` |\n\n1. **Investigate.** Run `\u002Fspec-investigate` with the description or spec name. In mode A, the refined requirements are not re-questioned; the investigation checks them against the code and gathers what the design needs.\n2. **Scope.** Run `\u002Fspec-scope` (advisory step).\n3. **Create.** Run `\u002Fspec-create-single` or `\u002Fspec-create-epic`. Both follow `\u002Fspec-decomposition` for `tasks.md`. Exit gate: `scripts\u002Fci\u002Flint-agent-harness.sh --only specs` passes. A spec with a missing task field or a dependency cycle is fixed now, not sent to review.\n4. **Validate.** Run `\u002Fspec-validate` on the new spec directories.\n5. **Fix or report.** Run `\u002Fspec-fix-and-report`. It fixes auto-fixable findings, re-runs phase 4 on what it changed (at most three validation rounds in all), and on Ready moves each validated spec to `todo\u002F` with `scripts\u002Fharness\u002Fspec-lifecycle.sh move \u003Cname> todo`. An epic's `plan.md` stays in `unrefined\u002F` until its first sub-spec starts.\n\n## Principles\n\n- **Investigate before decomposing.** The spec is only as good as the reading behind it. Every path, symbol and behaviour it names was opened, not assumed (`.claude\u002Frules\u002Fspec-premise-grounding.md`).\n- **Concrete over plausible.** Real paths, exact signatures, IDs from the master spec. An unknown goes in Open Questions with its conservative default; it is never papered over.\n- **Each spec stands alone.** An agent implementing any one task has everything it needs from that spec, the files it cites and its declared dependencies.\n- **Smaller is better.** A spec that can ship in two independent parts is two specs.\n- **Don't re-read what this session has already read** and nothing has changed since (check `git status`). A dispatched validator always starts fresh; that is the point of it.\n\n## Naming\n\nSpec directories are lowercase kebab-case, named for the change: \"Receipt shows unknown costs\" becomes `receipt-unknown-costs`. An epic's sub-specs are siblings of each other in the lifecycle directories, never nested inside the epic's directory.\n\n## Output\n\nThe report from `\u002Fspec-fix-and-report`: the spec (or epic plan and sub-specs), where each now lives, the validation verdict, what was fixed, and what needs a person.\n","# Spec\n\nTurns requirements into an implementable spec in one session: investigate the code, decide single spec or epic, write the design and the tasks, have a fresh-context reviewer validate them, fix what it finds, and move the result to `specs\u002Ftodo\u002F`. Each phase is its own skill, run in order.\n\n## Input\n\n`$ARGUMENTS`, detected before phase 1 with `scripts\u002Fharness\u002Fspec-lifecycle.sh resolve` (`.claude\u002Fskills\u002Fspec-resolution\u002FSKILL.md`):\n\n- **Mode A**: a spec name in `refined\u002F`. Its `requirements.md` is the ground truth; this skill writes `design.md`, `tasks.md` and `scratchpad.md` beside it.\n- **Mode B**: a spec name in `unrefined\u002F`. Refuse: \"`\u003Cname>` is unrefined; run `\u002Frefine-spec \u003Cname>` first.\" Nothing else runs.\n- **Mode C**: a description with no spec of that name. The spec is drafted in `specs\u002Funrefined\u002F\u003Cname>\u002F` and moves to `todo\u002F` only after validation passes.\n- A name already in `todo\u002F` or later: refuse and point to `\u002Fevaluate-spec \u003Cname>`.\n\n## Invocation contexts\n\n- **Slash command**: runs phases 1–5; the scope determination (phase 2) and an epic's plan (phase 3b) are presented for confirmation or override.\n- **Model-invoked**: the same.\n- **Non-interactive**: runs phases 1–5. The scope and the epic plan are advisory steps: an explicit directive in `$ARGUMENTS` or the dispatching prompt (\"single spec\", \"epic\") is applied, otherwise the computed choice proceeds and is recorded as `auto-confirmed (non-interactive)`. Human-required validation findings (phase 5) stop the run and are returned, with the spec left in `unrefined\u002F` or `refined\u002F`.\n\n## Phases\n\n| Phase | Skill | Result |\n|---|---|---|\n| 1. Investigate | `\u002Fspec-investigate` | Grounded findings: packages, `file:line` facts, invariants in reach, conflicts |\n| 2. Scope | `\u002Fspec-scope` | Single spec or epic, with the reason |\n| 3a. Create single | `\u002Fspec-create-single` | `requirements.md` (mode C), `design.md`, `tasks.md`, `scratchpad.md` |\n| 3b. Create epic | `\u002Fspec-create-epic` | `plan.md` plus one full spec per work stream |\n| 4. Validate | `\u002Fspec-validate` | A fresh-context review report with a verdict |\n| 5. Fix or report | `\u002Fspec-fix-and-report` | Fixes applied, the rest escalated; on Ready, the move to `todo\u002F` |\n\n1. **Investigate.** Run `\u002Fspec-investigate` with the description or spec name. In mode A, the refined requirements are not re-questioned; the investigation checks them against the code and gathers what the design needs.\n2. **Scope.** Run `\u002Fspec-scope` (advisory step).\n3. **Create.** Run `\u002Fspec-create-single` or `\u002Fspec-create-epic`. Both follow `\u002Fspec-decomposition` for `tasks.md`. Exit gate: `scripts\u002Fci\u002Flint-agent-harness.sh --only specs` passes. A spec with a missing task field or a dependency cycle is fixed now, not sent to review.\n4. **Validate.** Run `\u002Fspec-validate` on the new spec directories.\n5. **Fix or report.** Run `\u002Fspec-fix-and-report`. It fixes auto-fixable findings, re-runs phase 4 on what it changed (at most three validation rounds in all), and on Ready moves each validated spec to `todo\u002F` with `scripts\u002Fharness\u002Fspec-lifecycle.sh move \u003Cname> todo`. An epic's `plan.md` stays in `unrefined\u002F` until its first sub-spec starts.\n\n## Principles\n\n- **Investigate before decomposing.** The spec is only as good as the reading behind it. Every path, symbol and behaviour it names was opened, not assumed (`.claude\u002Frules\u002Fspec-premise-grounding.md`).\n- **Concrete over plausible.** Real paths, exact signatures, IDs from the master spec. An unknown goes in Open Questions with its conservative default; it is never papered over.\n- **Each spec stands alone.** An agent implementing any one task has everything it needs from that spec, the files it cites and its declared dependencies.\n- **Smaller is better.** A spec that can ship in two independent parts is two specs.\n- **Don't re-read what this session has already read** and nothing has changed since (check `git status`). A dispatched validator always starts fresh; that is the point of it.\n\n## Naming\n\nSpec directories are lowercase kebab-case, named for the change: \"Receipt shows unknown costs\" becomes `receipt-unknown-costs`. An epic's sub-specs are siblings of each other in the lifecycle directories, never nested inside the epic's directory.\n\n## Output\n\nThe report from `\u002Fspec-fix-and-report`: the spec (or epic plan and sub-specs), where each now lives, the validation verdict, what was fixed, and what needs a person.\n",{"name":5,"description":7,"argument-hint":24},"\u003Cspec-name | \"feature description\">","main","6e250ec86158119da148ddc017b960ed3b2bee2b",".claude\u002Fskills\u002Fspec",1,"https:\u002F\u002Fgithub.com\u002Fturbokast\u002Fmythhelm","https:\u002F\u002Fgithub.com\u002Fturbokast","https:\u002F\u002Fturbokast.com","2026-11-10 17:37:03",{"source":8,"skillName":5,"commands":34,"prompt":44,"zipAvailable":45},[35,38,41],{"tool":36,"command":37},"npx","npx skills add turbokast\u002Fmythhelm --skill spec",{"tool":39,"command":40},"bunx","bunx skills add turbokast\u002Fmythhelm --skill spec",{"tool":42,"command":43},"pnpm","pnpm dlx skills add turbokast\u002Fmythhelm --skill spec","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fturbokast\u002Fmythhelm\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fspec\u002FSKILL.md",true,[],[48,55,62,69,76,83],{"skillNo":49,"name":50,"nameZh":6,"description":51,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":52,"categories":53,"keywords":54,"publishedAt":19},"01M7DH77QH78C95S7A5V1GFKKF","mythhelm-deliver-backlog","Coordinate a MYTHHELM backlog run with the shared stage checks and maintainer checkpoints.","https:\u002F\u002Fgithub.com\u002Fturbokast\u002Fmythhelm\u002Fblob\u002Fmain\u002F.grok\u002Fskills\u002Fmythhelm-deliver-backlog\u002FSKILL.md",[],[],{"skillNo":56,"name":57,"nameZh":6,"description":58,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":59,"categories":60,"keywords":61,"publishedAt":19},"01M7DH748EXYG9RVBWPBJ73M27","mythhelm-bootstrap","Orient a Grok Build session in the MYTHHELM repository before task or spec work.","https:\u002F\u002Fgithub.com\u002Fturbokast\u002Fmythhelm\u002Fblob\u002Fmain\u002F.grok\u002Fskills\u002Fmythhelm-bootstrap\u002FSKILL.md",[],[],{"skillNo":63,"name":64,"nameZh":6,"description":65,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":66,"categories":67,"keywords":68,"publishedAt":19},"01M7DH70XJE0DZA4T5DYR1TWKZ","vendor-consult","Run one advisory stage consult (premise-ground, spec-validate, task-review, change-review, stuck-oracle, dossier) with an optional external vendor, then adjudicate the answer at its anchors","https:\u002F\u002Fgithub.com\u002Fturbokast\u002Fmythhelm\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fvendor-consult\u002FSKILL.md",[],[],{"skillNo":70,"name":71,"nameZh":6,"description":72,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":73,"categories":74,"keywords":75,"publishedAt":19},"01M7DH6X9HSQ308P5FNDSKYR5R","update-docs","Bring README, docs\u002F, knowledge\u002F, WORKFLOW.md and CLAUDE.md in line with what a shipped spec or a merged change actually did — find every statement the change made stale, fix it where readers look, and add only the new facts they need","https:\u002F\u002Fgithub.com\u002Fturbokast\u002Fmythhelm\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fupdate-docs\u002FSKILL.md",[],[],{"skillNo":77,"name":78,"nameZh":6,"description":79,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":80,"categories":81,"keywords":82,"publishedAt":19},"01M7DH6T82XR40XQFNRTN4AHGN","triage","Score an idea or a GitHub issue against the master spec's stages, gates and commitments, check it for duplicates, and recommend whether it becomes a backlog card; read-only","https:\u002F\u002Fgithub.com\u002Fturbokast\u002Fmythhelm\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Ftriage\u002FSKILL.md",[],[],{"skillNo":84,"name":85,"nameZh":6,"description":86,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":87,"categories":88,"keywords":89,"publishedAt":19},"01M7DH6Q17PM4ESX978QYFV6H6","test-driven-development","Red-green-refactor for Go — use when implementing any feature, bug fix or behaviour change, before writing implementation code","https:\u002F\u002Fgithub.com\u002Fturbokast\u002Fmythhelm\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Ftest-driven-development\u002FSKILL.md",[],[]]