[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M75JFPPEY86F2NWAJC5RD6X3":3,"skill-same:01M75JFPPEY86F2NWAJC5RD6X3":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":25,"commitSha":26,"skillDirectory":27,"forksCount":14,"repositoryHtmlUrl":28,"repositoryHomepageUrl":6,"ownerHtmlUrl":29,"ownerBlogUrl":30,"ownerDescription":6,"updatedAt":31,"install":32,"resources":44},"01M75JFPPEY86F2NWAJC5RD6X3","ask-cape",null,"Ask which cape skill or flow fits your situation. A router over the skills in this repo.","colenet-gmbh\u002Fcolenet-claude-code-plugin","colenet-gmbh","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F27774717?v=4","01M75JEMA67P2CEDJC0GZCWZDV","Colenet GmbH","01M75JEMASFM02PQP38B77BBN5",0,"MIT","https:\u002F\u002Fgithub.com\u002Fcolenet-gmbh\u002Fcolenet-claude-code-plugin\u002Fblob\u002Fmain\u002Fskills_source\u002Fmeta\u002Fask-cape\u002FSKILL.md",[],[],"2026-11-12 13:55:43",false,"---\nname: ask-cape\ndescription: Ask which cape skill or flow fits your situation. A router over the skills in this repo.\ndisable-model-invocation: true\n---\n\n# Ask Cape\n\nYou don't remember every skill, so ask.\n\nA **flow** is a path through the skills. Most work runs along one **main flow**; on-ramps merge onto it, and a few skills stand alone.\n\n## The main flow: idea → ship\n\nThe route most work travels. You have an idea and want it built, reviewed, and shipped.\n\n1. **`\u002Fgrill-with-docs`** — sharpen the idea by interview. Start here when you **have a codebase**: it's stateful, seeding and retaining what it learns in the documentation referenced in the `CONTEXT.md` map — like arc42 and the domain glossary. (No codebase? Use `\u002Fgrill-me` — see Standalone.)\n2. **Branch — can you settle every question in conversation?** If a question needs a runnable answer (state, business logic, a UI you have to see), detour through a prototype, bridged by **`\u002Fhandoff`** in both directions (see Crossing sessions):\n   - **`\u002Fhandoff`** out, then open a fresh session against that file,\n   - **`\u002Fprototype`** to answer the question with throwaway code,\n   - **`\u002Fhandoff`** back what you learned, and reference it from the original idea thread.\n3. **Branch — a whole feature, or a single issue?**\n   - **A whole feature** → **`\u002Ffeature`** → **`\u002Fsplit`** → **`\u002Fbuild`**:\n     - **`\u002Ffeature`** turns the thread into a **reviewed spec** — by designing its architecture and checking it through various review lenses; creating a spec with the key decisions on top for **human sign-off**, the internal design below.\n     - **`\u002Fsplit`** splits the spec into independently-grabbable, vertically-sliced issues.\n     - **`\u002Fbuild`** orchestrates the implementation of the feature by driving the issues to done, internally calling **`\u002Fimplement`** per issue and dynamically handling upcoming new issues and reviewing the result.\n   - **A single issue** → **`\u002Fimplement`** directly — it builds the one issue (`\u002Ftdd` inside) and reviews the implementation.\n\nEither way, done means **every review passed**.\n\n### Context hygiene\n\n**`\u002Fimplement` always starts in a fresh context.** Its input is a **written-out issue** plus the spec — never the live planning thread. One contract, whichever flow you're in:\n\n- **Whole feature** — keep `\u002Fgrill-with-docs` → `\u002Ffeature` → `\u002Fsplit` in **one unbroken context window** so they build on the same thinking. **`\u002Fbuild` then starts fresh** and orchestrates: it drives each issue through `\u002Fimplement`, each in its own fresh context.\n- **Single issue** — **write the issue out**, then run `\u002Fimplement` fresh against it. Even the one-off issue gets persisted first — that's the price of `\u002Fimplement` having a single, predictable contract.\n\nThe limit on the planning window is the **[smart zone](https:\u002F\u002Fwww.aihero.dev\u002Fai-coding-dictionary\u002Fsmart-zone)**: the window (~120k tokens on state-of-the-art models) within which the model still reasons sharply. If a session approaches it before `\u002Fsplit`, don't push on degraded — `\u002Fhandoff` and continue in a fresh thread.\n\n## On-ramps\n\nA starting situation that generates work, then merges onto the main flow.\n\n- **Bugs and requests piling up** → **`\u002Ftriage`**. It moves each raw item through triage roles into an agent-ready brief. Where that brief rejoins the main flow depends on its size:\n  - a **single issue** is already a written-out brief → it joins at **`\u002Fimplement`**;\n  - a **feature-sized** item joins earlier: the grilling session `\u002Ftriage` already runs to sharpen it *is* the main flow starting, so it continues through **`\u002Ffeature`** → **`\u002Fsplit`** → **`\u002Fbuild`**.\n\n  Triage is **backlog management** for anything **not yet worked up** — bug reports, incoming feature requests, your own captured ideas, whatever their origin. Only what's already agent-ready is out: issues `\u002Fsplit` produced need no triage, so **don't re-triage them**.\n\n- **Facts missing before you can plan** → **`\u002Fresearch`**. It delegates the reading legwork to a background agent — investigate the question against **primary sources** (official docs, source code, specs, first-party APIs) and leave a **cited note in the repo**. That note then feeds **`\u002Fgrill-with-docs`** upstream, so a sharp interview leans on facts rather than guesses. (For a broad, multi-source web report with adversarial fact-checking, that's `\u002Fdeep-research`, not this.)\n\n## Codebase & harness health\n\nNot feature work — upkeep.\n\n- **`\u002Farchitect`** — a deep skill for shaping structure, activated in context (and by hand). Its **deepening** aspect you run in a spare moment to keep the codebase good for agents to operate in: it surfaces opportunities, and picking one *generates an idea* you take into the main flow at `\u002Fgrill-with-docs`. Its other aspects — codebase\u002Fsoftware design, domain modelling, arc42 docs — fire *inside* other skills as the work needs them (e.g. `\u002Ftdd` reaching for design vocabulary, domain modelling during `\u002Fgrill-with-docs`), several at once.\n- **`\u002Fimprove`** — where `\u002Farchitect` keeps the *codebase* good for agents, `\u002Fimprove` keeps the *harness* good. The two also run very differently: `\u002Farchitect` is model-invokable and mostly fires *inside* other skills as the work needs it, while `\u002Fimprove` is an interactive session that the **user always starts** — the agent may suggest it, never launch it. Run it anytime a flow or session showed friction (\"that should've gone better\"): it finds the friction, traces the root cause, and applies a proportional fix in the project's own harness. When it finds a genuinely general improvement, it suggests sending it to the cape community as a GitHub issue.\n\n## Crossing sessions\n\n- **`\u002Fhandoff`** — when a thread is full or you need to branch off (e.g. into a `\u002Fprototype` session), this compacts the conversation into a markdown file. You don't continue in place — you **open a new session and reference that file** to carry the context across. It's the bridge between context windows, in either direction. Use it when you want a **fresh session** but need the **current conversation preserved**.\n- **`\u002Fcompact`** (built-in) — stay in the **same conversation**, letting the earlier turns be summarized. Use it at **intentional breaks between phases**, when you don't mind losing the verbatim history. Don't compact mid-phase — the agent can lose its way. `\u002Fhandoff` forks; `\u002Fcompact` continues.\n\n## Standalone\n\nOff the main flow entirely.\n\n- **`\u002Fgrill-me`** — the same relentless interview as `\u002Fgrill-with-docs`, but for when you have **no codebase**. Stateless: it saves nothing locally, builds no docs. Reach for it to sharpen any plan or design that doesn't live in a repo.\n- **`\u002Fwriting-great-skills`** — reference for writing and editing skills well. The authoring standard for every cape skill.\n\n## Precondition\n\n- **`\u002Fcape:setup`** — run once per repo before your first engineering flow: it scaffolds the docs the skills assume — the `docs\u002Fwork\u002F` board, the issue tracker, and `CONTEXT.md` with the domain glossary.\n\n---\n\nThe skill names above are shown in short form for reading. In the plugin they are namespaced as `cape:\u003Cname>` (e.g. `cape:feature`, `cape:implement`) — that is how one skill invokes another and how you call them by hand. Pull a newer cape into a repo with `\u002Fplugin update`.\n","# Ask Cape\n\nYou don't remember every skill, so ask.\n\nA **flow** is a path through the skills. Most work runs along one **main flow**; on-ramps merge onto it, and a few skills stand alone.\n\n## The main flow: idea → ship\n\nThe route most work travels. You have an idea and want it built, reviewed, and shipped.\n\n1. **`\u002Fgrill-with-docs`** — sharpen the idea by interview. Start here when you **have a codebase**: it's stateful, seeding and retaining what it learns in the documentation referenced in the `CONTEXT.md` map — like arc42 and the domain glossary. (No codebase? Use `\u002Fgrill-me` — see Standalone.)\n2. **Branch — can you settle every question in conversation?** If a question needs a runnable answer (state, business logic, a UI you have to see), detour through a prototype, bridged by **`\u002Fhandoff`** in both directions (see Crossing sessions):\n   - **`\u002Fhandoff`** out, then open a fresh session against that file,\n   - **`\u002Fprototype`** to answer the question with throwaway code,\n   - **`\u002Fhandoff`** back what you learned, and reference it from the original idea thread.\n3. **Branch — a whole feature, or a single issue?**\n   - **A whole feature** → **`\u002Ffeature`** → **`\u002Fsplit`** → **`\u002Fbuild`**:\n     - **`\u002Ffeature`** turns the thread into a **reviewed spec** — by designing its architecture and checking it through various review lenses; creating a spec with the key decisions on top for **human sign-off**, the internal design below.\n     - **`\u002Fsplit`** splits the spec into independently-grabbable, vertically-sliced issues.\n     - **`\u002Fbuild`** orchestrates the implementation of the feature by driving the issues to done, internally calling **`\u002Fimplement`** per issue and dynamically handling upcoming new issues and reviewing the result.\n   - **A single issue** → **`\u002Fimplement`** directly — it builds the one issue (`\u002Ftdd` inside) and reviews the implementation.\n\nEither way, done means **every review passed**.\n\n### Context hygiene\n\n**`\u002Fimplement` always starts in a fresh context.** Its input is a **written-out issue** plus the spec — never the live planning thread. One contract, whichever flow you're in:\n\n- **Whole feature** — keep `\u002Fgrill-with-docs` → `\u002Ffeature` → `\u002Fsplit` in **one unbroken context window** so they build on the same thinking. **`\u002Fbuild` then starts fresh** and orchestrates: it drives each issue through `\u002Fimplement`, each in its own fresh context.\n- **Single issue** — **write the issue out**, then run `\u002Fimplement` fresh against it. Even the one-off issue gets persisted first — that's the price of `\u002Fimplement` having a single, predictable contract.\n\nThe limit on the planning window is the **[smart zone](https:\u002F\u002Fwww.aihero.dev\u002Fai-coding-dictionary\u002Fsmart-zone)**: the window (~120k tokens on state-of-the-art models) within which the model still reasons sharply. If a session approaches it before `\u002Fsplit`, don't push on degraded — `\u002Fhandoff` and continue in a fresh thread.\n\n## On-ramps\n\nA starting situation that generates work, then merges onto the main flow.\n\n- **Bugs and requests piling up** → **`\u002Ftriage`**. It moves each raw item through triage roles into an agent-ready brief. Where that brief rejoins the main flow depends on its size:\n  - a **single issue** is already a written-out brief → it joins at **`\u002Fimplement`**;\n  - a **feature-sized** item joins earlier: the grilling session `\u002Ftriage` already runs to sharpen it *is* the main flow starting, so it continues through **`\u002Ffeature`** → **`\u002Fsplit`** → **`\u002Fbuild`**.\n\n  Triage is **backlog management** for anything **not yet worked up** — bug reports, incoming feature requests, your own captured ideas, whatever their origin. Only what's already agent-ready is out: issues `\u002Fsplit` produced need no triage, so **don't re-triage them**.\n\n- **Facts missing before you can plan** → **`\u002Fresearch`**. It delegates the reading legwork to a background agent — investigate the question against **primary sources** (official docs, source code, specs, first-party APIs) and leave a **cited note in the repo**. That note then feeds **`\u002Fgrill-with-docs`** upstream, so a sharp interview leans on facts rather than guesses. (For a broad, multi-source web report with adversarial fact-checking, that's `\u002Fdeep-research`, not this.)\n\n## Codebase & harness health\n\nNot feature work — upkeep.\n\n- **`\u002Farchitect`** — a deep skill for shaping structure, activated in context (and by hand). Its **deepening** aspect you run in a spare moment to keep the codebase good for agents to operate in: it surfaces opportunities, and picking one *generates an idea* you take into the main flow at `\u002Fgrill-with-docs`. Its other aspects — codebase\u002Fsoftware design, domain modelling, arc42 docs — fire *inside* other skills as the work needs them (e.g. `\u002Ftdd` reaching for design vocabulary, domain modelling during `\u002Fgrill-with-docs`), several at once.\n- **`\u002Fimprove`** — where `\u002Farchitect` keeps the *codebase* good for agents, `\u002Fimprove` keeps the *harness* good. The two also run very differently: `\u002Farchitect` is model-invokable and mostly fires *inside* other skills as the work needs it, while `\u002Fimprove` is an interactive session that the **user always starts** — the agent may suggest it, never launch it. Run it anytime a flow or session showed friction (\"that should've gone better\"): it finds the friction, traces the root cause, and applies a proportional fix in the project's own harness. When it finds a genuinely general improvement, it suggests sending it to the cape community as a GitHub issue.\n\n## Crossing sessions\n\n- **`\u002Fhandoff`** — when a thread is full or you need to branch off (e.g. into a `\u002Fprototype` session), this compacts the conversation into a markdown file. You don't continue in place — you **open a new session and reference that file** to carry the context across. It's the bridge between context windows, in either direction. Use it when you want a **fresh session** but need the **current conversation preserved**.\n- **`\u002Fcompact`** (built-in) — stay in the **same conversation**, letting the earlier turns be summarized. Use it at **intentional breaks between phases**, when you don't mind losing the verbatim history. Don't compact mid-phase — the agent can lose its way. `\u002Fhandoff` forks; `\u002Fcompact` continues.\n\n## Standalone\n\nOff the main flow entirely.\n\n- **`\u002Fgrill-me`** — the same relentless interview as `\u002Fgrill-with-docs`, but for when you have **no codebase**. Stateless: it saves nothing locally, builds no docs. Reach for it to sharpen any plan or design that doesn't live in a repo.\n- **`\u002Fwriting-great-skills`** — reference for writing and editing skills well. The authoring standard for every cape skill.\n\n## Precondition\n\n- **`\u002Fcape:setup`** — run once per repo before your first engineering flow: it scaffolds the docs the skills assume — the `docs\u002Fwork\u002F` board, the issue tracker, and `CONTEXT.md` with the domain glossary.\n\n---\n\nThe skill names above are shown in short form for reading. In the plugin they are namespaced as `cape:\u003Cname>` (e.g. `cape:feature`, `cape:implement`) — that is how one skill invokes another and how you call them by hand. Pull a newer cape into a repo with `\u002Fplugin update`.\n",{"name":5,"description":7,"disable-model-invocation":24},true,"main","47dbd725f1dd37f2c0deed3f223249b2cdcbb5a8","skills_source\u002Fmeta\u002Fask-cape","https:\u002F\u002Fgithub.com\u002Fcolenet-gmbh\u002Fcolenet-claude-code-plugin","https:\u002F\u002Fgithub.com\u002Fcolenet-gmbh","https:\u002F\u002Fwww.colenet.de","2026-10-05 07:40:40",{"source":8,"skillName":5,"commands":33,"prompt":43,"zipAvailable":24},[34,37,40],{"tool":35,"command":36},"npx","npx skills add colenet-gmbh\u002Fcolenet-claude-code-plugin --skill ask-cape",{"tool":38,"command":39},"bunx","bunx skills add colenet-gmbh\u002Fcolenet-claude-code-plugin --skill ask-cape",{"tool":41,"command":42},"pnpm","pnpm dlx skills add colenet-gmbh\u002Fcolenet-claude-code-plugin --skill ask-cape","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fcolenet-gmbh\u002Fcolenet-claude-code-plugin\u002Fblob\u002Fmain\u002Fskills_source\u002Fmeta\u002Fask-cape\u002FSKILL.md",[],[46,53,60,67,74,81],{"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},"01M75JGH2BKSPNHEFPK7GGWWX9","triage","Backlog management — move not-yet-worked-up items through triage roles into agent-ready briefs, whatever their origin. Use when the user wants to sort or work up a backlog of bug reports, feature requests, or ideas, mentions \"triage\", \"triagieren\", \"Backlog sichten\", \"Backlog pflegen\", \"sort the backlog\", \"manage the backlog\", or points at an untriaged item in docs\u002Fwork\u002F01-backlog.","https:\u002F\u002Fgithub.com\u002Fcolenet-gmbh\u002Fcolenet-claude-code-plugin\u002Fblob\u002Fmain\u002Fskills_source\u002Futility\u002Ftriage\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},"01M75JGDZC06KVV60HW33NKMZV","teach","Teach the user a new skill or concept over multiple sessions, turning the current directory into a stateful teaching workspace grounded in a mission, trusted resources, and beautiful lessons. Use when the user wants to learn or be taught something, or says \"teach me\", \"bring mir bei\", \"lehre mich\", \"ich will X lernen\", \"mach mich fit in\", \"onboard me on\", \"help me learn\".","https:\u002F\u002Fgithub.com\u002Fcolenet-gmbh\u002Fcolenet-claude-code-plugin\u002Fblob\u002Fmain\u002Fskills_source\u002Futility\u002Fteach\u002FSKILL.md",[],[],{"skillNo":61,"name":62,"nameZh":6,"description":63,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":64,"categories":65,"keywords":66,"publishedAt":19},"01M75JG9EF7354J71JQRZVQ82X","research","Delegate reading legwork to a background agent — investigate a question against high-trust primary sources (official docs, source code, specs, first-party APIs) and leave the findings as a cited Markdown note in the repo. Use when engineering work needs docs, API, or codebase facts gathered before building, or the user says \"recherchiere das\", \"schau in den Docs nach\", \"lies dich in X ein\", \"research this\", \"gather the facts on\", \"read up on\". For a broad, multi-source web report with adversarial fact-checking, reach for deep-research instead; cape:research is the lighter, repo-bound feeder that leaves a note behind.","https:\u002F\u002Fgithub.com\u002Fcolenet-gmbh\u002Fcolenet-claude-code-plugin\u002Fblob\u002Fmain\u002Fskills_source\u002Futility\u002Fresearch\u002FSKILL.md",[],[],{"skillNo":68,"name":69,"nameZh":6,"description":70,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":71,"categories":72,"keywords":73,"publishedAt":19},"01M75JG5R86XRG8B6EYX3JB06H","prototype","Build a throwaway prototype to answer a design question. Use when the user wants to sanity-check whether a state model or logic feels right, or explore what a UI should look like.","https:\u002F\u002Fgithub.com\u002Fcolenet-gmbh\u002Fcolenet-claude-code-plugin\u002Fblob\u002Fmain\u002Fskills_source\u002Futility\u002Fprototype\u002FSKILL.md",[],[],{"skillNo":75,"name":76,"nameZh":6,"description":77,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":78,"categories":79,"keywords":80,"publishedAt":19},"01M75JG46T7P54KKF80X22HF4Z","handoff","Compact the current conversation into a handoff document for another agent to pick up.","https:\u002F\u002Fgithub.com\u002Fcolenet-gmbh\u002Fcolenet-claude-code-plugin\u002Fblob\u002Fmain\u002Fskills_source\u002Futility\u002Fhandoff\u002FSKILL.md",[],[],{"skillNo":82,"name":83,"nameZh":6,"description":84,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":85,"categories":86,"keywords":87,"publishedAt":19},"01M75JG2EM4XSXC32SRRN37VRA","grilling","Interview the user relentlessly about a plan or design. Use when the user wants to stress-test a plan before building, or uses any 'grill' trigger phrases.","https:\u002F\u002Fgithub.com\u002Fcolenet-gmbh\u002Fcolenet-claude-code-plugin\u002Fblob\u002Fmain\u002Fskills_source\u002Futility\u002Fgrilling\u002FSKILL.md",[],[]]