grilling
Resolve material product and design decisions before planning or implementing an ambiguous large feature. Use when the user asks to be grilled or when several valid interpretations would produce meaningfully different behavior, architecture, cost, or migration work; do not use for questions, small changes, or facts the repository can answer.
- License 许可
- Apache-2.0
- 最近更新
- 2026-11-13
- Stars
- ★ 0
- Forks
- 0
01 前言 Front Matter
- name
- create-prototype
- description
- Builds new UI prototypes and makes targeted changes to existing prototypes using consumer-owned context. Use when asked to build a prototype, mock up a UI, or add, change, refine or recreate part of an existing prototype. Do not use for unrelated tasks, routine commands, Git lifecycle actions, acknowledgements or infrastructure work.
02 资源文件 4
- 参考文档references/authoring.md12.0 KB
- 参考文档workflows/clarify.md2.1 KB
- 参考文档workflows/create.md2.9 KB
- 参考文档workflows/iterate.md2.8 KB
03 SKILL.md 正文
Create prototype
When asked to create controls for variants, scenarios, adjustable previews or data-source switches, read the adjacent prototype-controls skill, even when the request never says "canvas controls". Product UI form fields and filters alone stay in this skill. Controls guidance does not establish panel availability.
Dispatch before reading workflows
Determine the branch from the original request, prior answers and current target before loading any workflow files. Use a short, targeted lookup of consumer repository instructions or the named target if needed, not a broad repository tour. The word "new" alone does not decide: "add a new tab to this prototype" is iteration.
| Request | Required read, only for this branch |
|---|---|
| Clear new build, such as "build an offline inventory prototype" | create |
| Change to an existing prototype, such as "add filtering to this page" | iterate |
| Candidate narrow edit to an existing prototype | iterate, then apply the root instructions' narrow-edit path |
| Consequential missing information, including an unresolved target or create-versus-iterate choice | clarify, then the selected build or iteration workflow |
| Behavior-preserving test-only work | Use the root instructions' task-scope guidance; do not load a prototype workflow |
| Unrelated task or routine command | Stop using this skill and continue the original task |
Load only the selected workflow, not all companions as a required-read bundle. Clarification is conditional, not a startup step. Either build or iteration can need it later, but a clear request skips it entirely. Do not load the other branch for orientation. If the selected companion is missing, report its exact installed path and stop. Do not improvise it or use the other branch.
Preserve the original request and every answer in session context throughout the work. A choice answers only its question, not unrelated scope. There is no mandatory brief file, schema, build-mode quiz or separate intake skill.
Stay in the current target and worktree unless the user explicitly requests a different destination. A reference URL is not permission to fork, duplicate, publish, delete or activate anything. Approved references are 1:1 requirements except for the requested delta, not design inspiration.
The selected create or normal iteration workflow reads the shared
authoring reference before implementation. A
candidate narrow edit follows the iteration workflow's classification step and,
when eligible, skips that reference and its context dependencies.
It owns context resolution, design evidence, data boundaries and completion.
It does not load another workflow. Install product-design, setup-skills
and the context-templates library alongside this skill.
Missing installed dependencies stop
the affected work with their exact paths, never trigger automatic installation.
Context research uses the centralized library through setup-skills. The shared authoring reference declares this consumer's context reads. Templates are setup inputs, not additional per-task reads or active product facts.
04 同仓库更多 Skills
pr-babysit
Watch a GitHub pull request through CI and bot reviews (Copilot, Codex, Claude, Greptile), fix feedback locally as it arrives, and push only once everything has settled. Use when asked to babysit, shepherd, or watch a PR until it is green and reviewed. Not for a one-off look at PR comments or CI status.
okf
Read, create, and update Open Knowledge Format (OKF) knowledge bundles — wikis of markdown concept files with YAML frontmatter, per Google's OKF v0.1 spec. Use when browsing or consuming an OKF bundle, when asked to create a knowledge wiki for a codebase or data system, or when editing a directory of .md files that carry `type:` frontmatter alongside index.md/log.md files. Covers the frontmatter schema, bundle-root-relative linking, index.md maintenance, and log.md changelog conventions.
design-doc
Write an explicit design document, RFC, or consequential architecture proposal with adversarial review. Use for decisions with material ownership, compatibility, migration, or operational consequences. Do not activate for ordinary comparisons, syntax questions, or small implementation choices.
background-terminals
Run and manage long-lived shell commands in background terminals. Use for dev servers, watchers, streaming builds, and other commands that should keep running while the agent continues working.
writing-for-agents
Write or revise skills, AGENTS.md, CLAUDE.md, prompts, and other instructions consumed by coding agents. Use when agent behavior, skill routing, instruction structure, or prompt maintenance is part of the task.
wait-what
Stop. That last message did not land: re-pitch it.