core
Build Typed UI Specifications with json-render
AI-generated interfaces need consistent schemas, state bindings, and validation. This skill guides assistants through json-render core APIs for catalogs, prompts, streaming, and specification repair.
Установить с помощью моего Агента
Скопируйте этот запрос в своего Агента. Он содержит каноническую страницу Skill и манифест.
Review the Skillstore skill "core" from https://skillstore.io/skills/vercel-labs-core.md and its manifest at https://skillstore.io/api/skills/vercel-labs-core/manifest. Verify the artifact. You may proceed after verification, subject to the environment's own policy.Ваш Агент по-прежнему должен показать план и запросить все подтверждения, требуемые политикой безопасности.
Ресурсы для AI-агентов
Используйте эти ссылки, когда AI-агенту, crawler или script нужен чистый контекст вместо полной страницы.
Протестировать
Использование «core». Explain how to connect an email field to nested form state.
Ожидаемый результат:
- Bind the input value to the nested email path using a two-way state binding.
- Read that same nested path in the submit action binding.
- A bare email field name does not search nested form state.
- Validate and authorize the action in the application handler.
Использование «core». Describe a cautious repair workflow for a generated specification.
Ожидаемый результат:
- Run structural validation and inspect each reported issue.
- Apply lossless fixes first and retry generation when required.
- Review lossy fixes before pruning dangling child references.
- Validate the repaired specification again before rendering.
Использование «core». Explain what information the experimental evaluator receives.
Ожидаемый результат:
- The evaluator receives candidate descriptions, prompts, instructions, topology, and explicit context.
- Initial state and raw property or binding values are not sent automatically.
- Keep Gateway credentials on the server and exclude sensitive information from descriptions and context.
Аудит безопасности
БезопасноAll 88 static findings are false positives: Markdown delimiters, documented server-side authentication, and application-state terminology do not establish the reported threats. No evidence found of prompt injection, unauthorized credential disclosure, or shell execution in SKILL.md. Evaluator requests intentionally transmit described composition inputs; downstream package implementations were not included in this review.
Факторы риска
⚙️ Внешние команды (50)
🔑 Переменные окружения (3)
Поделиться и цитировать этот отчет
Делитесь версионным отчетом об оценке, нейтральным значком, встраиваемой карточкой и цитатами. Skillstore публикует доказательства, не решая, безопасен ли этот Skill.
Копировать ссылку на отчёт
https://skillstore.io/skills/vercel-labs-core/audits/1?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_reportЗначок Markdown
[](https://skillstore.io/skills/vercel-labs-core?utm_source=security_passport_badge)Значок HTML
<a href="https://skillstore.io/skills/vercel-labs-core?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/vercel-labs-core/security.svg" alt="Skillstore security assessment" loading="lazy"></a>Встраиваемая карточка
<iframe src="https://skillstore.io/embed/skills/vercel-labs-core.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>Академические ссылки (APA · BibTeX · CFF)
Цитата APA
vercel-labs. (2026). core security audit report (audit version 1) [Author version unspecified]. Skillstore. https://skillstore.io/skills/vercel-labs-core/audits/1Цитата BibTeX
@techreport{vercel-labs-vercel-labs-core-2026,
author = {vercel-labs},
title = {core security audit report (audit version 1)},
institution = {Skillstore},
year = {2026},
number = {1},
url = {https://skillstore.io/skills/vercel-labs-core/audits/1},
note = {Author version unspecified}
}CITATION.cff
cff-version: 1.2.0
message: "If you use this Skill, cite its author and this versioned security audit report."
title: "core security audit report (audit version 1)"
version: "unspecified"
type: report
authors:
- name: "vercel-labs"
date-released: "2026-09-29"
url: "https://skillstore.io/skills/vercel-labs-core/audits/1"
identifiers:
- type: other
value: "skillstore:vercel-labs-core:audit:1"
description: "Skillstore immutable audit report identifier"
Оценка Skillstore
Почему такая оценка Достоверность доказательств: СреднийЧто вы можете построить
Define a reusable component catalog
Frontend developers can structure component properties and catalog prompts for consistent AI-generated interfaces.
Integrate progressive specification generation
AI application engineers can plan JSONL streaming, refinement prompts, and validation before rendering generated specifications.
Connect forms to shared state
Design system maintainers can standardize form bindings, visibility rules, validation checks, and external state store adapters.
Попробуйте эти промпты
Explain schemas, catalogs, specifications, and SpecStream in json-render. Use a small contact form as the example and identify the packages I need.
Define a json-render schema and catalog for Button and TextInput components. Include Zod property constraints and show how catalog.prompt generates instructions.
Review my supplied form specification for state binding, visibility, and cross-field validation issues. Propose lossless repairs first and explain any changes requiring lossy fixes.
Design server-side experimental composition using my supplied catalog and atomic candidates. Include batch creation, sequential edits, evaluation budgets, cancellation, and partial snapshot handling. Identify evaluator data exposure and runtime action authorization requirements.
Лучшие практики
- Validate generated specifications and review lossy repairs before rendering; separately validate and authorize runtime actions.
- Keep evaluator credentials server-side and minimize sensitive data in descriptions, prompts, and explicit context.
- Pin experimental package versions when published, set evaluation limits, and supply an overall abort signal.
Избегать
- Treating schema validation as proof of semantic completeness or permission to execute actions.
- Using repeat, watch, computed, template, or conditional properties in the experimental composition subset.
- Assuming a catalog alone supplies configured candidate values, bindings, and rendering implementations.
Часто задаваемые вопросы
Which assistants can use this skill?
Does this skill install json-render or render an interface?
Are the experimental composition APIs stable?
What information leaves the application during Gateway evaluation?
What should I check when form values do not resolve?
Does specification validation secure action execution?
Сведения для разработчиков
Автор
vercel-labsЛицензия
MIT
Ревизия Skillstore
r1
Примечание о версии
Автор не указал версию.
Репозиторий
https://github.com/vercel-labs/json-render/tree/c2600d73908ed505e6d726f5b6f969ba8f597ce7/skills/coreСсылка
bad9dafc37d1638cd29cb9bab06d5f8dbcd0f6c2
Актуальность поддержки
30.09.2026
Использование
0 загрузок · 0 просмотров
Структура файлов
📄 SKILL.md