📦

convex-sentinel

Name: convex-sentinel
Author: get-convex

Ревизия содержимого r1 Безопасно ⚙️ Внешние команды

Capture Convex Production Errors with Sentinel

Production failures can be difficult to collect and investigate without exposing sensitive data. This skill guides Sentinel setup, redaction, monitoring, and optional automated triage.

Поддерживает: Claude Codex Code(CC)

🥉 77 Бронза

Установить с помощью моего Агента

Скопируйте этот запрос в своего Агента. Он содержит каноническую страницу Skill и манифест.

Запрос агента

Review the Skillstore skill "convex-sentinel" from https://skillstore.io/skills/get-convex-convex-sentinel.md and its manifest at https://skillstore.io/api/skills/get-convex-convex-sentinel/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.

Markdown-детали GET /skills/get-convex-convex-sentinel.md Подписанный manifest GET /api/skills/get-convex-convex-sentinel/manifest Подписанный lockfile GET /api/skills/get-convex-convex-sentinel/lockfile

Протестировать

Использование «convex-sentinel». Help me add Sentinel to an existing Convex and React application.

Ожидаемый результат:

Add the Sentinel component to the Convex application configuration.
Connect a React error boundary and browser error event handlers.
Confirm that write-time redaction remains enabled before deployment.

Использование «convex-sentinel». How should I inspect and respond to recent production errors?

Ожидаемый результат:

Use Convex CLI data access to review recent records, then subscribe to production error events for a controlled response workflow.

Использование «convex-sentinel». Can recurring errors create fix pull requests automatically?

Ожидаемый результат:

Enable triage only after classification rules are defined. Route recurring non-transient errors to ai-runner, and retain human review before merging changes.

Аудит безопасности

Безопасно

v1 • 21.09.2026 Открыть версионный отчет

All five static findings are false positives caused by Markdown inline-code delimiters in SKILL.md. The file describes configuration and CLI workflows but contains no executable Ruby or shell command mechanism, prompt injection, or malicious intent.

Просканировано файлов

Проанализировано строк

Пункты проверки

Ложные срабатывания проигнорированы

Факторы риска

⚙️ Внешние команды (5)

SKILL.md:10 SKILL.md:14 SKILL.md:15 SKILL.md:17 SKILL.md:18

Последний завершенный статический и семантический аудит не обнаружил подтвержденных проблем безопасности. Это не доказывает отсутствие побочных эффектов у навыка.

Аудитор:: codex

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.

Открыть версионный отчет

Копировать ссылку на отчёт

https://skillstore.io/skills/get-convex-convex-sentinel/audits/1?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_report

Значок Markdown

[![Skillstore security assessment](https://skillstore.io/badges/skills/get-convex-convex-sentinel/security.svg)](https://skillstore.io/skills/get-convex-convex-sentinel?utm_source=security_passport_badge)

Значок HTML

<a href="https://skillstore.io/skills/get-convex-convex-sentinel?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/get-convex-convex-sentinel/security.svg" alt="Skillstore security assessment" loading="lazy"></a>

Встраиваемая карточка

<iframe src="https://skillstore.io/embed/skills/get-convex-convex-sentinel.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

get-convex. (2026). convex-sentinel security audit report (audit version 1) [Author version unspecified]. Skillstore. https://skillstore.io/skills/get-convex-convex-sentinel/audits/1

Цитата BibTeX

@techreport{get-convex-get-convex-convex-sentinel-2026,
  author = {get-convex},
  title = {convex-sentinel security audit report (audit version 1)},
  institution = {Skillstore},
  year = {2026},
  number = {1},
  url = {https://skillstore.io/skills/get-convex-convex-sentinel/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: "convex-sentinel security audit report (audit version 1)"
version: "unspecified"
type: report
authors:
  - name: "get-convex"
date-released: "2026-09-21"
url: "https://skillstore.io/skills/get-convex-convex-sentinel/audits/1"
identifiers:
  - type: other
    value: "skillstore:get-convex-convex-sentinel:audit:1"
    description: "Skillstore immutable audit report identifier"

Оценка Skillstore

Почему такая оценка Достоверность доказательств: Средний

Архитектура

Сопровождаемость

Контент

Сообщество

Соответствие спецификации

Что вы можете построить

Add Production Monitoring

Configure Sentinel to collect server and client failures inside an existing Convex deployment.

Protect Diagnostic Data

Apply mandatory write-time redaction before production error records can reach an AI model provider.

Automate Error Triage

Plan recurring error classification and optional pull request creation for maintainable production operations.

Попробуйте эти промпты

Plan a Basic Installation

Guide me through adding Sentinel to my Convex application. Identify the configuration file changes and prerequisites before making edits.

Configure Client Capture

Add Sentinel client error capture to my React application. Include an error boundary, browser error events, promise rejection handling, and breadcrumbs.

Review Redaction Controls

Review my Sentinel error capture design for sensitive data exposure. Verify default-deny key filtering, value redaction, sampling, and storage limits.

Design Automated Triage

Design a Sentinel workflow that detects recurring non-transient errors, classifies them, and safely hands qualified issues to ai-runner for pull requests.

Лучшие практики

Keep redaction enabled and test it against representative secret names and values.
Apply sampling and storage caps that match production traffic and cost limits.
Review automated triage classifications before allowing pull request creation.

Избегать

Do not store raw secrets or unredacted request data in error records.
Do not send production error data to an undeclared third-party service.
Do not enable automated fixes without recurrence checks and human review.

Часто задаваемые вопросы

Where does Sentinel store captured errors?

It stores errors in a table within the user's own Convex deployment.

Does production data leave the Convex deployment?

The workflow says data stays in the deployment. Agent reads can reach the configured model provider, so redaction is mandatory.

Which errors can the client integration capture?

It covers React boundary failures, browser errors, unhandled promise rejections, and related breadcrumbs.

Is redaction optional?

No. The skill requires write-time redaction and describes it as enabled by default.

Can I inspect errors from the command line?

Yes. The workflow references Convex CLI data access and one-time queries for recent errors.

Does the skill automatically repair every error?

No. Self-healing is optional and targets recurring non-transient errors through a separate ai-runner workflow.

Сведения для разработчиков

Автор

get-convex

Лицензия

MIT

Ревизия Skillstore

Примечание о версии

Автор не указал версию.

Репозиторий

https://github.com/get-convex/agent-skills/tree/0aa10576821c6928f6a0f498c087af4ee231536e/skills/convex-sentinel

Ссылка

ee1e1d0465cbfd49d2dffacf50000ec9a8812bf0

Актуальность поддержки

21.09.2026

Использование

0 загрузок · 0 просмотров

Структура файлов

📄 SKILL.md