Навыки xhs-content-research-assistant

📦

xhs-content-research-assistant

Name: xhs-content-research-assistant
Author: devinchen2014

Ревизия содержимого r1 Средний риск ⚙️ Внешние команды🌐 Доступ к сети🔑 Переменные окружения

Research Xiaohongshu Content and Audience Feedback

Content teams need evidence about topics, audience feedback, and creator activity before planning posts. This skill organizes SocialDataX retrieval workflows into focused Xiaohongshu research.

Поддерживает: Claude Codex Code(CC)

📊 63 Адекватно

Maintain this Skill? Claim it 📊 63 Адекватно

Установить с помощью моего Агента

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

Запрос агента

Review the Skillstore skill "xhs-content-research-assistant" from https://skillstore.io/skills/devinchen2014-xhs-content-research-assistant.md and its manifest at https://skillstore.io/api/skills/devinchen2014-xhs-content-research-assistant/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/devinchen2014-xhs-content-research-assistant.md Подписанный manifest GET /api/skills/devinchen2014-xhs-content-research-assistant/manifest Подписанный lockfile GET /api/skills/devinchen2014-xhs-content-research-assistant/lockfile

Протестировать

Использование «xhs-content-research-assistant». Illustrative request: Plan topics from a one-page search for small apartment storage.

Ожидаемый результат:

Illustrative format only; no live platform data was retrieved.
Observed evidence: List returned titles, authors, available engagement signals, and note IDs.
Topic patterns: Group the sampled posts by storage location and audience question.
Proposed angles: Explain which visible examples support each suggested topic.
Limit: One page is a sample, not evidence of platform-wide demand.

Использование «xhs-content-research-assistant». Illustrative request: Summarize audience questions from a selected post and its replies.

Ожидаемый результат:

Illustrative format only; no comments were retrieved.
Post facts: Record the returned author, publication time, and available engagement metrics.
Comment themes: Group visible questions and cite supporting comment evidence.
Coverage: State whether replies are complete, partial, or not yet expanded.
Next step: Continue only with the matching returned token and remaining retrieval allowance.

Аудит безопасности

Средний риск

v1 • 02.11.2026 Открыть версионный отчет

All 59 static findings are false positives involving Markdown, documented service configuration, platform identifiers, or readable Chinese text. Two semantic risks remain: automatic execution of an unpinned npm package and mandatory disclosure of token-bearing note URLs. No evidence found of malicious payloads, credential theft, or audit-directed prompt injection in the reviewed files; external package behavior was not inspected.

Просканировано файлов

172

Проанализировано строк

Пункты проверки

Ложные срабатывания проигнорированы

Подтверждённые проблемы безопасности (2)

Средний

SKILL.md:31-54 SKILL.md:59-81 SKILL.md:141

Automatic Execution of an Unpinned npm Dependency

Examples use npx -y socialdatax-skills@latest to retrieve and execute a mutable external package. Unreviewed updates could execute with the user's permissions and service credential.

Every CLI example uses @latest and -y, while runtime authentication requires an environment credential. The trust exposure is explicit; no evidence found of package compromise.

Средний

SKILL.md:113

Mandatory Disclosure of Token-Bearing Note URLs

The output rule requires retaining xsec_token in final answers, citations, storage, and forwarding, and prohibits redaction. This can expose potentially sensitive link tokens beyond their retrieval context.

The instruction explicitly requires token inclusion across output destinations without a sensitivity check. The files do not establish the token's privileges or lifetime.

Факторы риска

⚙️ Внешние команды (39)

🌐 Доступ к сети (3)

SKILL.md:7 SKILL.md:25 SKILL.md:153

🔑 Переменные окружения (6)

SKILL.md:7 SKILL.md:25 SKILL.md:141 SKILL.md:152 SKILL.md:154 SKILL.md:161

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/devinchen2014-xhs-content-research-assistant/audits/1?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_report

Значок Markdown

[![Skillstore security assessment](https://skillstore.io/badges/skills/devinchen2014-xhs-content-research-assistant/security.svg)](https://skillstore.io/skills/devinchen2014-xhs-content-research-assistant?utm_source=security_passport_badge)

Значок HTML

<a href="https://skillstore.io/skills/devinchen2014-xhs-content-research-assistant?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/devinchen2014-xhs-content-research-assistant/security.svg" alt="Skillstore security assessment" loading="lazy"></a>

Встраиваемая карточка

<iframe src="https://skillstore.io/embed/skills/devinchen2014-xhs-content-research-assistant.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

devinchen2014. (2026). xhs-content-research-assistant security audit report (audit version 1) [Author version unspecified]. Skillstore. https://skillstore.io/skills/devinchen2014-xhs-content-research-assistant/audits/1

Цитата BibTeX

@techreport{devinchen2014-devinchen2014-xhs-content-research-assistant-2026,
  author = {devinchen2014},
  title = {xhs-content-research-assistant security audit report (audit version 1)},
  institution = {Skillstore},
  year = {2026},
  number = {1},
  url = {https://skillstore.io/skills/devinchen2014-xhs-content-research-assistant/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: "xhs-content-research-assistant security audit report (audit version 1)"
version: "unspecified"
type: report
authors:
  - name: "devinchen2014"
date-released: "2026-11-02"
url: "https://skillstore.io/skills/devinchen2014-xhs-content-research-assistant/audits/1"
identifiers:
  - type: other
    value: "skillstore:devinchen2014-xhs-content-research-assistant: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

Plan Evidence-Based Content Topics

Compare hot-search signals and keyword samples to identify content angles, supporting examples, and unanswered audience questions.

Review Competitor Audience Feedback

Inspect relevant posts and comment threads to distinguish recurring product questions from isolated reactions.

Study Creator Positioning

Compare creator profile information with recent post evidence before choosing a content direction.

Попробуйте эти промпты

Explore a Topic

Search Xiaohongshu for [topic] using one page of results. Summarize relevant titles, authors, and three content angles. Separate observations from interpretations.

Analyze a Post and Its Comments

Retrieve details and one page of comments for [note ID or URL]. Summarize visible audience questions and recurring themes. State any missing data.

Compare Creator Evidence

Review [creator A] and [creator B] using their profiles and one page of posts each. Compare positioning, recurring topics, and available engagement signals. Cite evidence separately from judgments.

Build a Bounded Research Brief

Research [topic] for [audience] using hot-search signals, keyword results, and selected comments. Limit retrieval to [page limit] pages and [item limit] items. Preserve partial results and continue only within those limits. Report evidence, uncertainty, and proposed content angles. Do not expose credentials or sensitive URL tokens in shared outputs.

Лучшие практики

Start with one page and explicit item limits; preserve partial results before requesting more data.
Keep observations separate from interpretations, and state sample size and coverage gaps.
Use a reviewed, pinned CLI version and keep API keys and potentially sensitive URL tokens out of shared reports.

Избегать

Treating a small post sample or hot-search ranking as proof of platform-wide audience demand.
Restarting retrieval after partial failures without checking continuation tokens and remaining limits.
Executing unreviewed package updates or forwarding token-bearing URLs without checking authorization.

Часто задаваемые вопросы

What inputs can I provide?

Provide keywords, a research topic, post links or IDs, creator profile links, account share text, or platform account IDs.

Do I need a SocialDataX API key?

Yes. Configure SOCIALDATAX_API_KEY through your environment. Obtain or manage the key through the declared SocialDataX website.

Does this skill publish or interact with posts?

No. The documented workflow retrieves information and does not log in, post, like, comment, or change accounts.

Which AI tools are supported?

The report lists Claude, Codex, and Claude Code. Live retrieval still requires usable CLI execution or the documented MCP tools.

What happens when results are incomplete?

Keep retrieved items and report the error. Continue with the matching returned token only when limits allow; do not automatically restart.

Are there costs or installation concerns?

Service credits may be required. CLI examples use an unpinned npm package; review and pin it before execution.

Сведения для разработчиков

Автор

devinchen2014

Лицензия

Неизвестно

Ревизия Skillstore

Примечание о версии

Автор не указал версию.

Репозиторий

https://github.com/devinchen2014/socialdatax-skills/tree/c14b6adea4ba2b551557115cdfffde4c17375cbf/modelscope-skills/xhs-content-research-assistant

Ссылка

2d821629d9cf6def5fbc40718bb2420c989b2ca6

Актуальность поддержки

02.11.2026

Использование

0 загрузок · 0 просмотров

Структура файлов

📁 agents/

📄 openai.yaml

📄 SKILL.md