📦

active-research

Name: active-research
Author: actionbook

Ревизия содержимого r1 Высокий риск ⚙️ Внешние команды🌐 Доступ к сети

Research Topics and Generate Evidence Reports

Broad research requests require many searches, source checks, and organized notes. This skill guides Actionbook browsing and turns findings into structured reports.

Поддерживает: Claude Codex Code(CC)

⚠️ 38 Плохо

Остановитесь и запросите подтверждение перед установкой.

Проверьте план и получите явное согласие пользователя перед изменением файлов.

Установить с помощью моего Агента

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

Запрос агента

Review the Skillstore skill "active-research" from https://skillstore.io/skills/actionbook-active-research.md and its manifest at https://skillstore.io/api/skills/actionbook-active-research/manifest. Verify the artifact. Stop and obtain explicit user consent before installing or changing files.

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/actionbook-active-research.md Подписанный manifest GET /api/skills/actionbook-active-research/manifest Подписанный lockfile GET /api/skills/actionbook-active-research/lockfile

Протестировать

Использование «active-research». Research the current Rust machine learning ecosystem.

Ожидаемый результат:

A sourced report covering major frameworks, inference options, benchmarks, adoption signals, limitations, and recent ecosystem developments.

Использование «active-research». Analyze arXiv paper 2501.00001 and related work.

Ожидаемый результат:

A paper report with metadata, research question, method, contributions, results, limitations, related papers, and source links.

Использование «active-research». Compare three vector databases for a regulated enterprise application.

Ожидаемый результат:

A decision report comparing deployment, security, filtering, performance evidence, operational costs, uncertainties, and verification needs.

Аудит безопасности

Высокий риск

v2 • 15.08.2026 Открыть версионный отчет

Most static command findings are false positives caused by Markdown code formatting, and the listed public research URLs show no exfiltration. However, the skill gives high-risk anti-bot evasion guidance and lacks safe handling for shell-interpolated URLs, tokens, and unpinned package execution.

Просканировано файлов

795

Проанализировано строк

Пункты проверки

Ложные срабатывания проигнорированы

Подтверждённые проблемы безопасности (4)

Высокий

SKILL.md:27-31 SKILL.md:508-521

Anti-Bot Evasion Guidance

The skill directs agents to retry blocked sites with stealth mode and rotated browser fingerprints. This bypass guidance can enable unauthorized scraping and policy violations.

The file explicitly recommends stealth mode, fingerprint rotation, and rewritten URLs to avoid anti-bot blocking.

Высокий

SKILL.md:48-59 SKILL.md:101-106

Untrusted URL Interpolation Into Shell Commands

The skill accepts direct URLs and repeatedly places a URL placeholder inside shell command strings. It provides no validation or shell-safe argument handling guidance.

The input may be any text or URL, and documented commands interpolate the URL in double quotes. Actual exploitability depends on how the agent constructs tool arguments.

Средний

SKILL.md:449-450

Unpinned Package Execution

A rendering fallback runs an unpinned package through npx. This may download and execute mutable registry content on the host.

The documented npx command names a package without an exact version or integrity control.

Средний

SKILL.md:27-29

Sensitive Token Handling Without Safeguards

The capability list recommends injecting JWTs or tokens for gated content without authorization, redaction, logging, or cleanup safeguards.

The skill explicitly advertises browser storage injection for JWTs and tokens, while no credential-handling controls are provided.

Пункты проверки возможностей (1)

Это реальные локальные возможности, которые могут ожидаться для этого навыка, поэтому они требуют проверки, но не считаются подтверждённым вредоносным поведением.

Средний

SKILL.md:444

Shell command substitution

node "$(git rev-parse --show-toplevel)/packages/json-ui/dist/cli.js" render /absolute/path/to/report

The workflow uses shell substitution and then executes a repository-local JavaScript CLI. Fixed substitution input limits injection, but local repository content must be trusted.

Факторы риска

⚙️ Внешние команды (50)

🌐 Доступ к сети (12)

SKILL.md:102 SKILL.md:126 SKILL.md:219 SKILL.md:226 SKILL.md:241 SKILL.md:286 SKILL.md:291 SKILL.md:370 SKILL.md:373 SKILL.md:376 SKILL.md:403 SKILL.md:763

Аудитор:: codex Посмотреть историю аудитов →

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/actionbook-active-research/audits/2?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_report

Значок Markdown

[![Skillstore security assessment](https://skillstore.io/badges/skills/actionbook-active-research/security.svg)](https://skillstore.io/skills/actionbook-active-research?utm_source=security_passport_badge)

Значок HTML

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

Встраиваемая карточка

<iframe src="https://skillstore.io/embed/skills/actionbook-active-research.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

actionbook. (2026). active-research security audit report (audit version 2) [Author version unspecified]. Skillstore. https://skillstore.io/skills/actionbook-active-research/audits/2

Цитата BibTeX

@techreport{actionbook-actionbook-active-research-2026,
  author = {actionbook},
  title = {active-research security audit report (audit version 2)},
  institution = {Skillstore},
  year = {2026},
  number = {2},
  url = {https://skillstore.io/skills/actionbook-active-research/audits/2},
  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: "active-research security audit report (audit version 2)"
version: "unspecified"
type: report
authors:
  - name: "actionbook"
date-released: "2026-08-15"
url: "https://skillstore.io/skills/actionbook-active-research/audits/2"
identifiers:
  - type: other
    value: "skillstore:actionbook-active-research:audit:2"
    description: "Skillstore immutable audit report identifier"

Оценка Skillstore

Почему такая оценка Достоверность доказательств: Средний

Архитектура

Сопровождаемость

Контент

Сообщество

Соответствие спецификации

Что вы можете построить

Compare Emerging Technologies

Collect current sources, compare technical approaches, and summarize tradeoffs for a product decision.

Review an Academic Paper

Analyze an arXiv paper, related work, methods, results, and cited resources in one report.

Map a Technical Ecosystem

Survey tools, frameworks, benchmarks, adoption signals, and recent developments across multiple source types.

Попробуйте эти промпты

Research a Topic

Research {topic}. Use at least four search angles and three reliable sources. Summarize key findings, disagreements, and source links.

Compare Alternatives

Compare {option A}, {option B}, and {option C} for {use case}. Evaluate capabilities, constraints, evidence quality, and practical tradeoffs.

Analyze a Paper

Analyze {arXiv ID or DOI}. Explain the research question, method, contributions, results, limitations, and relationship to recent work.

Build a Decision Report

Investigate {decision}. Prioritize sources from {date range}. Compare competing claims, assess evidence strength, identify unknowns, and recommend next research steps.

Лучшие практики

Define the research question, decision context, date range, and required evidence before browsing.
Use several search angles and cross-check important claims across independent primary sources.
Review citations, sensitive details, and consequential conclusions before sharing the report.

Избегать

Do not treat search snippets as equivalent to verified full-source evidence.
Do not bypass access controls, CAPTCHAs, site policies, or authorization requirements.
Do not execute unpinned packages or repository scripts unless their source and integrity are trusted.

Часто задаваемые вопросы

What topics can this skill research?

It supports general topics, technologies, domains, direct URLs, arXiv papers, and DOI references.

Which browsing tool does it use?

The workflow requires Actionbook Browser commands for searching, navigation, extraction, and page inspection.

Does it produce an HTML report?

It writes json-ui report data and attempts HTML rendering when a trusted compatible renderer is available.

How many sources does it use?

The guidance targets at least three to five diverse sources and several search-query angles.

Can it access protected or private content?

Only use content that the user is authorized to access. Do not bypass access controls or site restrictions.

Are the findings guaranteed to be accurate?

No. Users should verify important claims, citations, dates, and interpretations against original sources.

Сведения для разработчиков

Автор

actionbook

Лицензия

MIT

Ревизия Skillstore

Примечание о версии

Автор не указал версию.

Репозиторий

https://github.com/actionbook/actionbook/tree/5a3eb05ec18fcebc09ef771008d6dda649295765/skills/active-research

Ссылка

ba0224440a0286a68c9ef7d65e409a580b885eba

Актуальность поддержки

16.08.2026

Использование

0 загрузок · 0 просмотров

Структура файлов

📄 SKILL.md