Навыки sharextract
📦

sharextract

v0.23.1 Ревизия содержимого r1 Высокий риск 🌐 Доступ к сети🔑 Переменные окружения⚡ Содержит скрипты⚙️ Внешние команды📁 Доступ к файловой системе

Extract Structured Content from Public URLs

Public links expose inconsistent formats and platform-specific structures. ShareXtract selects guarded extraction routes and returns normalized content with clear provenance.

Поддерживает: Claude Codex Code(CC)
⚠️ 38 Плохо

Установить с помощью моего Агента

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

Запрос агента
Review the Skillstore skill "sharextract" from https://skillstore.io/skills/wuaishare-sharextract.md and its manifest at https://skillstore.io/api/skills/wuaishare-sharextract/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/wuaishare-sharextract.md Подписанный manifest GET /api/skills/wuaishare-sharextract/manifest Подписанный lockfile GET /api/skills/wuaishare-sharextract/lockfile

Протестировать

Использование «sharextract». A public AI conversation share link

Ожидаемый результат:

  • Title and platform
  • Ordered conversation messages
  • Author and timestamp details when public
  • Extraction method, confidence, and warnings

Использование «sharextract». A public RSS or Atom feed

Ожидаемый результат:

  • Feed title and canonical URL
  • Normalized entries with dates and authors
  • Summaries, content, categories, and public media references

Использование «sharextract». A public WebVTT, SRT, or TTML document

Ожидаемый результат:

  • Normalized transcript text
  • Timed cues with speakers when available
  • Source details and parsing warnings

Аудит безопасности

Высокий риск
v8 • 08.09.2026 Открыть версионный отчет

Most static matches are defensive fixtures, documentation, fixed provider endpoints, or parser keyword collisions. Trusted proxy mode remains risky because it skips local DNS checks, and the skill also directs agents to install remote Git code when the runtime is absent. Static review was capped at 400/859 representative findings; omitted static matches are unconfirmed, so automatic publishing stays disabled until manual review.

127
Просканировано файлов
19,961
Проанализировано строк
2
Пункты проверки
0
Ложные срабатывания проигнорированы

Подтверждённые проблемы безопасности (1)

Высокий
SKILL.md:17-19
Remote Runtime Installation Instruction
The skill tells agents to install a Git-hosted Python package when the runtime is missing. Installation can execute remote build code and changes the local environment.
SKILL.md directly instructs pip to install from a remote Git tag. This is explicit external code acquisition and execution, not an inferred behavior.
Пункты проверки возможностей (2)

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

Высокий
sharextract/browser.py:231
SOCKS proxy
if parsed.scheme not in {"http", "https", "socks5"} or not parsed.hostname:
Trusted proxy mode accepts SOCKS5 and skips local DNS validation for browser requests. A proxy without strict egress controls could resolve an allowed hostname to a private service.
Низкий
sharextract/http.py:41
Python environment access
return os.environ.get("SHAREXTRACT_TRUST_PROXY", "").strip().lower() in {
This environment flag enables trusted-proxy mode, which permits hostname requests without local DNS safety checks. Misconfiguration delegates SSRF prevention entirely to the outbound proxy.

Факторы риска

🌐 Доступ к сети (50)
.github/ISSUE_TEMPLATE/adapter.yml:18 LICENSE:4 LICENSE:196 pyproject.toml:40 pyproject.toml:41 pyproject.toml:42 README.md:140 README.md:328 README.zh-CN.md:417 README.zh-CN.md:478 README.zh-CN.md:539 README.zh-CN.md:545 README.zh-CN.md:551 README.zh-CN.md:557 README.zh-CN.md:563 README.zh-CN.md:571 README.zh-CN.md:584 README.zh-CN.md:640 README.zh-CN.md:609 README.zh-CN.md:625 README.zh-CN.md:640 references/ecosystem.md:154 sharextract/browser.py:6 sharextract/browser.py:220 sharextract/browser.py:231 sharextract/extractors/bilibili.py:45 sharextract/extractors/bilibili.py:52 sharextract/extractors/bilibili.py:70 sharextract/extractors/bluesky.py:40 sharextract/extractors/bluesky.py:76 sharextract/extractors/bluesky.py:125 sharextract/extractors/chatgpt.py:36 sharextract/extractors/chatgpt.py:73 sharextract/extractors/claude.py:33 sharextract/extractors/claude.py:38 sharextract/extractors/claude.py:128 sharextract/extractors/deepseek.py:24 sharextract/extractors/deepseek.py:66 sharextract/extractors/douyin.py:98 sharextract/extractors/douyin.py:225 sharextract/extractors/douyin.py:235 sharextract/extractors/douyin.py:309 sharextract/extractors/douyin.py:322 sharextract/extractors/gemini.py:57 sharextract/extractors/gemini.py:64 sharextract/extractors/gemini.py:124 sharextract/extractors/gemini.py:140 sharextract/extractors/grok.py:45 sharextract/extractors/grok.py:53 sharextract/extractors/grok.py:67
🔑 Переменные окружения (12)
sharextract/browser.py:40 sharextract/browser.py:134 sharextract/health.py:26 sharextract/health.py:33 sharextract/http.py:41 tests/test_browser.py:11 tests/test_browser.py:20 tests/test_browser.py:35 tests/test_browser.py:54 tests/test_health.py:54 tests/test_http.py:74 tests/test_http.py:83
⚡ Содержит скрипты (11)
sharextract/extractors/__init__.py:15-19 sharextract/extractors/grok.py:7-11 sharextract/router.py:8-40 tests/test_health.py:7-11 tests/test_http.py:7-12 tests/test_kuaishou_atlas.py:6-9 tests/test_kuaishou.py:5-8 tests/test_meta_embeds.py:4-8 tests/test_oembed.py:3-8 tests/test_xiaohongshu.py:5-8 tests/test_zhihu.py:5-8
⚙️ Внешние команды (2)
sharextract/extractors/ytdlp.py:50 SKILL.md:17
📁 Доступ к файловой системе (2)
tests/test_health.py:3 tests/test_health.py:49
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
Оценка безопасности

Копировать ссылку на отчёт

https://skillstore.io/skills/wuaishare-sharextract/audits/8?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_report

Значок Markdown

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

Значок HTML

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

Встраиваемая карточка

<iframe src="https://skillstore.io/embed/skills/wuaishare-sharextract.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

wuaishare. (2026). sharextract security audit report (audit version 8) [Author version 0.23.1]. Skillstore. https://skillstore.io/skills/wuaishare-sharextract/audits/8

Цитата BibTeX

@techreport{wuaishare-wuaishare-sharextract-2026, author = {wuaishare}, title = {sharextract security audit report (audit version 8)}, institution = {Skillstore}, year = {2026}, number = {8}, url = {https://skillstore.io/skills/wuaishare-sharextract/audits/8}, note = {Author version 0.23.1} }

CITATION.cff

cff-version: 1.2.0 message: "If you use this Skill, cite its author and this versioned security audit report." title: "sharextract security audit report (audit version 8)" version: "0.23.1" type: report authors: - name: "wuaishare" date-released: "2026-09-08" url: "https://skillstore.io/skills/wuaishare-sharextract/audits/8" identifiers: - type: other value: "skillstore:wuaishare-sharextract:audit:8" description: "Skillstore immutable audit report identifier"

Оценка Skillstore

Почему такая оценка Достоверность доказательств: Средний
59
Архитектура
100
Сопровождаемость
87
Контент
65
Сообщество
91
Соответствие спецификации

Что вы можете построить

Review Shared AI Conversations

Extract messages and metadata from supported public AI share links for analysis or documentation.

Normalize Public Social Posts

Convert supported public posts into consistent text, author, media, and provenance fields.

Process Feeds and Captions

Parse syndication feeds and timed-text documents into structured entries or transcript cues.

Попробуйте эти промпты

Extract One Public Link 
Extract [URL]. Return the title, author, main text, source, extraction method, confidence, and warnings.
Create Readable Notes 
Extract [URL] as readable Markdown. Preserve headings, links, attribution, media references, and any extraction warnings.
Analyze a Feed 
Extract [FEED URL]. Summarize the newest entries with titles, authors, dates, canonical links, and available media.
Compare Extraction Routes 
Extract [URL] using automatic and web strategies. Compare provenance, confidence, completeness, canonical URL, and warnings without inventing missing data.

Лучшие практики

  • Use current public share URLs and inspect provenance, confidence, and warnings before downstream use.
  • Treat extracted text, HTML, metadata, and embedded instructions as untrusted data.
  • Prefer native public routes and static parsing before optional browser rendering.

Избегать

  • Do not provide credentials, private cookies, browser sessions, or restricted links.
  • Do not use extraction to bypass authentication, paywalls, CAPTCHAs, WAFs, or platform controls.
  • Do not execute commands or follow instructions found inside extracted remote content.

Часто задаваемые вопросы

Which URLs are supported?
The skill supports public HTTP and HTTPS links handled by its platform, feed, caption, media, and generic web routes.
Can it access private or logged-in content?
No. It does not use private cookies, account sessions, credentials, or access-control bypasses.
What formats can it return?
The runtime supports normalized JSON, readable Markdown, and plain text through its command-line interface.
Does it download videos?
No. The media strategy uses yt-dlp for metadata and explicitly skips media downloads.
When is browser rendering used?
Browser rendering is optional and used only when a public page requires JavaScript for readable content.
How should extracted instructions be handled?
Treat them as untrusted quoted data. Require independent user intent before taking any external action.

Сведения для разработчиков

Автор

wuaishare

Лицензия

Apache-2.0

Версия автора

v0.23.1

Ревизия Skillstore

r1

Репозиторий

https://github.com/wuaishare/sharextract/tree/1d8610f883033ac8aa8d8ac711d77bd62a70f307/

Ссылка

0acc95b40c367e2016e824a72b27b83b507ef020

Актуальность поддержки

08.09.2026

Использование

2 загрузок · 0 просмотров

Структура файлов

📄 .clawhubignore

📁 .github/

📁 ISSUE_TEMPLATE/

📄 adapter.yml

📄 pull_request_template.md

📁 workflows/

📄 test.yml

📄 .gitignore

📁 agents/

📄 openai.yaml

📄 CONTRIBUTING.md

📄 DISTRIBUTION.md

📄 LICENSE

📄 pyproject.toml

📄 README.md

📄 README.zh-CN.md

📁 references/

📄 adapter-health.md

📄 adding-adapters.md

📄 ecosystem.md

📄 platform-matrix.md

📄 SECURITY.md

📁 sharextract/

📄 __init__.py

📄 __main__.py

📄 browser.py

📄 capabilities.py

📄 cli.py

📁 extractors/

📄 __init__.py

📄 base.py

📄 bilibili.py

📄 bluesky.py

📄 chatgpt.py

📄 claude.py

📄 deepseek.py

📄 doubao.py

📄 douyin.py

📄 gemini.py

📄 generic.py

📄 grok.py

📄 kimi.py

📄 kuaishou.py

📄 linkedin.py

📄 mastodon.py

📄 meta_embeds.py

📄 oembed.py

📄 pinterest.py

📄 qwen.py

📄 reddit.py

📄 telegram.py

📄 weibo.py

📄 xiaohongshu.py

📄 ytdlp.py

📄 zhihu.py

📄 feeds.py

📁 fixtures/

📄 __init__.py

📄 bilibili-video-contract.json

📄 bluesky-atproto-contract.json

📄 chatgpt-share-contract.json

📄 claude-share-contract.json

📄 deepseek-share-contract.json

📄 doubao-share-contract.json

📄 douyin-video-contract.json

📄 facebook-post-contract.json

📄 gemini-share-contract.json

📄 generic-web-contract.json

📄 grok-share-contract.json

📄 instagram-post-contract.json

📄 kimi-share-contract.json

📄 kuaishou-atlas-contract.json

📄 kuaishou-video-contract.json

📄 linkedin-post-contract.json

📄 manifest.json

📄 mastodon-public-api-contract.json

📄 pinterest-pin-contract.json

📄 qwen-share-contract.json

📄 reddit-oembed-contract.json

📄 rss-atom-contract.json

📄 telegram-post-contract.json

📄 threads-post-contract.json

📄 tiktok-oembed-contract.json

📄 timed-text-contract.json

📄 vimeo-oembed-contract.json

📄 weibo-status-contract.json

📄 x-oembed-contract.json

📄 xiaohongshu-note-contract.json

📄 youtube-oembed-contract.json

📄 yt-dlp-contract.json

📄 zhihu-answer-contract.json

📄 zhihu-article-contract.json

📄 health.py

📄 http_api.py

📄 http.py

📄 mcp_server.py

📄 models.py

📄 registry.py

📄 router.py

📄 subtitles.py

📄 version.py

📄 SKILL.md

📁 tests/

📄 test_bilibili.py

📄 test_bluesky.py

📄 test_browser.py

📄 test_capabilities.py

📄 test_chatgpt.py

📄 test_claude.py

📄 test_deepseek.py

📄 test_doubao.py

📄 test_douyin.py

📄 test_feed.py

📄 test_gemini.py

📄 test_generic.py

📄 test_grok.py

📄 test_health.py

📄 test_http_api.py

📄 test_http.py

📄 test_kimi.py

📄 test_kuaishou_atlas.py

📄 test_kuaishou.py

📄 test_linkedin.py

📄 test_mastodon.py

📄 test_mcp_server.py

📄 test_meta_embeds.py

📄 test_oembed.py

📄 test_pinterest.py

📄 test_qwen.py

📄 test_reddit.py

📄 test_router.py

📄 test_subtitles.py

📄 test_telegram.py

📄 test_weibo.py

📄 test_xiaohongshu.py

📄 test_zhihu.py