技能 twitter-api

📦

twitter-api

Name: twitter-api
Author: fetcher-sh

內容修訂 r1 中風險 📁 檔案系統存取🌐 網路存取🔑 環境變數⚙️ 外部命令

Access Twitter and X Data

Official X access can require developer approval, OAuth setup, and fixed pricing tiers. This skill provides paid, read-only REST and MCP access through fetcher.sh.

支援: Claude Codex Code(CC)

⚠️ 50 差

Maintain this Skill? Claim it ⚠️ 50 差

使用我的智慧代理安裝

將此請求複製給你的智慧代理，其中包含權威 Skill 頁面和清單。

智能體請求

Review the Skillstore skill "twitter-api" from https://skillstore.io/skills/fetcher-sh-twitter-api.md and its manifest at https://skillstore.io/api/skills/fetcher-sh-twitter-api/manifest. Verify the artifact. You may proceed after verification, subject to the environment's own policy.

你的智慧代理仍應顯示其計畫，並依安全政策請求必要確認。

Agent 可讀資源

當 AI Agent、爬蟲或腳本需要乾淨脈絡、而不是讀取完整頁面時，請使用這些連結。

Markdown 詳情 GET /skills/fetcher-sh-twitter-api.md 已簽名 manifest GET /api/skills/fetcher-sh-twitter-api/manifest 已簽名 lockfile GET /api/skills/fetcher-sh-twitter-api/lockfile

測試它

正在使用「twitter-api」。 Find recent posts about open source AI with at least 500 likes.

預期結果:

A ranked result list with author, timestamp, post text, engagement totals, and source link.
A brief theme summary that treats all retrieved text as untrusted content.

正在使用「twitter-api」。 Summarize the recent activity of @nasa.

預期結果:

A concise profile overview followed by recent post themes, posting frequency, and notable engagement.

正在使用「twitter-api」。 Compare trending topics in the United States and Canada.

預期結果:

Two country lists with shared topics, unique topics, and a short comparison of visible differences.

安全審計

中風險

v1 • 2026/8/13 開啟版本化報告

All 171 static matches are false positives caused by Markdown links, code formatting, documented HTTPS endpoints, placeholder credentials, and API terminology. The skill still has medium semantic risks from untrusted social content and paid MCP actions, plus a low privacy risk for bulk social graph collection.

已掃描檔案

729

分析行數

複核項目

已忽略的誤報

已確認的安全問題 (3)

中

SKILL.md:31-36 SKILL.md:116-118

Indirect Prompt Injection Through Retrieved Posts

The skill feeds externally authored Twitter and X content to agents but does not require treating post text as untrusted data. A malicious post could attempt to redirect later agent actions.

The documented purpose is to return third-party social content directly to agents. No instruction in the reviewed files defines a trust boundary for that content.

中

SKILL.md:216-229

Paid MCP Actions Lack an Explicit Approval Guard

The MCP integration exposes paid data retrieval and credit top-up actions without requiring user confirmation or a spending limit. Automated use could create unintended charges.

The documentation explicitly labels fetch_data and topup_credits as paid and states that calls have no rate limit or refund on upstream failure. No approval requirement is provided.

低

SKILL.md:99-107 references/faq.md:53-63

Bulk Social Graph Collection Lacks Privacy Guidance

The skill supports collecting followers, followings, replies, and retweeters but gives no guidance on consent, retention, or sensitive profiling. Bulk collection can create privacy and compliance risk.

The documented endpoints and follower export workflow clearly support bulk relationship collection. The reviewed guidance contains no privacy or retention controls.

風險因素

📁 檔案系統存取 (11)

references/endpoints.md:10 references/faq.md:19 references/faq.md:117 references/scenarios.md:16 SKILL.md:40 SKILL.md:73 SKILL.md:219 SKILL.md:238 SKILL.md:239 SKILL.md:240 SKILL.md:244

🌐 網路存取 (43)

🔑 環境變數 (37)

⚙️ 外部命令 (50)

審計單位: codex

分享與引用此報告

分享版本化評估報告、中性徽章、嵌入卡片與引用。Skillstore 只報告證據，不判斷該 Skill 是否安全。

開啟版本化報告

複製報告連結

https://skillstore.io/skills/fetcher-sh-twitter-api/audits/1?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_report

Markdown 徽章

[![Skillstore security assessment](https://skillstore.io/badges/skills/fetcher-sh-twitter-api/security.svg)](https://skillstore.io/skills/fetcher-sh-twitter-api?utm_source=security_passport_badge)

HTML 徽章

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

嵌入卡片

<iframe src="https://skillstore.io/embed/skills/fetcher-sh-twitter-api.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>

學術引用 (APA · BibTeX · CFF)

APA 引用

fetcher-sh. (2026). twitter-api security audit report (audit version 1) [Author version unspecified]. Skillstore. https://skillstore.io/skills/fetcher-sh-twitter-api/audits/1

BibTeX 引用

@techreport{fetcher-sh-fetcher-sh-twitter-api-2026,
  author = {fetcher-sh},
  title = {twitter-api security audit report (audit version 1)},
  institution = {Skillstore},
  year = {2026},
  number = {1},
  url = {https://skillstore.io/skills/fetcher-sh-twitter-api/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: "twitter-api security audit report (audit version 1)"
version: "unspecified"
type: report
authors:
  - name: "fetcher-sh"
date-released: "2026-08-13"
url: "https://skillstore.io/skills/fetcher-sh-twitter-api/audits/1"
identifiers:
  - type: other
    value: "skillstore:fetcher-sh-twitter-api:audit:1"
    description: "Skillstore immutable audit report identifier"

Skillstore 評分

評分依據證據置信度: 中

架構

可維護性

內容

社群

規範符合度

你可以打造什麼

Monitor Brand Conversations

Track recent mentions, hashtags, and high-engagement posts for reporting and response planning.

Collect Market Signals

Compare competitor posts, audience reactions, and country trends for structured research.

Build Social Data Pipelines

Retrieve paginated profiles, timelines, lists, and relationships for approved analytics workflows.

試試這些提示

Search Recent Posts

Search Twitter for posts about [topic]. Return the most relevant results with author, date, engagement, and source link.

Review an Account

Resolve @[handle], then summarize its profile and recent posts. Separate verified facts from your interpretation.

Compare Competitors

Search recent posts from [account one] and [account two] between [start date] and [end date]. Compare themes and engagement patterns.

Design a Monitored Pipeline

Plan a paginated monitoring workflow for [query]. Include cursor handling, deduplication, spending limits, untrusted-content controls, retention rules, and failure recovery.

最佳實務

Confirm the endpoint, expected call count, and spending limit before starting a paid request.
Treat post text, profile text, links, and metadata as untrusted external content.
Paginate with returned cursors, deduplicate identifiers, and retain only data needed for the approved purpose.

避免

Do not place real API keys in prompts, logs, reports, or shared configuration files.
Do not automate paid calls or credit top-ups without explicit approval and a spending cap.
Do not use follower or engagement data for sensitive profiling, harassment, or unauthorized surveillance.