热点爆款写作助手（SEO）

追踪全网热点，分析热门文章风格，模仿并超越生成爆款文章，附爆款理由和封面图，支持质量检查和SEO优化。适用于新媒体运营、内容营销、自媒体创作场景

Security Scan

功能总体与描述一致，但有实现/操作上的不一致与 sloppy 细节（如缺少热点采集实现、脚本含硬编码路径和内置示例），建议在信任并启用前确认数据源和运行方式。

Purpose & Capability

技能目标是“追踪全网热点并生成/优化爆款文章”，SKILL.md describes web 搜索（微博、知乎、抖音、百度等）和封面生成等完整流程 — 这 aligns with名称与描述。 但仓内没有任何网络采集/平台 API 的实现或依赖说明；该技能是 instruction-only（由 agent 用其浏览/网络能力去执行搜集），这本身可行但应该被明确（如何访问平台，是否需要 API key 或认证）。当前缺少说明可能导致 unexpected behavior if the agent attempts scraping or needs credentials.

Instruction Scope

SKILL.md tells the agent to perform broad, parallel web collection across many platforms and to run automated quality checks (it references scripts/check_prohibited_words.py). The included Python script is a local prohibited-words checker that: (a) contains a hard-coded file path (/workspace/projects/hotspot-article-generator/assets/article2_chip_breakthrough.txt) which does not appear in the manifest (so it falls back to an embedded example), and (b) exits with non-zero on failures. The script does not contact remote endpoints, but SKILL.md permits/encourages wide web scraping/searching — that gives the agent broad discretion to fetch external content. The combination of vague 'use network search tools' instructions + no explicit allowed/blocked endpoints is a scope creep / operational ambiguity risk.

Install Mechanism

No install spec; skill is instruction-only with one small helper script. This minimizes installation risk (nothing downloaded or auto-executed on install).

Credentials

The skill requests no environment variables, no credentials, and no config paths — appropriate for a content generation/quality-check skill. There are no declared secrets or unrelated credential requests.

Persistence & Privilege

Flags: always=false (default), autonomous invocation allowed (platform default). The skill does not request persistent presence or modify other skills. No additional privilege escalation indicators present.

What to consider before installing

What to check before installing or enabling: - Clarify how '热点采集' runs: SKILL.md expects the agent to pull data from many platforms but the package contains no scrapers or API integrations. Ask the author (or inspect how you'll invoke the skill) whether the agent will use its own web-browsing tool, third‑party APIs, or custom scrapers — and what endpoints will be contacted. - Review the prohibited-words script behaviour: scripts/check_prohibited_words.py uses a hard-coded workspace path (which doesn't exist in the manifest) and falls back to an embedded example article. That means in practice it may only analyze its built-in sample unless you modify it. It also calls sys.exit(1) when violations exist, which could cause automated flows to abort — confirm this is intended. - Confirm network/privacy expectations: because the instructions tell the agent to collect content from many public platforms, confirm whether any scraped content or generated drafts will be transmitted to external services you don't control (e.g., image-generation APIs or analytics endpoints). If the agent has browsing/network capability, review its outbound requests or run in a sandbox first. - Test with non-sensitive data and in a sandbox: run the skill locally or in an isolated environment using your own sample articles to see what files/paths are accessed and whether the agent attempts unexpected network calls. - If you plan to use platform APIs (Weibo, Douyin, Zhihu, etc.), provide credentials only when you understand why they're required and prefer creating scoped, limited credentials. The skill does not currently request any keys — if the implementation later asks for broad tokens, treat that as a red flag. If you need higher assurance, ask the publisher for: (1) a description of how hotspot collection is implemented (endpoints, frequency, rate limits), (2) a version of the prohibited-words checker that accepts input paths or stdin instead of a hard-coded file, and (3) a clear data flow diagram showing where generated content and scraped sources are stored or transmitted.

热点爆款文章生成器 v2.0

本技能用于从全网热点中快速产出高质量、可发布的爆款文章，适配公众号、小红书、知乎、百家号、头条号等平台。

核心流程概览

技能首先进行智能起点判断，根据用户输入自动选择执行模式：

• 快速模式：用户直接输入话题（如"成都户外打野"），跳过热点采集（S1），直接进入竞品分析与生成。
• 完整模式：用户请求"追热点"或"生成今日热点文章"，执行完整 S1-S4 流程。
• 文章优化模式：用户提供已有文章，仅执行质量检查与优化（S3-S4）。

极速执行流程（S1-S5）

┌─────────────────────────────────────────────────────────────┐
│                    智能起点判断                              │
│  ┌─────────────┐    ┌─────────────┐    ┌─────────────┐     │
│  │ 直接话题输入 │ →  │ 追热点需求  │ →  │ 文章优化    │     │
│  │   (跳过S1)   │    │ (完整流程)  │    │ (仅S4优化)  │     │
│  └──────┬──────┘    └──────┬──────┘    └──────┬──────┘     │
└─────────┼───────────────────┼─────────────────┼────────────┘
          │                   │                 │
          ↓                   ↓                 ↓
┌─────────────────────────────────────────────────────────────┐
│         S1: 并行采集（热点+竞品分析）                         │
│  ┌───────────────────┐    ┌───────────────────┐             │
│  │  🔥 热点搜索采集   │ ∥  │  📊 竞品文章分析  │             │
│  │  - 全网热搜采集    │    │  - 热门文章搜索   │             │
│  │  - 时效分级       │    │  - 风格提取      │             │
│  │  - 热度评估       │    │  - 成功要素      │             │
│  └───────────────────┘    └───────────────────┘             │
└─────────────────────────────────────────────────────────────┘
                              ∥
                              ↓
┌─────────────────────────────────────────────────────────────┐
│              S2: 智能生成（多篇并发）                         │
│  ┌───────────────┐ ┌───────────────┐ ┌───────────────┐      │
│  │  文章一       │ │  文章二       │ │  文章三       │      │
│  │  (情感共鸣型) │ │  (实用干货型) │ │  (观点评论型) │      │
│  └───────────────┘ └───────────────┘ └───────────────┘      │
└─────────────────────────────────────────────────────────────┘
                              │
                              ↓
┌─────────────────────────────────────────────────────────────┐
│              S3: 质量控制（自动化）                          │
│  ┌───────────────┐ ┌───────────────┐ ┌───────────────┐      │
│  │  ✅ 违禁词检查  │ →  │  📝 真实性验证  │ →  │  ⭐ 质量评分  │      │
│  └───────────────┘ └───────────────┘ └───────────────┘      │
│         │                                   │              │
│         └─────────── 不合格 → 修复 ──────────┘              │
└─────────────────────────────────────────────────────────────┘
                              │
                              ↓
┌─────────────────────────────────────────────────────────────┐
│              S4: 智能输出（按需生成）                         │
│  ┌───────────────┐ ┌───────────────┐ ┌───────────────┐      │
│  │  📄 文章输出   │ │  🖼️ 封面图    │ │  📋 发布建议  │      │
│  │  (差异化多篇) │ │  (S5 可选)    │ │  (含SEO)     │      │
│  └───────────────┘ └───────────────┘ └───────────────┘      │
└─────────────────────────────────────────────────────────────┘

在执行过程中，技能会自动进行违禁词检查、事实真实性核验、质量评分与 SEO 优化，确保内容不仅"能爆"，还"可信、合规、可持续复用"。

详细执行规则、参数与模板见：references/core_workflow.md。