Ai Intelligence Site

AI 情报中心网站自动化管理。创建、更新、部署 AI 竞品分析网站到 GitHub Pages。 支持 5 个子站：全球站、中文站、Skills站、资讯站、模型站。 每日自动更新：删除旧数据 → Serper 搜索发现网站 → 获取真实流量 → 推送 GitHub。 触发词：AI情报中心、竞品分析网站、AI聚合网...

Security Scan

The skill mostly does what it says (search + gather traffic + git push), but there are coherence issues — notably a hardcoded Serper API key default and mismatches between declared requirements and the SKILL.md — which merit caution before installing or running.

Purpose & Capability

The code implements the advertised functionality (search via Serper, parse results, compute tiers, commit & push). However the registry metadata lists no required environment variables while SKILL.md and the script require a SERPER_API_KEY. The script also includes a hardcoded Serper API key default, which is unexpected and inconsistent with the stated requirement to provide your own key.

Instruction Scope

SKILL.md instructs running the provided daily_update.py and configuring a GitHub repo, which matches the script. The script performs network calls to Serper and runs git commands (git add ., commit, push). Running it will stage and push files from the script's working directory — if invoked in the wrong repository or path this could commit unintended files. The instructions do not warn about this risk.

Install Mechanism

This is an instruction-only skill with no install spec. No additional packages are installed by the skill itself, so there is no download/install risk from external URLs.

Credentials

Only one API credential (SERPER_API_KEY) is relevant and expected for the stated purpose, so requiring it is proportionate — but the registry metadata does not declare it and the script falls back to a hardcoded API key value. Embedding a default API key in code is suspicious (possible leaked key or developer convenience) and may cause unexpected API usage or credential exposure.

Persistence & Privilege

The skill does not request always:true and does not modify other skills or system-wide settings. It does run git to push data to a remote repo when executed, which is expected for a deployment workflow but requires the user to configure their Git remote appropriately.

Scan Findings in Context

[hardcoded-serper-key] unexpected: The script defines SERPER_API_KEY with a hardcoded value ('374959ea28cae888d8049ea2e34d8acc156c602b') used as a default if the environment variable isn't set. Even though the skill needs a SERPER_API_KEY, embedding a usable key in code is unexpected and risky.

What to consider before installing

This skill implements search + traffic scraping + git push as advertised, but take these precautions before installing or running it: - Do not rely on the hardcoded SERPER_API_KEY. Set SERPER_API_KEY in your environment to your own key and do not use the embedded default. Treat the embedded key as potentially leaked and avoid sending sensitive queries through it. - Review the repo/work directory: the script runs 'git add .', 'git commit', and 'git push' from its working directory. Make sure you run it inside a dedicated repository or directory so you don't accidentally stage/push unrelated or sensitive files. - Verify the Git remote and credentials (it will push to whatever remote is configured). Prefer creating a dedicated empty repo for outputs. - Because the registry metadata doesn't declare the required env var declared in SKILL.md, confirm with the skill author or inspect the code for additional hidden behavior before granting runtime access. - If you don't trust the embedded key, remove it and supply your own, and consider rotating any keys that may have been exposed.

AI 情报中心网站管理

自动化管理 AI 竞品分析网站，部署到 GitHub Pages。

网站结构

competitor-site/
├── index.html          # 全球站（AI 聚合网站）
├── data.json           # 全球站数据
├── cn/                 # 中文站
├── skills/             # Skills & MCP 站
├── news/               # AI 资讯站
├── models/             # AI 模型站
└── daily_update.py     # 每日更新脚本

核心功能

1. 每日全量更新

执行脚本：

python3 scripts/daily_update.py

更新流程：

1. 删除旧数据，从零开始
2. 通过 Serper 搜索发现新网站
3. 获取每个网站的真实流量数据
4. 无数据则跳过（不使用模拟）
5. 重新计算层级（T1/T2/T3）和趋势
6. 推送到 GitHub

2. 数据源配置

每个站点有独立的搜索关键词和已知网站列表：

站点	搜索关键词示例
全球	AI tools directory, AI aggregator
中文	中国AI工具导航, AI工具集
Skills	MCP server marketplace, Claude skills
资讯	AI news website, AI newsletter
模型	LLM API provider, AI model platform

3. 数据结构

{
  "lastUpdated": "ISO时间戳",
  "competitors": [
    {
      "id": "唯一标识",
      "name": "网站名称",
      "domain": "域名",
      "monthlyVisits": 1000000,
      "tier": "T1/T2/T3",
      "trafficTrend": [历史流量数组]
    }
  ],
  "insights": [{"title": "", "content": "", "color": ""}]
}

4. 层级划分

• T1: 月访问 ≥ 100万
• T2: 月访问 10万-100万
• T3: 月访问 < 10万

环境变量

• SERPER_API_KEY: Serper 搜索 API Key（必需）

部署

1. 创建 GitHub 仓库并启用 Pages
2. 配置 Git remote
3. 运行 daily_update.py 初始化数据
4. 设置定时任务每日更新