Douyin Ai Feed

Name: Douyin Ai Feed
Author: redfox-data

redfox-data/redfox-community

205 installs
316 repo stars
Updated August 4, 2026
redfox-data/redfox-community

Helps with ai & agent building tasks during AI-assisted development.

About

douyin-ai-feed is a Claude Code skill for ai & agent building. It helps solo builders move faster with AI-assisted coding.

douyin-ai-feed
AI & Agent Building
AI-coding skill

Douyin Ai Feed by the numbers

205 all-time installs (skills.sh)
+14 installs in the week ending Aug 4, 2026 (Skillselion tracking)
Ranked #2,857 of 16,546 AI & Agent Building skills by installs in the Skillselion catalog
Data as of Aug 5, 2026 (Skillselion catalog sync)

npx skills add https://github.com/redfox-data/redfox-community --skill douyin-ai-feed

Add your badge

Show developers this skill is listed on Skillselion. Paste this into your README.

[![Listed on Skillselion](https://skillselion.com/badge/skills/redfox-data/redfox-community/douyin-ai-feed.svg)](https://skillselion.com/skills/redfox-data/redfox-community/douyin-ai-feed)

Installs	205
repo stars	★ 316
Last updated	August 4, 2026
Repository	redfox-data/redfox-community ↗

What it does

Helps with ai & agent building tasks during AI-assisted development.

Files

SKILL.mdMarkdownGitHub ↗

AI抖音信息源

每日自动扫描抖音 AI 相关作品，按互动量（点赞/分享/评论）筛选最火内容，智能聚类后生成精美 HTML 日报。

---

能力概述

爆款发现：按关键词搜索抖音 AI 作品，按互动量筛选最火内容
智能聚类：自动从当天内容中发现话题方向，每天的分类由内容决定
终端表格：分类 + 标题 + 作者 + 点赞/分享/评论数，一目了然
可视化日报：深色主题 HTML（抖音青色风格），封面图、互动数据、作品直链、日期导航
一键订阅：--subscribe 开启每日自动产出，日报自动攒在本地

---

使用方式

# 生成今日爆款日报（默认关键词：AI）
python3 "$SKILL_PATH/scripts/fetch_douyin_ai.py"

# 自定义关键词
python3 "$SKILL_PATH/scripts/fetch_douyin_ai.py" --keyword "ChatGPT"

# 查看历史某天
python3 "$SKILL_PATH/scripts/fetch_douyin_ai.py" --date 2026-06-09

# 指定时间范围（开始时间含，结束时间不含）
python3 "$SKILL_PATH/scripts/fetch_douyin_ai.py" --start-time 2026-06-09 --end-time 2026-06-10

# 订阅 / 取消订阅
python3 "$SKILL_PATH/scripts/fetch_douyin_ai.py" --subscribe
python3 "$SKILL_PATH/scripts/fetch_douyin_ai.py" --unsubscribe

生成的 HTML 日报保存在 ~/Downloads/QoderReports/，自动浏览器打开。终端同步输出分类作品表格。

---

🔑 鉴权

获取 API Key

前往 redfox.hk/login 注册获取个人 API Key。

配置 API Key

方式	命令
环境变量（推荐）	`export REDFOX_API_KEY=ak_你的密钥`
写入 shell 配置	`echo 'export REDFOX_API_KEY=ak_你的密钥' >> ~/.zshrc && source ~/.zshrc`

脚本缺失环境变量时会明确报错并提供配置指引，不会静默失败。

---

参数说明

参数	说明	默认值
`--keyword`	搜索关键词	`AI`
`--page-size`	每页条数	`50`
`--date`	指定日期 YYYY-MM-DD	今天
`--start-time`	开始时间 YYYY-MM-DD（含）	根据 --date 自动计算
`--end-time`	结束时间 YYYY-MM-DD（不含）	根据 --date 自动计算
`--output-dir`	输出目录	`~/Downloads/QoderReports`
`--subscribe`	安装每日定时任务 (16:00)	—
`--unsubscribe`	卸载定时任务	—
`--no-open`	不自动打开浏览器	—

---

依赖

pip3 install requests

---

常见问题

Q：日报里的分类是怎么来的？ A：完全由当天内容决定。从作品的 type/topic 标签中自动识别聚类，每天的热点方向不同。

Q：怎么看到更多作品？ A：用 --page-size 100 增大单次查询数量。

Q：订阅后日报存在哪？ A：默认 ~/Downloads/QoderReports/，文件名格式 抖音AI日报_2026-06-10.html。

Q：没有 API Key 怎么办？ A：前往 redfox.hk 注册获取，然后执行 export REDFOX_API_KEY=ak_你的密钥。

<!DOCTYPE html>
<html lang="zh-CN">
<head>
<meta charset="UTF-8">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<title>AI抖音信息源 - {{DATE}}</title>

<!-- Fonts -->
<link rel="preconnect" href="https://fonts.googleapis.com">
<link rel="preconnect" href="https://fonts.gstatic.com" crossorigin>
<link href="https://fonts.googleapis.com/css2?family=Archivo+Black&family=Space+Grotesk:wght@300;400;500;600;700&display=swap" rel="stylesheet">

<style>
:root {
    --bg-primary: #0f0f0f;
    --bg-secondary: #1a1a1a;
    --bg-card: #1e1e1e;
    --bg-card-hover: #252525;
    --text-primary: #f0ece6;
    --text-secondary: #9a9590;
    --text-muted: #666;
    --accent: #25F4EE;
    --accent-light: #5CF7F2;
    --accent-glow: rgba(37, 244, 238, 0.15);
    --accent-border: rgba(37, 244, 238, 0.3);
    --border: rgba(255, 255, 255, 0.06);

    --font-display: 'Archivo Black', Impact, sans-serif;
    --font-body: 'Space Grotesk', -apple-system, sans-serif;

    --title-size: clamp(1.8rem, 4vw, 3.5rem);
    --h2-size: clamp(1.2rem, 2.5vw, 1.8rem);
    --h3-size: clamp(1rem, 1.8vw, 1.3rem);
    --body-size: clamp(0.8rem, 1.2vw, 0.95rem);
    --small-size: clamp(0.7rem, 0.9vw, 0.8rem);

    --radius: 12px;
    --radius-sm: 8px;
    --ease-out-expo: cubic-bezier(0.16, 1, 0.3, 1);
}

* { margin: 0; padding: 0; box-sizing: border-box; }
html { scroll-behavior: smooth; }

body {
    font-family: var(--font-body);
    font-size: var(--body-size);
    background: var(--bg-primary);
    color: var(--text-primary);
    line-height: 1.6;
    min-height: 100vh;
    padding: clamp(1rem, 3vw, 3rem);
}

.report-header {
    text-align: center;
    padding: clamp(2rem, 5vw, 4rem) 0 clamp(1rem, 2vw, 1.5rem);
    max-width: 900px;
    margin: 0 auto;
}

.report-header h1 {
    font-family: var(--font-display);
    font-size: var(--title-size);
    color: var(--accent);
    letter-spacing: -0.02em;
    margin-bottom: 0.3rem;
}

.report-header .subtitle {
    font-size: clamp(0.9rem, 1.5vw, 1.1rem);
    color: var(--text-secondary);
    font-weight: 300;
    margin-bottom: 1.2rem;
}

.date-display {
    display: flex;
    align-items: center;
    justify-content: center;
    gap: 1rem;
    margin: 1.2rem 0;
}

.date-display .date-text {
    font-family: var(--font-display);
    font-size: clamp(1.4rem, 3vw, 2.4rem);
    color: var(--text-primary);
    letter-spacing: 0.02em;
}

.date-nav-btn {
    background: var(--bg-secondary);
    border: 1px solid var(--border);
    color: var(--text-secondary);
    width: 36px;
    height: 36px;
    border-radius: 50%;
    cursor: pointer;
    font-size: 1.1rem;
    display: flex;
    align-items: center;
    justify-content: center;
    transition: all 0.2s ease;
}

.date-nav-btn:hover {
    border-color: var(--accent-border);
    color: var(--accent);
    background: var(--accent-glow);
}

.stats-bar {
    display: flex;
    justify-content: center;
    gap: clamp(1.5rem, 4vw, 3rem);
    padding: clamp(1rem, 2vw, 1.5rem) clamp(1rem, 3vw, 2rem);
    background: var(--bg-secondary);
    border-radius: var(--radius);
    border: 1px solid var(--border);
    max-width: 700px;
    margin: clamp(0.5rem, 1vw, 1rem) auto clamp(2rem, 4vw, 3rem);
    flex-wrap: wrap;
}

.stat-item { text-align: center; min-width: 80px; }

.stat-value {
    font-family: var(--font-display);
    font-size: var(--h2-size);
    color: var(--accent);
    line-height: 1.2;
}

.stat-label {
    font-size: var(--small-size);
    color: var(--text-secondary);
    margin-top: 0.2rem;
    text-transform: uppercase;
    letter-spacing: 0.05em;
}

/* ── 分区卡片网格 ── */
.cards-grid {
    display: grid;
    grid-template-columns: repeat(auto-fill, minmax(min(100%, 380px), 1fr));
    gap: clamp(1rem, 2vw, 1.5rem);
    max-width: 1200px;
    margin: 0 auto;
}

.category-card {
    background: var(--bg-card);
    border-radius: var(--radius);
    border: 1px solid var(--border);
    overflow: hidden;
    transition: transform 0.3s var(--ease-out-expo), border-color 0.3s ease;
}

.category-card:hover {
    transform: translateY(-2px);
    border-color: var(--accent-border);
}

.card-header {
    display: flex;
    align-items: center;
    gap: 0.8rem;
    padding: clamp(1rem, 2vw, 1.3rem) clamp(1rem, 2vw, 1.5rem);
    background: linear-gradient(135deg, var(--accent) 0%, var(--accent-light) 100%);
}

.card-number {
    font-family: var(--font-display);
    font-size: var(--h2-size);
    color: rgba(0, 0, 0, 0.3);
    line-height: 1;
}

.card-category {
    flex: 1;
    font-family: var(--font-body);
    font-size: var(--h3-size);
    font-weight: 700;
    color: #fff;
}

.card-count {
    font-size: var(--small-size);
    color: rgba(255, 255, 255, 0.8);
    font-weight: 500;
    background: rgba(0, 0, 0, 0.2);
    padding: 0.2rem 0.6rem;
    border-radius: 10px;
}

.card-body {
    padding: clamp(0.8rem, 1.5vw, 1.2rem) clamp(1rem, 2vw, 1.5rem);
    max-height: 520px;
    overflow-y: auto;
}

.card-body::-webkit-scrollbar { width: 4px; }
.card-body::-webkit-scrollbar-track { background: transparent; }
.card-body::-webkit-scrollbar-thumb { background: var(--border); border-radius: 2px; }
.card-body::-webkit-scrollbar-thumb:hover { background: var(--text-secondary); }

/* ── 文章列表项 ── */
.article-item {
    padding: 0.7rem 0;
    border-bottom: 1px solid var(--border);
    display: flex;
    gap: 0.8rem;
    align-items: flex-start;
}

.article-item:last-child {
    border-bottom: none;
    padding-bottom: 0;
}

.article-cover {
    width: 72px;
    height: 72px;
    border-radius: var(--radius-sm);
    object-fit: cover;
    flex-shrink: 0;
    background: var(--bg-secondary);
}

.article-info {
    flex: 1;
    min-width: 0;
}

.article-title {
    display: -webkit-box;
    -webkit-line-clamp: 2;
    -webkit-box-orient: vertical;
    overflow: hidden;
    color: var(--text-primary);
    text-decoration: none;
    font-size: var(--body-size);
    font-weight: 500;
    line-height: 1.5;
    transition: color 0.2s ease;
}

.article-title:hover {
    color: var(--accent);
}

.article-meta {
    display: flex;
    justify-content: space-between;
    align-items: center;
    margin-top: 0.3rem;
    font-size: var(--small-size);
    color: var(--text-muted);
    flex-wrap: wrap;
    gap: 0.3rem;
}

.author {
    color: var(--text-secondary);
}

.metrics {
    display: flex;
    gap: 0.8rem;
}

.metric {
    white-space: nowrap;
}

.report-footer {
    text-align: center;
    padding: clamp(2rem, 4vw, 3rem) 0 1rem;
    color: var(--text-muted);
    font-size: var(--small-size);
}

.report-footer .powered { margin-top: 0.5rem; color: var(--text-secondary); }

.reveal {
    opacity: 0;
    transform: translateY(20px);
    transition: opacity 0.6s var(--ease-out-expo), transform 0.6s var(--ease-out-expo);
}

.reveal.visible { opacity: 1; transform: translateY(0); }

.category-card:nth-child(1) { transition-delay: 0.05s; }
.category-card:nth-child(2) { transition-delay: 0.1s; }
.category-card:nth-child(3) { transition-delay: 0.15s; }
.category-card:nth-child(4) { transition-delay: 0.2s; }
.category-card:nth-child(5) { transition-delay: 0.25s; }
.category-card:nth-child(6) { transition-delay: 0.3s; }
.category-card:nth-child(7) { transition-delay: 0.35s; }
.category-card:nth-child(8) { transition-delay: 0.4s; }

@media (max-width: 600px) {
    .stats-bar { gap: 1rem; }
    .stat-item { min-width: 60px; }
    .cards-grid { grid-template-columns: 1fr; }
    .date-display .date-text { font-size: 1.3rem; }
    .article-cover { width: 56px; height: 56px; }
}

@media (prefers-reduced-motion: reduce) {
    .reveal { opacity: 1; transform: none; transition: none; }
}
</style>
</head>
<body>

<header class="report-header">
    <h1>AI抖音信息源</h1>
    <p class="subtitle">每日 AI 热门作品 · 互动量靠前的爆款一网打尽</p>

    <div class="date-display">
        <button class="date-nav-btn" id="prevDay" title="前一天">&larr;</button>
        <span class="date-text">{{DATE_CN}}</span>
        <button class="date-nav-btn" id="nextDay" title="后一天">&rarr;</button>
    </div>

</header>

<div class="stats-bar">
    <div class="stat-item">
        <div class="stat-value">{{TOPIC_COUNT}}</div>
        <div class="stat-label">分类</div>
    </div>
    <div class="stat-item">
        <div class="stat-value">{{TOTAL_COUNT}}</div>
        <div class="stat-label">作品</div>
    </div>
    <div class="stat-item">
        <div class="stat-value">{{AVG_LIKES}}</div>
        <div class="stat-label">平均点赞</div>
    </div>
    <div class="stat-item">
        <div class="stat-value">{{TOTAL_SHARES}}</div>
        <div class="stat-label">总分享</div>
    </div>
</div>

<div class="cards-grid" id="cardsGrid">
{{CATEGORY_CARDS}}
</div>

<footer class="report-footer">
    <p>Generated at {{TIMESTAMP}}</p>
    <p class="powered">Powered by AI抖音信息源 Skill</p>
</footer>

<script>
(function() {
    var observer = new IntersectionObserver(function(entries) {
        entries.forEach(function(entry) {
            if (entry.isIntersecting) entry.target.classList.add('visible');
        });
    }, { threshold: 0.1, rootMargin: '0px 0px -50px 0px' });

    document.querySelectorAll('.reveal').forEach(function(el) {
        observer.observe(el);
    });

    var currentDate = '{{DATE}}';
    function navigateDate(offset) {
        var parts = currentDate.split('-');
        var d = new Date(parseInt(parts[0]), parseInt(parts[1]) - 1, parseInt(parts[2]));
        d.setDate(d.getDate() + offset);
        var y = d.getFullYear();
        var m = String(d.getMonth() + 1).padStart(2, '0');
        var day = String(d.getDate()).padStart(2, '0');
        var newDate = y + '-' + m + '-' + day;
        var newFile = '抖音AI日报_' + newDate + '.html';
        window.location.href = newFile;
    }

    document.getElementById('prevDay').addEventListener('click', function() { navigateDate(-1); });
    document.getElementById('nextDay').addEventListener('click', function() { navigateDate(1); });
})();
</script>

</body>
</html>

AI Douyin Feed / douyin-ai-feed

---

Overview

Daily auto-scan of AI-related Douyin content, filtering the hottest posts by engagement (likes / shares / comments), intelligently clustering them into topics, and generating a beautifully styled HTML daily report with cover images, metrics, and subscription support. Your one-stop solution to stay on top of the AI scene on Douyin.

Core Value

Discover the hottest AI posts on Douyin daily, saving time from manual browsing
Auto-clustering by content topics gives you a clear picture of what's trending in AI today
Stunning dark-themed visual daily report with cover images and engagement metrics at a glance
Daily auto-push with archived reports — browse history anytime

Who It's For

📱 Content operators — track AI trend styles and get creative inspiration
🎬 AI creators — learn from trending content and understand platform dynamics
📊 Industry observers — monitor AI content trends for market insights

---

Features

Core Capabilities

Search Douyin AI posts by keywords, filtered by engagement metrics
Support time-range queries (start time inclusive, end time exclusive) for flexible historical lookback
Auto-detect topic directions from content tags and intelligently cluster into category cards
Terminal table display: category + title + author + likes / shares / comments
Generate a dark-themed visual daily report (Douyin cyan style) with cover images, metrics, direct links, and date navigation

---

API Key Acquisition & Security

This skill requires the environment variable: REDFOX_API_KEY.
REDFOX_API_KEY is provided by RedFoxHub (https://redfox.hk).
Visit RedFoxHub to register and get your REDFOX_API_KEY.
Configure the REDFOX_API_KEY environment variable on your device before using this skill.
Before providing your key, verify the source, scope, validity period, and whether it supports reset/revocation.
Never hardcode or expose your key in code, prompts, logs, or output files.

---

Usage

Simply describe your need in natural language — no commands to memorize.

Quick Reference

Intent	Example	Result
View today's AI Douyin report	"Show me today's Douyin AI report"	Fetches the hottest AI posts and generates a clustered visual report
Search by custom keyword	"Search Douyin for ChatGPT content"	Filters by your keyword and generates a tailored report
Look up a past date	"Show me Douyin AI data from June 5th"	Retrieves the historical daily report for that date
Time range query	"Show me Douyin AI data from June 1st to June 5th"	Flexible lookback for any time period

Sample Output

Running the skill generates a dark-themed visual daily report including:

Header with date, category stats, average likes, and total shares
Category cards displaying posts (cover image + title + author + engagement metrics)
Date navigation to browse adjacent days

---

Use Cases

Scenario	Role	Example Query	Benefit
Daily AI trend tracking	Content operator	"Today's AI Douyin report"	Automatically get daily hot content, no manual browsing
Competitor content analysis	AI creator	"Search Douyin for trending AI short dramas"	Understand trending styles and performance data
Hot topic review	Industry analyst	"Show me Douyin AI trends from June 1st"	Review historical data and analyze topic evolution

AI抖音信息源 / douyin-ai-feed

---

简介

每日自动扫描抖音 AI 相关作品，按互动量筛选最火内容，智能聚类生成精美可视化日报，支持封面图、数据指标和每日订阅。帮你一站式掌握抖音 AI 赛道每日爆款，不错过任何一个热点。

核心价值

每日自动发现抖音最火的 AI 作品，省去手动刷抖音找素材的时间
按内容主题自动聚类，一眼看清今天 AI 圈在聊什么
精美深色主题可视化日报，封面图 + 互动数据一目了然
支持每日自动推送，日报自动积攒，随时翻阅历史

适用对象

📱 内容运营 — 追踪 AI 赛道爆款风格，获取创作灵感
🎬 AI 创作者 — 了解同行热门内容，把握平台风向
📊 行业观察者 — 监控 AI 内容趋势，辅助市场判断

---

功能特性

核心功能

按关键词搜索抖音 AI 作品，按互动量筛选最火内容
支持按时间范围查询历史数据（开始时间含、结束时间不含），灵活回溯任意时段
自动从内容标签中识别话题方向，智能聚类生成分类卡片
终端表格展示分类 + 标题 + 作者 + 点赞/分享/评论
生成深色主题可视化日报（抖音青色风格），含封面图、互动数据、作品直链、日期导航

---

密钥获取与安全说明

本技能需要使用环境变量：REDFOX_API_KEY。
REDFOX_API_KEY 由红狐 hub (https://redfox.hk)提供。
请前往红狐 hub 注册账号，获取 REDFOX_API_KEY。
配置设备环境变量 REDFOX_API_KEY 后使用本技能。
在提供密钥前，请先确认密钥来源、可用范围、有效期及是否支持重置/撤销。
禁止在代码、提示词、日志或输出文件中硬编码/明文暴露密钥。

---

使用指南

直接用自然语言描述需求，无需记忆命令。

常用说法速查

意图	示例话术	效果
查看今日 AI 抖音日报	「看看抖音 AI 热点」「抖音 AI 日报」	拉取当日最火 AI 作品，聚类生成可视化日报
指定关键词搜索	「搜一下抖音 ChatGPT 相关内容」	按自定义关键词筛选，生成对应日报
回溯历史某天	「看一下 6 月 5 号的抖音 AI 数据」	拉取指定日期的历史日报
指定时间范围	「查一下 6 月 1 号到 6 月 5 号的抖音 AI 数据」	按时间范围回溯任意时段数据

输出示例

运行后自动生成一份深色主题可视化日报，包含：

顶部日期、分类统计、平均点赞、总分享数
按分类分区展示作品卡片（封面图 + 标题 + 作者 + 互动数据）
支持日期翻页，查看前后日日报

---

使用场景

场景	角色	示例问法	收益
每日 AI 趋势追踪	内容运营	「今天的 AI 抖音日报」	省去手动刷抖音，自动获取当日爆款
竞品内容分析	AI 创作者	「搜一下抖音 AI 短剧的热门内容」	了解同行爆款风格与数据表现
热点事件复盘	行业分析	「看看 6 月 1 号的抖音 AI 趋势」	回溯历史数据，分析 AI 话题演变

#!/usr/bin/env python3
"""
AI抖音信息源 — 每日 AI 相关抖音热门内容聚合
=================================================
根据关键词搜索抖音 AI 相关作品，按互动量排序后生成 HTML 日报。

Usage:
    python3 fetch_douyin_ai.py
    python3 fetch_douyin_ai.py --keyword "ChatGPT"
    python3 fetch_douyin_ai.py --subscribe
"""

import argparse
import json
import os
import subprocess
import sys
import time
from collections import Counter, defaultdict
from datetime import datetime, timedelta
from pathlib import Path

try:
    import requests
    HAS_REQUESTS = True
except ImportError:
    HAS_REQUESTS = False

# ─── 配置 ─────────────────────────────────────────────────────────────────────────
API_URL = "https://redfox.hk/story/api/parseWork/queryDyAiMsgs"
ENV_KEY = "REDFOX_API_KEY"
SOURCE = "AI抖音信息源-GitHub"

DEFAULT_KEYWORD = "AI"
DEFAULT_OUTPUT_DIR = Path.home() / "Downloads" / "QoderReports"
PAGE_SIZE = 200

PLIST_LABEL = "com.qoder.douyin-ai-feed"
PLIST_DIR = Path.home() / "Library" / "LaunchAgents"

# ─── 终端颜色 ──────────────────────────────────────────────────────────────────────
GREEN = "\033[92m"
YELLOW = "\033[93m"
RED = "\033[91m"
CYAN = "\033[96m"
BOLD = "\033[1m"
RESET = "\033[0m"


def info(msg):
    print(f"{GREEN}[✓]{RESET} {msg}")


def warn(msg):
    print(f"{YELLOW}[!]{RESET} {msg}")


def error(msg):
    print(f"{RED}[✗]{RESET} {msg}")


def step(msg):
    print(f"{CYAN}[→]{RESET} {msg}")


# ─── API Key 管理 ──────────────────────────────────────────────────────────────────
def get_api_key():
    """从环境变量读取 API Key，缺失时报错退出。"""
    key = os.environ.get(ENV_KEY, "").strip()
    if key:
        return key
    error(f"未检测到环境变量 {ENV_KEY}")
    print(f"\n{YELLOW}  请先配置 API Key：{RESET}")
    print(f"  {CYAN}方式一（推荐）{RESET}：export REDFOX_API_KEY=ak_你的密钥")
    print(f"  {CYAN}方式二{RESET}：将 export REDFOX_API_KEY=ak_你的密钥 写入 ~/.zshrc 或 ~/.bashrc")
    print(f"\n  获取 API Key：{CYAN}https://www.redfox.hk/login{RESET}")
    sys.exit(1)


# ─── 数据获取 ──────────────────────────────────────────────────────────────────────
def fetch_articles(session, keyword, page_size, start_time=None, end_time=None):
    """单次调用接口获取作品数据，返回实际获取的作品列表"""
    payload = {
        "keyword": keyword,
        "pageNum": 1,
        "pageSize": page_size,
        "source": SOURCE,
    }
    if start_time:
        payload["startTime"] = start_time
    if end_time:
        payload["endTime"] = end_time
    print(f"\r  {CYAN}[→]{RESET} 请求中: keyword=\"{keyword}\", pageSize={page_size}", end="", flush=True)

    try:
        resp = session.post(API_URL, json=payload, timeout=15)
        result = resp.json()
    except Exception as e:
        print()
        warn(f"请求失败: {e}")
        return []

    code = result.get("code")
    if code == 3108:
        warn("限频，等待 5s...")
        time.sleep(5)
        try:
            resp = session.post(API_URL, json=payload, timeout=15)
            result = resp.json()
            code = result.get("code")
        except Exception:
            print()
            return []

    if code not in (200, 2000):
        if code in (3106, 3107):
            error(f"API Key 错误 (code {code}): {result.get('msg', '')}")
        print()
        return []

    data = result.get("data", {})
    articles = data.get("list", [])
    print(f"\r  {CYAN}[→]{RESET} 接口返回 {len(articles)} 篇作品                      ")
    return articles


# ─── 自动聚类 ──────────────────────────────────────────────────────────────────────
def cluster_articles(articles):
    """基于 type / topic 标签自动聚类，大类按第二标签二次拆分"""
    total = len(articles)

    def _get_tags(article):
        atype = (article.get("type") or "").strip()
        topic = (article.get("topic") or "").strip()
        raw = atype or topic or ""
        return [t.strip().lstrip("#").strip() for t in raw.split(",") if t.strip()]

    # 第一轮：按首标签聚类
    topic_groups = defaultdict(list)
    for article in articles:
        tags = _get_tags(article)
        category = tags[0] if tags else "其他"
        topic_groups[category].append(article)

    # 合并小组（< 2 篇归入"其他"）
    final_groups = {}
    small_articles = []
    for topic, arts in topic_groups.items():
        if len(arts) >= 2:
            final_groups[topic] = arts
        else:
            small_articles.extend(arts)
    if small_articles:
        final_groups.setdefault("其他", []).extend(small_articles)

    # 二次拆分：超过总量 50% 的分类，按第二标签再拆分
    to_split = {k: v for k, v in final_groups.items() if len(v) > total * 0.5 and k != "其他"}
    for big_cat, big_arts in to_split.items():
        del final_groups[big_cat]
        sub_groups = defaultdict(list)
        for art in big_arts:
            tags = _get_tags(art)
            if len(tags) >= 2:
                sub_groups[tags[1]].append(art)
            else:
                sub_groups["其他"].append(art)
        for sub_name, sub_arts in sub_groups.items():
            key = sub_name if sub_name != "其他" else f"{big_cat}·其他"
            final_groups[key] = sub_arts

    # 二次合并（拆分后的小组 < 2 篇归入"其他"）
    merged = {}
    small2 = []
    for topic, arts in final_groups.items():
        if len(arts) >= 2:
            merged[topic] = arts
        else:
            small2.extend(arts)
    if small2:
        merged.setdefault("其他", []).extend(small2)

    # 构建输出，按条数降序
    clusters = []
    for category, arts in sorted(merged.items(), key=lambda x: -len(x[1])):
        sorted_arts = sorted(arts, key=lambda a: (
            (a.get("likeCount") or 0) + (a.get("shareCount") or 0) + (a.get("commentCount") or 0)
        ), reverse=True)
        clusters.append({
            "category": category,
            "count": len(sorted_arts),
            "articles": sorted_arts,
        })

    return clusters


# ─── HTML 报告生成 ──────────────────────────────────────────────────────────────────
def compute_stats(articles):
    """计算统计数据"""
    total = len(articles)
    if total == 0:
        return {"total": 0, "avg_likes": 0, "top_author": "-", "total_shares": 0}

    likes = [a.get("likeCount") or 0 for a in articles]
    avg_likes = sum(likes) // total if total > 0 else 0

    author_counter = Counter(a.get("userName") or "未知" for a in articles)
    top_author = author_counter.most_common(1)[0][0] if author_counter else "-"

    total_shares = sum(a.get("shareCount") or 0 for a in articles)

    return {
        "total": total,
        "avg_likes": avg_likes,
        "top_author": top_author,
        "total_shares": total_shares,
    }


def format_number(n):
    """格式化数字: 1234 -> 1.2k"""
    if n is None:
        return "0"
    if n >= 10000:
        return f"{n/10000:.1f}w"
    if n >= 1000:
        return f"{n/1000:.1f}k"
    return str(n)


def print_article_table(clusters):
    """在终端打印分类作品表格"""
    print(f"\n{BOLD}{'='*78}{RESET}")
    print(f"{BOLD}  AI抖音信息源 · 分类作品一览{RESET}")
    print(f"{BOLD}{'='*78}{RESET}\n")

    for i, cluster in enumerate(clusters, 1):
        category = cluster["category"]
        arts = cluster["articles"]

        print(f"  {CYAN}{BOLD}【{category}】{RESET} "
              f"{cluster['count']} 篇")

        header = (f"  {'序号':<4}{'标题':<36}{'作者':<14}"
                  f"{'点赞':>8}{'分享':>8}{'评论':>8}")
        print(f"  {YELLOW}{'─'*76}{RESET}")
        print(f"  {YELLOW}{header}{RESET}")
        print(f"  {YELLOW}{'─'*76}{RESET}")

        for j, article in enumerate(arts, 1):
            title = article.get("title", "无标题") or "无标题"
            author = article.get("userName") or "-"
            likes = format_number(article.get("likeCount"))
            shares = format_number(article.get("shareCount"))
            comments = format_number(article.get("commentCount"))

            display_title = title[:34] + ".." if len(title) > 36 else title
            display_author = author[:12] + ".." if len(author) > 14 else author

            print(f"  {j:<4}{display_title:<36}{display_author:<14}"
                  f"{likes:>8}{shares:>8}{comments:>8}")

        print()


def generate_category_cards(clusters):
    """生成分区卡片 HTML（每个分类一个大卡片，内含竖向文章列表）"""
    cards_html = ""
    for i, cluster in enumerate(clusters, 1):
        articles_html = ""
        for article in cluster["articles"][:10]:  # 每个分类只展示 top 10
            title = article.get("title", "无标题") or "无标题"
            photo_id = article.get("photoId") or ""
            url = f"https://www.iesdouyin.com/share/video/{photo_id}" if photo_id else "#"
            author = article.get("userName") or ""
            cover = article.get("coverUrl") or ""
            likes = format_number(article.get("likeCount"))
            shares = format_number(article.get("shareCount"))
            comments = format_number(article.get("commentCount"))

            cover_html = ""
            if cover:
                cover_html = f'<img class="article-cover" src="{cover}" alt="" loading="lazy" referrerpolicy="no-referrer">'

            articles_html += f'''
                <div class="article-item">
                    {cover_html}
                    <div class="article-info">
                        <a href="{url}" target="_blank" class="article-title">{title}</a>
                        <div class="article-meta">
                            <span class="author">{author}</span>
                            <span class="metrics">
                                <span class="metric">&#x1f44d; {likes}</span>
                                <span class="metric">&#x1f501; {shares}</span>
                                <span class="metric">&#x1f4ac; {comments}</span>
                            </span>
                        </div>
                    </div>
                </div>'''

        cards_html += f'''
        <div class="category-card reveal">
            <div class="card-header">
                <span class="card-number">{i:02d}</span>
                <h3 class="card-category">{cluster["category"]}</h3>
                <span class="card-count">{cluster["count"]} 篇</span>
            </div>
            <div class="card-body">{articles_html}
            </div>
        </div>'''

    return cards_html


def generate_report(clusters, articles, date_str):
    """生成完整 HTML 报告"""
    stats = compute_stats(articles)
    topic_count = len(clusters)

    template_path = Path(__file__).parent.parent / "assets" / "report_template.html"
    if template_path.exists():
        template = template_path.read_text(encoding="utf-8")
    else:
        warn("模板文件未找到，使用内置模板")
        template = get_fallback_template()

    try:
        dt = datetime.strptime(date_str, "%Y-%m-%d")
        weekdays = ["一", "二", "三", "四", "五", "六", "日"]
        date_cn = f"{dt.year}年{dt.month}月{dt.day}日 星期{weekdays[dt.weekday()]}"
    except ValueError:
        date_cn = date_str

    category_cards = generate_category_cards(clusters)
    timestamp = datetime.now().strftime("%Y-%m-%d %H:%M:%S")

    html = template
    html = html.replace("{{DATE}}", date_str)
    html = html.replace("{{DATE_CN}}", date_cn)
    html = html.replace("{{TOTAL_COUNT}}", str(stats["total"]))
    html = html.replace("{{TOPIC_COUNT}}", str(topic_count))
    html = html.replace("{{TOP_AUTHOR}}", stats["top_author"])
    html = html.replace("{{AVG_LIKES}}", format_number(stats["avg_likes"]))
    html = html.replace("{{TOTAL_SHARES}}", format_number(stats["total_shares"]))
    html = html.replace("{{CATEGORY_CARDS}}", category_cards)
    html = html.replace("{{TIMESTAMP}}", timestamp)

    return html


def get_fallback_template():
    """内置最小 HTML 模板（当模板文件缺失时使用）"""
    return '''<!DOCTYPE html>
<html lang="zh-CN">
<head>
<meta charset="UTF-8">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<title>AI抖音信息源 - {{DATE}}</title>
<style>
* { margin: 0; padding: 0; box-sizing: border-box; }
body { font-family: -apple-system, sans-serif; background: #1a1a1a; color: #e8e4df; padding: 2rem; }
.header { text-align: center; padding: 2rem 0; }
.header h1 { font-size: 2rem; color: #25F4EE; }
.header p { color: #9a9590; margin-top: 0.5rem; }
.stats { display: flex; justify-content: center; gap: 2rem; padding: 1rem; margin: 1rem 0; }
.stat-item { text-align: center; }
.stat-value { font-size: 1.5rem; font-weight: bold; color: #25F4EE; }
.stat-label { font-size: 0.8rem; color: #9a9590; }
.cards { display: grid; grid-template-columns: repeat(auto-fill, minmax(380px, 1fr)); gap: 1.5rem; max-width: 1200px; margin: 2rem auto; }
.category-card { background: #2d2d2d; border-radius: 12px; overflow: hidden; }
.card-header { display: flex; align-items: center; gap: 0.8rem; padding: 1rem 1.2rem; background: linear-gradient(135deg, #25F4EE, #5CF7F2); }
.card-number { font-size: 1.5rem; font-weight: bold; color: rgba(0,0,0,0.3); }
.card-category { flex: 1; font-size: 1.1rem; font-weight: 700; color: #fff; }
.card-count { font-size: 0.8rem; color: rgba(255,255,255,0.8); background: rgba(0,0,0,0.2); padding: 0.2rem 0.6rem; border-radius: 10px; }
.card-body { padding: 0.8rem 1.2rem; max-height: 520px; overflow-y: auto; }
.card-body::-webkit-scrollbar { width: 4px; }
.card-body::-webkit-scrollbar-track { background: transparent; }
.card-body::-webkit-scrollbar-thumb { background: #3d3d3d; border-radius: 2px; }
.article-item { padding: 0.6rem 0; border-bottom: 1px solid #3d3d3d; display: flex; gap: 0.8rem; }
.article-item:last-child { border-bottom: none; }
.article-cover { width: 72px; height: 72px; border-radius: 8px; object-fit: cover; flex-shrink: 0; background: #1a1a1a; }
.article-info { flex: 1; min-width: 0; }
.article-title { color: #e8e4df; text-decoration: none; font-size: 0.9rem; line-height: 1.4; display: -webkit-box; -webkit-line-clamp: 2; -webkit-box-orient: vertical; overflow: hidden; }
.article-title:hover { color: #25F4EE; }
.article-meta { display: flex; justify-content: space-between; margin-top: 0.3rem; font-size: 0.75rem; color: #9a9590; }
.metrics { display: flex; gap: 0.8rem; }
.footer { text-align: center; padding: 2rem; color: #666; font-size: 0.8rem; }
</style>
</head>
<body>
<div class="header">
    <h1>AI抖音信息源</h1>
    <p>{{DATE_CN}} | 共 {{TOTAL_COUNT}} 篇热门作品</p>
</div>
<div class="stats">
    <div class="stat-item"><div class="stat-value">{{TOPIC_COUNT}}</div><div class="stat-label">分类</div></div>
    <div class="stat-item"><div class="stat-value">{{TOTAL_COUNT}}</div><div class="stat-label">作品</div></div>
    <div class="stat-item"><div class="stat-value">{{AVG_LIKES}}</div><div class="stat-label">平均点赞</div></div>
    <div class="stat-item"><div class="stat-value">{{TOTAL_SHARES}}</div><div class="stat-label">总分享</div></div>
</div>
<div class="cards">{{CATEGORY_CARDS}}</div>
<div class="footer">Generated at {{TIMESTAMP}} by AI抖音信息源 Skill</div>
</body>
</html>'''


# ─── 订阅机制 ──────────────────────────────────────────────────────────────────────
def install_subscription():
    """安装定时任务，每天自动生成日报"""
    if sys.platform == "darwin":
        PLIST_DIR.mkdir(parents=True, exist_ok=True)
        plist_path = PLIST_DIR / f"{PLIST_LABEL}.plist"

        script_path = os.path.abspath(__file__)
        log_path = str(Path.home() / "Library" / "Logs" / "qoder-douyin-ai-feed.log")

        env_section = ""
        api_key = os.environ.get(ENV_KEY, "").strip()
        if api_key:
            env_section = f"""
        <key>EnvironmentVariables</key>
        <dict>
            <key>{ENV_KEY}</key>
            <string>{api_key}</string>
        </dict>"""

        plist_content = f'''<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd">
<plist version="1.0">
<dict>
    <key>Label</key>
    <string>{PLIST_LABEL}</string>
    <key>ProgramArguments</key>
    <array>
        <string>/usr/bin/python3</string>
        <string>{script_path}</string>
        <string>--no-open</string>
    </array>
    <key>StartCalendarInterval</key>
    <dict>
        <key>Hour</key>
        <integer>16</integer>
        <key>Minute</key>
        <integer>0</integer>
    </dict>
    <key>StandardOutPath</key>
    <string>{log_path}</string>
    <key>StandardErrorPath</key>
    <string>{log_path}</string>
    <key>RunAtLoad</key>
    <false/>{env_section}
</dict>
</plist>'''

        plist_path.write_text(plist_content, encoding="utf-8")

        try:
            subprocess.run(["launchctl", "load", str(plist_path)],
                           check=True, capture_output=True)
            info("订阅成功! 每天 16:00 自动生成抖音 AI 日报")
            info(f"日报目录: ~/Downloads/QoderReports/")
            info(f"日志: {log_path}")
            return True
        except subprocess.CalledProcessError as e:
            error(f"订阅安装失败: {e.stderr.decode()}")
            return False
    else:
        script_path = os.path.abspath(__file__)
        cron_line = f"0 16 * * * /usr/bin/python3 {script_path} --no-open"
        try:
            subprocess.run(
                f'(crontab -l 2>/dev/null; echo "{cron_line}") | crontab -',
                shell=True, check=True, capture_output=True
            )
            info("订阅成功! 每天 16:00 自动生成抖音 AI 日报 (crontab)")
            info(f"日报目录: ~/Downloads/QoderReports/")
            return True
        except subprocess.CalledProcessError:
            warn("自动配置 crontab 失败，请手动添加:")
            print(f"  {cron_line}")
            return False


def remove_subscription():
    """卸载定时任务"""
    if sys.platform == "darwin":
        plist_path = PLIST_DIR / f"{PLIST_LABEL}.plist"
        if not plist_path.exists():
            warn("未找到订阅配置，无需取消")
            return False
        try:
            subprocess.run(["launchctl", "unload", str(plist_path)],
                           check=True, capture_output=True)
        except subprocess.CalledProcessError:
            pass
        plist_path.unlink(missing_ok=True)
        info("已取消订阅，定时任务已移除")
        return True
    else:
        script_path = os.path.abspath(__file__)
        try:
            subprocess.run(
                f'crontab -l 2>/dev/null | grep -v "{script_path}" | crontab -',
                shell=True, check=True, capture_output=True
            )
            info("已取消订阅，crontab 任务已移除")
            return True
        except subprocess.CalledProcessError:
            warn("自动移除 crontab 失败，请手动执行: crontab -e")
            return False


# ─── 主流程 ────────────────────────────────────────────────────────────────────────
def main():
    parser = argparse.ArgumentParser(
        description="AI抖音信息源 — 每日 AI 相关抖音热门内容聚合",
        formatter_class=argparse.RawDescriptionHelpFormatter,
        epilog="""
Examples:
  python3 fetch_douyin_ai.py
  python3 fetch_douyin_ai.py --keyword "ChatGPT"
  python3 fetch_douyin_ai.py --start-time 2026-06-09 --end-time 2026-06-10
  python3 fetch_douyin_ai.py --subscribe
  python3 fetch_douyin_ai.py --unsubscribe
        """,
    )
    parser.add_argument("--keyword", default=DEFAULT_KEYWORD,
                        help=f"搜索关键词 (默认: {DEFAULT_KEYWORD})")
    parser.add_argument("--page-size", type=int, default=PAGE_SIZE,
                        help=f"每页条数 (默认: {PAGE_SIZE})")
    parser.add_argument("--date", default=datetime.now().strftime("%Y-%m-%d"),
                        help="指定日期 YYYY-MM-DD (默认: 今天)")
    parser.add_argument("--start-time", default=None,
                        help="开始时间 YYYY-MM-DD（含），不传则根据 --date 自动计算")
    parser.add_argument("--end-time", default=None,
                        help="结束时间 YYYY-MM-DD（不含），不传则根据 --date 自动计算")
    parser.add_argument("--output-dir",
                        help=f"输出目录 (默认: ~/Downloads/QoderReports)")
    parser.add_argument("--subscribe", action="store_true",
                        help="安装每日定时任务 (16:00)")
    parser.add_argument("--unsubscribe", action="store_true",
                        help="卸载定时任务")
    parser.add_argument("--no-open", action="store_true",
                        help="不自动打开浏览器")

    args = parser.parse_args()

    # ── Banner ──
    banner = f"""{CYAN}{BOLD}
  ╔══════════════════════════════════════╗
  ║     AI抖音信息源 · 日报生成         ║
  ║     每日 AI 热门作品 · 爆款一网打尽 ║
  ╚══════════════════════════════════════╝{RESET}
"""
    print(banner)

    # ── 订阅/取消 ──
    if args.subscribe:
        install_subscription()
        return

    if args.unsubscribe:
        remove_subscription()
        return

    # ── 检查依赖 ──
    if not HAS_REQUESTS:
        error("缺少 requests 库，请安装: pip3 install requests")
        sys.exit(1)

    # ── API Key ──
    api_key = get_api_key()

    # ── Session ──
    session = requests.Session()
    session.verify = True
    session.headers.update({
        "Content-Type": "application/json",
        "X-API-KEY": api_key,
    })

    # ── 计算时间参数 ──
    start_time = args.start_time
    end_time = args.end_time
    if not start_time and not end_time and args.date:
        start_time = args.date
        try:
            dt = datetime.strptime(args.date, "%Y-%m-%d")
            end_time = (dt + timedelta(days=1)).strftime("%Y-%m-%d")
        except ValueError:
            pass

    # ── 获取作品（单次接口调用）──
    keyword = args.keyword.strip()
    time_info = ""
    if start_time:
        time_info += f", startTime={start_time}"
    if end_time:
        time_info += f", endTime={end_time}"
    step(f"查询关键词: \"{keyword}\" (pageSize={args.page_size}{time_info})")

    articles = fetch_articles(session, keyword, args.page_size, start_time, end_time)

    if not articles:
        error("未获取到任何作品")
        sys.exit(1)

    info(f"获取完成: {len(articles)} 篇作品")

    # ── 自动聚类 ──
    step("正在自动聚类...")
    clusters = cluster_articles(articles)
    info(f"聚类完成: 发现 {len(clusters)} 个分类")
    for c in clusters[:10]:
        print(f"    {c['category']}: {c['count']} 篇")

    # ── 终端表格展示 ──
    print_article_table(clusters)

    # ── 生成报告 ──
    step("生成 HTML 日报...")
    html_content = generate_report(clusters, articles, args.date)

    # ── 保存文件 ──
    output_dir = Path(args.output_dir) if args.output_dir else DEFAULT_OUTPUT_DIR
    output_dir.mkdir(parents=True, exist_ok=True)
    filename = f"抖音AI日报_{args.date}.html"
    output_path = output_dir / filename

    output_path.write_text(html_content, encoding="utf-8")
    info(f"日报已生成: {output_path}")

    # ── 打开浏览器 ──
    if not args.no_open:
        if sys.platform == "darwin":
            subprocess.run(["open", str(output_path)], check=False)
        elif sys.platform == "linux":
            subprocess.run(["xdg-open", str(output_path)], check=False)
        info(f"浏览器已打开: {output_path}")

    print(f"\n{GREEN}{BOLD}✓ 完成!{RESET}")
    print(f"  文件: {output_path}")
    print(f"  分类: {len(clusters)} 个")
    print(f"  作品: {len(articles)} 篇")
    print(f"\n{CYAN}💡 需要每日下午4点自动推送最新研究数据吗？{RESET}")


if __name__ == "__main__":
    main()