# Trending Skills

> Get skills.sh trending rankings with AI summaries, trend analysis, and category insights. Use when asking about skill rankings, trends, or skill details.

- Skill: `geekjourneyx/trending-skills` (Agent Skill)
- Install (CLI): `npx skillmds@latest add geekjourneyx/trending-skills`
- Raw SKILL.md: https://api.skillmd.com/api/skills/geekjourneyx/trending-skills/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: geekjourneyx (https://skillmd.com/u/geekjourneyx)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/geekjourneyx/trending-skills

---


# Trending Skills

Fetches the latest skills rankings from skills.sh/trending, provides AI-powered skill summaries, and analyzes ranking trends (rising, falling, new entries).

## Quick Start

```
# View today's rankings
今天技能排行榜
Top 10 skills
技能榜单

# View skill details
remotion-best-practices 是什么
frontend-design 技能介绍

# View trends
技能趋势分析
哪些技能上升了
今天有新技能吗
新晋榜单
```

---

## Supported Query Types

| Type | Examples | Description |
|------|----------|-------------|
| **今日榜单** | "今天技能排行榜" "Top 10" "技能榜单" | Get current rankings |
| **技能详情** | "xxx是什么" "xxx技能介绍" | Get AI summary of a skill |
| **趋势分析** | "技能趋势" "哪些上升了" | Get trend analysis |

---

## Workflow

```
Progress:
- [ ] Step 1: Parse user query type
- [ ] Step 2: Fetch/Load ranking data
- [ ] Step 3: Format and display results
```

---

## Step 1: Parse Query Type

Determine what the user is asking for:

| User Input | Query Type | Action |
|------------|------------|--------|
| "今天技能排行榜" | `rankings` | Show top N rankings |
| "Top 10 skills" | `rankings` | Show top N rankings |
| "xxx是什么" | `detail` | Show skill details |
| "哪些技能上升了" | `trends` | Show rising skills |
| "新晋榜单" | `trends` | Show new entries |
| "技能趋势分析" | `trends` | Show full trend report |

---

## Step 2: Fetch/Load Data

### Option A: Use Database (Preferred)

If `data/trends.db` exists and has recent data:

```bash
# Check available dates
sqlite3 data/trends.db "SELECT date FROM skills_daily ORDER BY date DESC LIMIT 1;"

# Get latest rankings
sqlite3 data/trends.db "SELECT rank, name, owner, installs, installs_delta, rank_delta FROM skills_daily WHERE date = '2026-01-23' ORDER BY rank LIMIT 20;"
```

### Option B: Fetch from skills.sh

If no database or data is stale:

```python
# Run the fetcher
from src.skills_fetcher import SkillsFetcher

fetcher = SkillsFetcher()
skills = fetcher.fetch()  # Returns Top 100
```

---

## Step 3: Format Results

### Rankings Output Format

```markdown
# 📊 Skills Trending - 2026-01-23

| # | 技能 | 拥有者 | 安装量 | 变化 |
|---|------|--------|--------|------|
| 1 | remotion-best-practices | remotion-dev/skills | 5.6K | ↑ 50 |
| 2 | vercel-react-best-practices | vercel-labs/agent-skills | 5.4K | - |
| 3 | web-design-guidelines | vercel-labs/agent-skills | 4.0K | ↓ 2 |
...
```

### Detail Output Format

```markdown
# remotion-best-practices

**拥有者**: remotion-dev/skills
**排名**: #1 (5.6K 安装)

## 简介
用 React 代码创建视频的最佳实践

## 详细说明
程序化视频生成框架 Remotion 的最佳实践集合，包含 27 个规则。

## 解决问题
- 程序化视频
- 字幕生成
- 3D 动效
- 音频处理

## 使用场景
视频自动化、个性化视频生成、数据可视化视频

**分类**: 视频/动画

🔗 https://skills.sh/remotion-dev/skills/remotion-best-practices
```

### Trends Output Format

```markdown
# 📈 技能趋势分析 - 2026-01-23

## 上升 Top 5

| # | 技能 | 变化 |
|---|------|------|
| 7 | seo-audit | ↑ 38 |
| 15 | copywriting | ↑ 12 |
...

## 下降 Top 5

| # | 技能 | 变化 |
|---|------|------|
| 10 | old-skill | ↓ 15 |
...

## 新晋榜单

- new-skill (#82)
- another-new (#95)

## 跌出榜单

- dropped-skill (昨日 #75)
```

---

## Data Schema

### skills_daily Table

```sql
CREATE TABLE skills_daily (
    date TEXT,           -- YYYY-MM-DD
    rank INTEGER,        -- 排名
    name TEXT,           -- 技能名称
    owner TEXT,          -- 拥有者
    installs INTEGER,    -- 安装量
    installs_delta INTEGER,  -- 安装量变化
    rank_delta INTEGER   -- 排名变化 (正=上升)
);
```

### skills_details Table

```sql
CREATE TABLE skills_details (
    name TEXT PRIMARY KEY,
    summary TEXT,        -- 一句话摘要
    description TEXT,    -- 详细描述
    use_case TEXT,       -- 使用场景
    solves TEXT,         -- JSON: 解决的问题
    category TEXT,       -- 分类
    category_zh TEXT,    -- 中文分类
    rules_count INTEGER,
    owner TEXT,
    url TEXT
);
```

---

## Configuration

Environment variables (optional, for fetching):

```bash
# For AI summaries
ZHIPU_API_KEY=your_key

# For database
DB_PATH=data/trends.db
```

---

## Examples

### Example 1: Today's Rankings

**User Input**: "今天技能排行榜"

**Process**:
1. Query type: `rankings`
2. Load latest data from database or fetch from skills.sh
3. Format as table

**Output**:
```markdown
# 📊 Skills Trending - 2026-01-23

| # | 技能 | 安装量 | 变化 |
|---|------|--------|------|
| 1 | remotion-best-practices | 5.6K | - |
| 2 | vercel-react-best-practices | 5.4K | - |
...
```

### Example 2: Skill Detail

**User Input**: "remotion-best-practices 是什么"

**Process**:
1. Query type: `detail`
2. Parse skill name: `remotion-best-practices`
3. Get details from database or fetch from skills.sh
4. Format with AI summary

**Output**: (See Detail Output Format above)

### Example 3: Trends

**User Input**: "哪些技能上升了"

**Process**:
1. Query type: `trends` (filter: rising)
2. Get yesterday's data for comparison
3. Filter skills with `rank_delta > 0`
4. Sort by `rank_delta` DESC

**Output**:
```markdown
# 📈 上升中的技能

| # | 技能 | 上升 |
|---|------|------|
| 7 | seo-audit | ↑ 38 |
| 15 | copywriting | ↑ 12 |
...
```

---

## Troubleshooting

### No database found

If `data/trends.db` doesn't exist, fetch fresh data:

```python
from src.skills_fetcher import SkillsFetcher
from src.database import Database
from datetime import datetime

today = datetime.now().strftime("%Y-%m-%d")
fetcher = SkillsFetcher()
skills = fetcher.fetch()

db = Database()
db.init_db()
db.save_today_data(today, skills)
```

### Data is stale

If data is old, re-run `main_trending.py` or fetch fresh data.

---

## CLI Reference

```bash
# Run full trending analysis
python src/main_trending.py

# Query database
sqlite3 data/trends.db "SELECT * FROM skills_daily WHERE date = '2026-01-23' ORDER BY rank;"
```