← Back to Skills Marketplace
csyr

crypto briefing

by rocinsky · GitHub ↗ · v1.0.0
cross-platform ⚠ suspicious
385
Downloads
0
Stars
1
Active Installs
1
Versions
Install in OpenClaw
/install crypto-briefing
Description
Generate crypto market briefing reports with latest news, price analysis, trend predictions, sentiment analysis, and Fear & Greed Index. Use when user asks f...
README (SKILL.md)

Crypto Briefing Skill

Generate comprehensive crypto market briefing reports by browsing news sites, collecting price data, and analyzing market sentiment.


⚠️ MANDATORY: News Source Checklist

You MUST visit ALL 5 news sources. This is NON-NEGOTIABLE.

Before generating the report, verify you have visited:

  • foresightnews.pro/news
  • panewslab.com/zh/newsflash
  • theblockbeats.info/newsflash
  • techflowpost.com/newsletter/index.html
  • odaily.news/newsflash

If any source is skipped, the report is INCOMPLETE and INVALID.

Do NOT rationalize skipping sources. Do NOT assume one source "has enough news." Each source has unique coverage. Completeness > Speed.

At the end of data collection, you must be able to confirm:

✅ foresightnews.pro - visited
✅ panewslab.com - visited
✅ theblockbeats.info - visited
✅ techflowpost.com - visited
✅ odaily.news - visited

If any shows ❌, go back and visit it before proceeding.


Workflow

Step 1: Collect Price Data

Use browser to fetch from CoinMarketCap homepage (preferred - gets all data in one page):

This provides:

  • Current prices and 24h changes for major coins
  • Fear & Greed Index (official)
  • Market Cap
  • Average Crypto RSI
  • Altcoin Season Index

For BGB specifically, also visit:

Coin mapping:

  • BTC → Bitcoin
  • ETH → Ethereum
  • SOL → Solana
  • SUI → Sui
  • WLD → Worldcoin
  • WIF → dogwifhat
  • BGB → Bitget Token

Step 2: Browse ALL News Sites (MANDATORY)

Use browser tool with profile=openclaw. Visit EACH site and extract headlines from last 12 hours.

Required sites (must visit ALL 5):

# Site URL
1 Foresight News https://foresightnews.pro/news
2 PANews https://www.panewslab.com/zh/newsflash
3 BlockBeats https://www.theblockbeats.info/newsflash
4 TechFlow https://www.techflowpost.com/newsletter/index.html
5 Odaily https://www.odaily.news/newsflash

For each site:

  1. browser action=navigate to the URL
  2. browser action=snapshot with maxChars=15000
  3. Parse headlines and timestamps
  4. Extract news from last 12 hours
  5. Move to next site

After visiting all 5 sites, close the browser tab.

Step 3: Calculate Fear & Greed Index

Use the official Fear & Greed Index from CoinMarketCap (collected in Step 1).

Also analyze the following factors to provide AI analysis context:

Factors to consider:

  1. Price Action (40% weight)

    • 24h price changes across major coins
    • Volatility levels
    • Support/resistance breaks
  2. News Sentiment (35% weight)

    • Ratio of positive/negative headlines from ALL 5 sources
    • Severity of negative news (hacks, bans, crashes)
    • Strength of positive news (adoption, institutional buying)
  3. Market Volume & Liquidity (15% weight)

    • ETF flows (inflows = greed, outflows = fear)
    • Exchange volume trends
  4. Social Sentiment (10% weight)

    • Panic indicators in news language
    • Regulatory concerns
    • Market commentary tone

Classification:

  • 0-25: Extreme Fear
  • 25-45: Fear
  • 45-55: Neutral
  • 55-75: Greed
  • 75-100: Extreme Greed

Step 4: Analyze & Generate Report

Based on collected news (from ALL 5 sources) and price data, generate the briefing report:

# 🪙 加密货币市场简报
> 生成时间: {timestamp} (北京时间)

## 📊 价格概览

| 币种 | 当前价格 | 24h 变化 | 12h 目标价 | 趋势预测 |
|------|----------|----------|------------|----------|
| BTC  | $XX,XXX  | +X.XX%   | $XX,XXX    | 📈 上涨 (XX%) |
| ETH  | $X,XXX   | +X.XX%   | $X,XXX     | 📉 下跌 (XX%) |
| SOL  | $XXX     | +X.XX%   | $XXX       | ➡️ 横盘 (XX%) |
| SUI  | $X.XX    | +X.XX%   | $X.XX      | 📈 上涨 (XX%) |
| WLD  | $X.XX    | +X.XX%   | $X.XX      | 📉 下跌 (XX%) |
| WIF  | $X.XX    | +X.XX%   | $X.XX      | 📈 上涨 (XX%) |
| BGB  | $X.XX    | +X.XX%   | $X.XX      | 📈 上涨 (XX%) |

## 🎭 市场情绪

- 😊 积极: XX%
- 😐 中性: XX%
- 😟 消极: XX%

**综合判断**: {积极/中性/消极}

## 😱 恐慌与贪婪指数

**{value}** - {classification}

[██████████░░░░░░░░░░] {value}/100

**AI 分析依据:**
- {list key indicators from news and data that influenced the F&G score}
- {example: 24小时全网爆仓 XX 亿美元(多单 XX 亿美元)}
- {example: BTC/ETH ETF 连续 X 日净流出共约 XX 亿美元}
- {example: 主流币种全线承压/反弹}
- {example: 重大事件影响(如监管、黑客攻击、机构动向)}

## 📰 重要资讯 (过去 12 小时)

### 利好消息 📈
- 🏦 **{category}**: {headline and details}
- 💰 **{category}**: {headline and details}
- 🚀 **{category}**: {headline and details}

### 利空消息 📉
- 📉 **{category}**: {headline and details}
- 💥 **{category}**: {headline and details}
- ⚠️ **{category}**: {headline and details}

### 其他动态 📋
- 🎯 **{category}**: {headline and details}
- 🔓 **{category}**: {headline and details}

---

## 💡 分析师观点

**短期展望**: {对市场短期(12-24h)走势的综合判断}

**关键支撑位**:
- BTC: $XX,XXX - $XX,XXX (描述)
- ETH: $X,XXX - $X,XXX (描述)

**积极信号**:
1. {正面因素 1}
2. {正面因素 2}
3. {正面因素 3}

**风险因素**:
1. {风险因素 1}
2. {风险因素 2}
3. {风险因素 3}

**策略建议**: {具体操作建议,包括仓位、风险管理等}

**今日关注**:
- {关键事件 1}
- {关键事件 2}
- {关键事件 3}

---

## 📈 技术分析

**比特币 (BTC)**
- 支撑位: $XX,XXX / $XX,XXX
- 阻力位: $XX,XXX / $XX,XXX
- RSI: {超买/中性/超卖}

**以太坊 (ETH)**
- 支撑位: $X,XXX / $X,XXX
- 阻力位: $X,XXX / $X,XXX
- 相对 BTC 表现: {强势/弱势/同步}

---

*数据来源: CoinMarketCap, Foresight News, PANews, BlockBeats, TechFlow, Odaily*  
*免责声明: 本简报仅供参考,不构成投资建议。加密货币市场波动巨大,请谨慎决策。*

Emoji Usage Guidelines

利好消息: 🏦 🏛️ 🔐 🌐 💰 🎯 🚀 ✅ 📈 利空消息: 📉 💥 🐋 ⚖️ 🚨 ⚠️ ❌ 📊 其他动态: 🎯 🔓 📦 🔥 🌐 🔒 💼 🇰🇿 🇸🇻 (country flags for regional news)


Trend Prediction Guidelines

Analyze news sentiment and price momentum to predict 12h trends:

上涨信号 (📈):

  • Positive news (ETF approvals, institutional buying, regulatory clarity)
  • Strong 24h momentum (>3% gain)
  • Low fear index (\x3C30) with accumulation signs

下跌信号 (📉):

  • Negative news (hacks, regulatory crackdowns, large sell-offs)
  • Weak momentum (\x3C-3% loss)
  • High fear or extreme greed (contrarian)

横盘信号 (➡️):

  • Mixed or no significant news
  • Low volatility (-3% to +3%)
  • Neutral sentiment

Sentiment Analysis

Categorize each news headline as:

  • 积极: Bullish news, adoption, partnerships, upgrades
  • 消极: Hacks, bans, crashes, lawsuits
  • 中性: Neutral updates, technical changes, mixed news

Calculate percentages based on headline count across ALL sources.


Final Checklist Before Submitting Report

Before outputting the report, confirm:

  • CoinMarketCap data collected (prices, F&G, RSI)
  • ✅ foresightnews.pro - visited
  • ✅ panewslab.com - visited
  • ✅ theblockbeats.info - visited
  • ✅ techflowpost.com - visited
  • ✅ odaily.news - visited
  • News from all sources merged and deduplicated
  • Report includes data sources footer mentioning all 5 news sites

If any checkbox is unchecked, DO NOT submit the report. Go back and complete the missing steps.

Usage Guidance
This skill appears to do what it says (collect prices, Fear & Greed index, and news), but take precautions before installing: 1) Review and vet the five mandatory news domains — if you don't trust them, do not allow the browser to visit them. 2) Ensure the browser profile used (profile=openclaw) is isolated and does not contain personal cookies, saved logins, or sensitive headers, since the skill requires navigating third-party sites. 3) Note the SKILL.md prefers CoinMarketCap but the included script uses CoinGecko and alternative.me — confirm which data source you want and consider editing the script if necessary. 4) Be aware that visiting external sites can leak IP and other metadata; run the skill in a sandboxed environment or behind a proxy if privacy is a concern. 5) If you need stronger assurances, request the author to (a) make the list of news sources configurable, (b) remove the absolute 'MUST visit ALL 5' requirement, and (c) document any risk mitigations (robots.txt, timeouts, sanitization).
Capability Analysis
Type: OpenClaw Skill Name: crypto-briefing Version: 1.0.0 The OpenClaw skill bundle is designed to generate crypto market briefing reports. All network requests in `SKILL.md` and `scripts/fetch_data.py` are directed to legitimate and well-known cryptocurrency data sources (CoinMarketCap, CoinGecko, alternative.me) and reputable crypto news sites. The `SKILL.md` instructions, while strict about data collection, are focused on ensuring a comprehensive report from specified sources and do not contain any prompt injection attempts to subvert the agent, exfiltrate data, or perform unauthorized actions. The Python script is a straightforward data fetcher with no malicious capabilities.
Capability Assessment
Purpose & Capability
Name and description match the behavior: fetching prices, F&G index, news and producing a briefing. Minor inconsistency: SKILL.md instructs the agent to prefer CoinMarketCap for price and F&G, while the included script fetches prices from CoinGecko and F&G from alternative.me. Both are plausible data sources, but the mismatch should be clarified.
Instruction Scope
SKILL.md mandates visiting five specific external news sites (foresightnews.pro, panewslab.com, theblockbeats.info, techflowpost.com, odaily.news) and insists none may be skipped. This is within the stated purpose (collecting news) but forces the agent to navigate arbitrary third‑party domains with a specific browser profile (profile=openclaw). That rigid 'MUST visit ALL 5' requirement increases risk: visiting untrusted sites can leak IP, headers, or profile cookies and may expose the agent to malicious content. The instructions do not mention respecting robots.txt, rate limits, or validating content authenticity.
Install Mechanism
No install spec; skill is instruction-driven with a small helper script. No packages are downloaded or written to disk by an installer, which is lower risk.
Credentials
Skill does not request environment variables or credentials (proportionate). However, it requires using a browser with profile=openclaw, which could expose stored cookies, auth tokens, or identifying headers if that profile is not isolated. The included script performs outbound network requests to public APIs (CoinGecko, alternative.me) — expected, but network access is required.
Persistence & Privilege
Skill does not request always:true or any elevated persistence. It is user-invocable and does not declare system-wide changes or privileged modifications.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install crypto-briefing
  3. After installation, invoke the skill by name or use /crypto-briefing
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Initial release: Generates comprehensive crypto market briefing reports with multi-source news, price, and sentiment analysis. - Collects real-time price data and market indices from CoinMarketCap. - Mandates news aggregation from 5 designated crypto news sites for complete coverage. - Integrates official Fear & Greed Index with multi-factor AI sentiment analysis. - Provides clear structure for bullish/bearish/neutral forecasting, technicals, and analyst commentary. - Features explicit step-by-step workflow and strict checklist to ensure report validity and completeness.
Metadata
Slug crypto-briefing
Version 1.0.0
License
All-time Installs 1
Active Installs 1
Total Versions 1
Frequently Asked Questions

What is crypto briefing?

Generate crypto market briefing reports with latest news, price analysis, trend predictions, sentiment analysis, and Fear & Greed Index. Use when user asks f... It is an AI Agent Skill for Claude Code / OpenClaw, with 385 downloads so far.

How do I install crypto briefing?

Run "/install crypto-briefing" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is crypto briefing free?

Yes, crypto briefing is completely free (open-source). You can download, install and use it at no cost.

Which platforms does crypto briefing support?

crypto briefing is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created crypto briefing?

It is built and maintained by rocinsky (@csyr); the current version is v1.0.0.

💬 Comments