MCP HubMCP Hub
返回技能列表

generating-trading-signals

jeremylongshore
更新于 Today
80 次查看
712
74
712
在 GitHub 上查看
ai

关于

This Claude Skill generates crypto trading signals by analyzing technical indicators and on-chain metrics. It is triggered by phrases like "get trading signals" and requires configured connections to crypto data APIs and blockchain RPC endpoints. The skill provides actionable alerts for trading decisions.

快速安装

Claude Code

推荐
插件命令推荐
/plugin add https://github.com/jeremylongshore/claude-code-plugins-plus
Git 克隆备选方式
git clone https://github.com/jeremylongshore/claude-code-plugins-plus.git ~/.claude/skills/generating-trading-signals

在 Claude Code 中复制并粘贴此命令以安装该技能

技能文档

Prerequisites

Before using this skill, ensure you have:

  • Access to crypto market data APIs (CoinGecko, CoinMarketCap, or similar)
  • Blockchain RPC endpoints or node access (Infura, Alchemy, or self-hosted)
  • API keys for exchanges if trading or querying account data
  • Web3 libraries installed (ethers.js, web3.py, or equivalent)
  • Understanding of blockchain concepts and crypto market dynamics

Instructions

Step 1: Configure Data Sources

Set up connections to crypto data providers:

  1. Use Read tool to load API credentials from {baseDir}/config/crypto-apis.env
  2. Configure blockchain RPC endpoints for target networks
  3. Set up exchange API connections if required
  4. Verify rate limits and subscription tiers
  5. Test connectivity and authentication

Step 2: Query Crypto Data

Retrieve relevant blockchain and market data:

  1. Use Bash(crypto:signals-*) to execute crypto data queries
  2. Fetch real-time prices, volumes, and market cap data
  3. Query blockchain for on-chain metrics and transactions
  4. Retrieve exchange order book and trade history
  5. Aggregate data from multiple sources for accuracy

Step 3: Analyze and Process

Process crypto data to generate insights:

  • Calculate key metrics (returns, volatility, correlation)
  • Identify patterns and anomalies in data
  • Apply technical indicators or on-chain signals
  • Compare across timeframes and assets
  • Generate actionable insights and alerts

Step 4: Generate Reports

Document findings in {baseDir}/crypto-reports/:

  • Market summary with key price movements
  • Detailed analysis with charts and metrics
  • Trading signals or opportunity recommendations
  • Risk assessment and position sizing guidance
  • Historical context and trend analysis

Output

The skill generates comprehensive crypto analysis:

Market Data

Real-time and historical metrics:

  • Current prices across exchanges with spread analysis
  • 24h volume, market cap, and circulating supply
  • Price changes across multiple timeframes (1h, 24h, 7d, 30d)
  • Trading volume distribution by exchange
  • Liquidity metrics and slippage estimates

On-Chain Metrics

Blockchain-specific analysis:

  • Transaction count and network activity
  • Active addresses and user growth metrics
  • Token holder distribution and concentration
  • Smart contract interactions and DeFi TVL
  • Gas usage and network congestion indicators

Technical Analysis

Trading indicators and signals:

  • Moving averages (SMA, EMA) and trend identification
  • RSI, MACD, Bollinger Bands technical indicators
  • Support and resistance levels
  • Chart patterns and breakout signals
  • Volume profile and accumulation zones

Risk Metrics

Portfolio and position risk assessment:

  • Value at Risk (VaR) calculations
  • Portfolio correlation and diversification metrics
  • Volatility analysis and beta to market
  • Drawdown statistics and recovery times
  • Liquidation risk for leveraged positions

Error Handling

Common issues and solutions:

API Rate Limit Exceeded

  • Error: Too many requests to crypto data API
  • Solution: Implement request throttling; use caching for frequently accessed data; upgrade API tier if needed

Blockchain RPC Errors

  • Error: Cannot connect to blockchain node or timeout
  • Solution: Switch to backup RPC endpoint; verify network connectivity; check if node is synced

Invalid Address or Transaction

  • Error: Blockchain address format invalid or transaction not found
  • Solution: Validate address checksums; verify network (mainnet vs testnet); allow time for transaction confirmation

Exchange API Authentication Failed

  • Error: Invalid API key or signature mismatch
  • Solution: Regenerate API keys; verify permissions (read/trade); check system clock synchronization for signatures

Resources

Crypto Data Providers

  • CoinGecko API for market data across thousands of assets
  • Etherscan API for Ethereum blockchain data
  • Dune Analytics for on-chain SQL queries
  • The Graph for decentralized blockchain indexing

Web3 Libraries

  • ethers.js for Ethereum smart contract interaction
  • web3.py for Python-based blockchain queries
  • viem for TypeScript Web3 development
  • Hardhat for local blockchain testing

Trading and Analysis Tools

  • TradingView for technical analysis and charting
  • Glassnode for advanced on-chain metrics
  • DeFi Llama for DeFi protocol analytics
  • Nansen for wallet tracking and smart money flows

Best Practices

  • Never store private keys or seed phrases in code
  • Always verify smart contract addresses from official sources
  • Use testnet for experimentation before mainnet
  • Implement proper error handling for network failures
  • Monitor gas prices before submitting transactions
  • Validate all user inputs to prevent injection attacks

GitHub 仓库

jeremylongshore/claude-code-plugins-plus
路径: plugins/crypto/crypto-signal-generator/skills/crypto-signal-generator
aiautomationclaude-codedevopsmarketplacemcp

相关推荐技能

sglang

SGLang是一个专为LLM设计的高性能推理框架,特别适用于需要结构化输出的场景。它通过RadixAttention前缀缓存技术,在处理JSON、正则表达式、工具调用等具有重复前缀的复杂工作流时,能实现极速生成。如果你正在构建智能体或多轮对话系统,并追求远超vLLM的推理性能,SGLang是理想选择。

查看技能

evaluating-llms-harness

测试

该Skill通过60+个学术基准测试(如MMLU、GSM8K等)评估大语言模型质量,适用于模型对比、学术研究及训练进度追踪。它支持HuggingFace、vLLM和API接口,被EleutherAI等行业领先机构广泛采用。开发者可通过简单命令行快速对模型进行多任务批量评估。

查看技能

langchain

LangChain是一个用于构建LLM应用程序的框架,支持智能体、链和RAG应用开发。它提供多模型提供商支持、500+工具集成、记忆管理和向量检索等核心功能。开发者可用它快速构建聊天机器人、问答系统和自主代理,适用于从原型验证到生产部署的全流程。

查看技能

llamaguard

其他

LlamaGuard是Meta推出的7-8B参数内容审核模型,专门用于过滤LLM的输入和输出内容。它能检测六大安全风险类别(暴力/仇恨、性内容、武器、违禁品、自残、犯罪计划),准确率达94-95%。开发者可通过HuggingFace、vLLM或Sagemaker快速部署,并能与NeMo Guardrails集成实现自动化安全防护。

查看技能