Binance Spot Trader

Autonomous Binance spot trading bot with LLM-powered market analysis. Supports momentum trading, mean reversion, and DCA strategies on any Binance spot pair. Use when user wants to trade on Binance, set up automated crypto trading, build a spot trading bot, or automate DCA buying. Features technical analysis, LLM sentiment evaluation, position sizing, and portfolio tracking.

Security Scan

The skill's code, instructions, and required credentials are coherent with a Binance spot trading bot that uses an LLM for sentiment — there are no obvious unrelated permissions, unexpected installs, or hidden endpoints.

Purpose & Capability

Name/description (Binance spot trading with LLM sentiment) match what the files and SKILL.md request and implement: the scripts call Binance REST endpoints and an Anthropic LLM. Required env vars (BINANCE_API_KEY, BINANCE_SECRET_KEY, LLM_API_KEY) are appropriate for the stated purpose.

Instruction Scope

SKILL.md instructs installing dependencies, creating a .env with API keys, and running trader.py (or via cron). The code reads those env vars, writes a local trades.jsonl log, and sends market context to api.anthropic.com — all consistent with the described behavior. Note: the skill persists trade logs and will transmit summary market data to Anthropic (privacy/cost implications).

Install Mechanism

setup.sh uses pip to install httpx and python-dotenv — reasonable, minimal, and from standard packaging. No downloads from unknown hosts or archive extraction are present.

Credentials

Only three env vars are required (Binance API key + secret, LLM API key), which are proportionate to a trading bot that places orders and calls an LLM. No unrelated credentials or broad system paths are requested.

Persistence & Privilege

Skill is not force-included (always:false) and does not request elevated platform privileges. It writes local logs (trades.jsonl, trader.log if run via cron) and does not modify other skills or system-wide settings.

Assessment

This package appears to be what it says: an autonomous Binance spot trader that also calls an Anthropic LLM for sentiment. Before installing: (1) only use Binance API keys with trading enabled and withdrawals explicitly disabled; prefer a sub-account with limited funds; (2) secure the .env (chmod 600, limit server access) and IP-restrict the Binance key; (3) review and test the code locally in paper-trade mode or with tiny amounts first — the bot will place real market orders; (4) be aware that the bot sends recent price/volume data to Anthropic (costs and privacy) and will append trade activity to trades.jsonl on disk; (5) monitor rate limits and broker/account limits to avoid accidental rapid trading; (6) if you need higher assurance, run it in an isolated machine and/or audit the code for edge-case error handling and signature construction. If any of the above conditions are unacceptable (e.g., you cannot restrict API key withdrawals or cannot tolerate external LLM calls), do not install.

Autonomous spot trading bot for Binance. Combines technical indicators with LLM-powered market sentiment analysis to execute trades on any Binance spot pair.

Prerequisites

• Binance account with API keys (spot trading enabled, withdrawal DISABLED)
• Anthropic API key (uses Haiku ~$0.001/eval)
• Python 3.10+

Setup

1. Install

bash {baseDir}/scripts/setup.sh

2. Configure

Create .env:

BINANCE_API_KEY=<your-api-key>
BINANCE_SECRET_KEY=<your-secret-key>
LLM_API_KEY=<anthropic-api-key>
PAIRS=BTCUSDT,ETHUSDT,SOLUSDT
STRATEGY=momentum
TRADE_SIZE_PCT=5
MAX_POSITIONS=5

3. Run

python3 {baseDir}/scripts/trader.py

Or via cron:

*/5 * * * * cd /opt/trader && python3 trader.py >> trader.log 2>&1

Strategies

Momentum (default)

• Buys when price crosses above 20-EMA with volume spike
• Sells when price crosses below 20-EMA or hits TP/SL
• Best for trending markets (BTC, ETH, SOL)

Mean Reversion

• Buys when RSI < 30 (oversold) and price near Bollinger Band lower
• Sells when RSI > 70 (overbought) or price near upper band
• Best for range-bound markets

DCA (Dollar Cost Average)

• Buys fixed amount at regular intervals regardless of price
• Configurable interval (hourly, daily, weekly)
• Lowest risk strategy for long-term accumulation

LLM-Enhanced (all strategies)

• Before each trade, asks Claude Haiku for market sentiment
• Evaluates: recent news, price action, volume patterns, market structure
• Can veto a trade signal if sentiment is strongly against

Trading Parameters

Parameter	Default	Description
PAIRS	BTCUSDT	Comma-separated trading pairs
STRATEGY	momentum	momentum, mean_reversion, or dca
TRADE_SIZE_PCT	5	% of portfolio per trade
MAX_POSITIONS	5	Max concurrent open positions
TAKE_PROFIT_PCT	5	Take profit %
STOP_LOSS_PCT	3	Stop loss %
DCA_INTERVAL	daily	For DCA: hourly, daily, weekly
DCA_AMOUNT_USDT	50	USDT per DCA buy
USE_LLM	true	Enable LLM sentiment filter

Monitoring

# Check portfolio
python3 {baseDir}/scripts/portfolio.py

# View trade history
tail -50 trades.jsonl

# Check logs
tail -f trader.log

⚠️ Security Considerations

• NEVER enable withdrawal on API keys — trading only
• IP-restrict your API keys on Binance
• Use a sub-account with limited funds for bot trading
• Start with tiny amounts ($50-100) and paper trade first
• Monitor actively during first 24 hours
• Set up Binance email alerts for all trades
• API keys on disk — secure your server (SSH keys only, firewall, chmod 600)

References

• See references/binance-api.md for REST API docs
• See references/indicators.md for technical analysis details