Agents Trade — AI skill for Claude Code
Disciplined Claude-assisted intraday options trading system for NIFTY/BANKNIFTY — break detection, trade cards with entry/stop/target, risk levels.
How to install Agents Trade
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open MrSpideyNihal/agents-trade and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Agents Trade does
Disciplined Claude-assisted intraday options trading system for NIFTY/BANKNIFTY — break detection, trade cards with entry/stop/target, risk levels. MCP, works with any AI agent.
Alternatives in AI
- Pal Configuration Docs — for model and thinking mode options 11.3k ★
- Cursor Tools — by eastlondoner - Creates a versatile AI command interface supporting multiple providers and models with flexi 4.7k ★
- Surface — Show ranked attack surface for a target from its recon manifest + hunt memory 3.8k ★
README
angel-trader
A disciplined, **Claude-assisted intraday options trading system** for NIFTY/BANKNIFTY — live index data, break-detection with full trade cards (entry/stop/target/R:R), position management, and hard discipline rules. Claude finds and manages the trade; **you place every order yourself.** Claude never sends a live order.
Built and refined over real trading days. The planned trades were green; the one red day came from chasing. This system exists so that doesn't happen again.
What's in the box
| File | Role |
|---|---|
server.py |
Angel SmartAPI layer: live quotes, option chain, budget-fitted candidates, and a guarded preview→place order flow (MCP server). |
SKILL.md |
The methodology for Claude: the workflow, the TRADE CARD format, and the hard rules (2 trades/day, confirmed break only, no revenge/zero-hero). Upload as a skill. |
signal_engine.py |
Background monitor: scans both indices for a confirmed break, emits a full trade card, and once you're in a position, pings Claude at decision moments with net P&L. Read-only. |
trade.py |
CLI to size / validate / dry-run an order (buy/sell/exit). Not used for live orders. |
.env |
Your Angel API creds + risk caps. Never commit it. |
The hard rules (the actual edge)
- Max 2 trades/day. After trade 2, done — win or lose.
- Proper setup only — entry, stop, target decided before you buy.
- Confirmed break only — a candle closes past the level with momentum. No wicks, no front-running.
- No rush / no FOMO / no revenge. Never fund up or buy a cheap far-OTM "zero-hero" to win a loss back.
- Respect the stop, book the target. No averaging, no flipping, no "a bit more."
- Don't fight a divergence (NIFTY vs BANKNIFTY disagreeing).
- Claude never places live orders — you click buy/sell in the Angel web console yourself.
The edge is discipline, not prediction. Most of the day, the answer is **no trade.**
Setup (PowerShell, in
Related Skills
Sentinel Skill
Risk-aware AI trading copilot for Binance Agent OS — analyzes your portfolio, challenges risky trades before e
Deepstack
Autonomous AI-Powered Trading System with Claude Integration Web App at deepstack.trade
Token Goat Mem
Local-first long-term memory for AI coding agents (Claude Code, Copilot, Codex): durable facts with trust leve
Dash Sales Agent
Agentic AI sales dashboard for General Trade. Cleans messy CSV exports into a DuckDB warehouse, then serves li
Avoid AI Writing
Audits and rewrites content to remove 21 categories of AI writing patterns with a 43-entry replacement table a
Humanise Text Skill
Strip AI writing patterns from any text. Claude Code skill with 507-entry banned word list, structural pattern
Related Agents
Options Strategist
Options strategy specialist for spread design, volatility trading, hedging strategies, and options portfolio m
Desk Lead
Runs the Hyperliquid trading desk. The user's main entry point; routes work to specialists, runs the trade lif
Trade Executor
The only agent allowed to place, modify, or cancel orders. Executes a risk-manager-approved trade plan exactly