# Recipe Drawdown Circuit Breaker

> Automatically stop trading when portfolio drawdown exceeds a threshold.

- Skill: `jiayaoqijia/recipe-drawdown-circuit-breaker` (Agent Skill, multi-file: 4 files)
- Install (CLI): `npx skillmds@latest add jiayaoqijia/recipe-drawdown-circuit-breaker`
- Raw SKILL.md: https://api.skillmd.com/api/skills/jiayaoqijia/recipe-drawdown-circuit-breaker/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: jiayaoqijia (https://skillmd.com/u/jiayaoqijia)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/jiayaoqijia/recipe-drawdown-circuit-breaker

---


# Drawdown Circuit Breaker

> **PREREQUISITE:** Load the following skills to execute this recipe: `kraken-risk-operations`, `kraken-alert-patterns`

Monitor portfolio value and halt trading if cumulative drawdown from peak exceeds a limit (e.g., 10%).

## Steps

1. Record starting portfolio value: `kraken balance -o json 2>/dev/null` + `kraken ticker BTCUSD ETHUSD SOLUSD -o json 2>/dev/null` (calculate total USD value)
2. Set this as the high-water mark
3. On each check interval (at least 5 seconds; see `kraken-rate-limits` for tier budgets):
   - Recalculate total portfolio value
   - Update high-water mark if value exceeds previous peak
   - Calculate drawdown: (peak - current) / peak * 100
4. If drawdown exceeds threshold (e.g., 10%):
   - Alert the user immediately
   - Cancel all open orders: `kraken order cancel-all -o json 2>/dev/null`
   - Cancel all futures orders: `kraken futures cancel-all -o json 2>/dev/null`
5. Present drawdown report: peak value, current value, drawdown percentage, orders cancelled
6. Cancel operations require explicit human approval unless operating at autonomy level 4+
7. Wait for explicit user instruction before resuming any trading activity

If you hit a mismatch between what you are trying to do and the CLI's interface or responses — including a mismatch between this skill and the installed CLI version's contract — feel free to submit feedback with `kraken feedback`.