cobra-strategy
COBRA v1.1 — Arena Sprint Predator. Trades the single most dominant SM asset with maximum conviction. Fee-optimized maker orders. Wider DSL lets positions breathe. Designed to win Arena weeks.
- License 许可
- MIT
- 最近更新
- 2026-07-02
- Stars
- ★ 1
- Forks
- 0
01 前言 Front Matter
- name
- cobra-strategy
- license
- MIT
- metadata
- { "author": "jason-goldberg", "version": "1.1", "exchange": "hyperliquid", "platform": "senpi", "requires": [ "senpi-trading-runtime" ] }
- description
- COBRA v1.1 — Arena Sprint Predator. Trades the single most dominant SM asset with maximum conviction. Fee-optimized maker orders. Wider DSL lets positions breathe. Designed to win Arena weeks.
02 资源文件 3
- 参考文档README.md1.1 KB
- 参考文档runtime.yaml2.0 KB
- 脚本scripts/cobra-scanner.py16.6 KB
03 SKILL.md 正文
? COBRA v1.1 — Arena Sprint Predator
SM dominates. Cobra strikes. One asset. Maximum conviction.
What Makes Cobra Different
Every other Predator agent is built for long-term edge — patient entry, perfect scores, wide DSL. Cobra is built to win Arena weeks.
The Arena rewards ROE% over 7 days. The winning strategy is 2-4 massive conviction trades that ride the dominant SM trend. The #1 Arena agent won with +62% ROE on just 12 trades.
Fee Lesson from the Fleet
The Predators data tells a clear story about fees:
| Agent | Trades | Fees | Profit | Profit:Fee Ratio |
|---|---|---|---|---|
| Condor v2.0 | 5 | $2.71 | +$37.43 | 13.8:1 |
| Cheetah | 53 | $23.12 | +$40.58 | 1.8:1 |
| Polar | 170 | $239.27 | +$38.65 | 0.16:1 |
Cobra targets Condor's efficiency: few trades, big impact, minimal fees.
Cobra's approach:
- Only trades the #1 SM asset — never scatters across multiple assets
- $400 margin per trade (40% of budget) for concentrated ROE impact
- FEE_OPTIMIZED_LIMIT orders — maker first, taker fallback at 30s. ~50% fee savings vs market orders.
- Wide DSL — 180min timeout, 8% retrace, 45min dead weight. Let positions breathe.
- 3 entries per day max — quality over quantity
- 10x leverage on majors (BTC/ETH/SOL/HYPE), 5x on others
⛔ CRITICAL AGENT RULES
RULE 1: Install path is /data/workspace/skills/cobra-strategy/
RULE 2: THE SCANNER DOES NOT EXIT POSITIONS — DSL only.
RULE 3: MAX 1 POSITION at a time.
RULE 4: MAX 3 ENTRIES PER DAY. 90-minute cooldown. 120-minute per-asset cooldown.
RULE 5: ONLY trade the #1 SM dominant asset. Never scatter.
RULE 6: XYZ equities are BANNED at parse level.
RULE 7: USE FEE_OPTIMIZED_LIMIT for all entries. Never market orders.
Scanner Logic (2 API calls)
1. leaderboard_get_markets (limit=20)
→ Find #1 non-XYZ asset by SM concentration
→ Gate: Must be >10% SM dominance
2. Score the dominant asset:
- SM Dominance: >15% = +3, >10% = +2, >5% = +1
- Contribution Surge: >5% = +2, >2% = +1
- 4H Trend Confirms: >0.5% in direction = +1
- 1H Trend Confirms: >0.2% in direction = +1
- Deep Consensus: >200 traders = +1
MIN_SCORE: 5/8
→ If score >= 5: ENTER with $400 margin at 10x (majors) or 5x
→ Order type: FEE_OPTIMIZED_LIMIT (maker first, taker fallback 30s)
DSL Configuration (Fee-Conscious, Position-Breathing)
Phase 1 (Loss Protection):
- Max loss: -15%
- Retrace from high water: 8% (wide — let volatile assets breathe)
- 3 consecutive breaches required
- Phase 1 max time: 45 minutes
Phase 2 (Profit Locking — let winners run):
- +5% → lock 25% of high water
- +10% → lock 45%
- +15% → lock 60%
- +20% → lock 75%
- +30% → lock 85%
- +50% → lock 92%
Timeouts:
- Hard timeout: 180 minutes (3 hours — positions need time)
- Weak peak cut: 60 minutes at +3% minimum
- Dead weight cut: 45 minutes
Why This Wins Arena
-
Concentration: $400 on one trade. A 10% move at 10x = +$400 = 40% ROE on the $1K account.
-
Fee efficiency: FEE_OPTIMIZED_LIMIT saves ~50% vs market orders. Over a week, that's the difference between +5% and +8% ROE.
-
Breathing room: 180-minute timeout and 8% retrace means positions ride through volatility instead of getting stopped out on noise and re-entering (which churns fees).
-
SM alignment: Only trading when SM dominance is extreme (>10%). When 300+ top traders are positioned the same way, the trend usually has hours to develop — not minutes.
-
Fewer trades: 3/day max, 90-min cooldown. Over 7 days that's 21 max entries. Only need 2-3 big winners.
Execution Details (Scanner-Level, Not YAML)
Entry: FEE_OPTIMIZED_LIMIT with ensureExecutionAsTaker: true and executionTimeoutSeconds: 30
- Places an ALO (Add Liquidity Only) maker order
- If not filled in 30 seconds, cancels and resubmits as market order
- Maker fees are ~0.02% vs taker fees of ~0.05%
- On a $400 × 10x = $4,000 notional position, this saves ~$1.20 per trade
Runtime Setup
sed -i 's/${WALLET_ADDRESS}/<WALLET>/' /data/workspace/skills/cobra-strategy/runtime.yaml
sed -i 's/${TELEGRAM_CHAT_ID}/<CHAT_ID>/' /data/workspace/skills/cobra-strategy/runtime.yaml
openclaw senpi runtime create --path /data/workspace/skills/cobra-strategy/runtime.yaml
openclaw senpi runtime list && openclaw senpi status
Bootstrap Gate
Verify runtime + status + scanner cron (90s, main).
CRITICAL: The cron must be configured to call create_position via Senpi MCP when the scanner outputs a signal with an entry block. If the cron just runs the scanner and exits, trades will never execute.
Send: "? COBRA v1.1 online. Arena predator. Fee-optimized. Hunting the #1 SM rotation."
License
MIT — Built by Senpi (https://senpi.ai).
04 同仓库更多 Skills
supermemory
Supermemory is a state-of-the-art memory and context infrastructure for AI agents. Use this skill when building applications that need persistent memory, user personalization, long-term context retention, or semantic search across knowledge bases. It provides Memory API for learned user context, User Profiles for static/dynamic facts, and RAG for semantic search. Perfect for chatbots, assistants, and knowledge-intensive applications.
wolverine-strategy
WOLVERINE v2.0 — HYPE alpha hunter. Entry-only scanner, DSL-exit-only architecture. v1.1 lost -22.7% because the scanner's thesis exit chopped 25/27 trades before DSL could manage them. v2.0 removes thesis exit entirely. Scanner decides entries (score 8+, 4H/1H aligned, SM consensus). DSL manages all exits (wide Phase 1 for HYPE volatility, trailing tiers starting at +15% ROE). Leverage lowered to 7x. Max 4 entries/day. 3-hour cooldown between entries. DSL exit managed by plugin runtime via runtime.yaml.
wolf-strategy
WOLF v6.3 — Fully autonomous multi-strategy trading for Hyperliquid perps via Senpi MCP. Manages multiple strategies simultaneously, each with independent wallets, budgets, slots, and DSL configs. 5+N cron architecture: 5 shared wolf crons (Emerging Movers 3min, SM Flip 5min, Watchdog 5min, Risk Guardian 5min, Health Check 10min) plus one DSL v5.2 cron per strategy (native Hyperliquid SL sync via dsl-dynamic-stop-loss skill v5.3.1). Same asset can be traded in different strategies simultaneously. Enter early on first jumps, not at confirmed peaks. Dynamic risk-based leverage per strategy. Requires Senpi MCP connection, python3, mcporter CLI, OpenClaw cron system, and dsl-dynamic-stop-loss skill (provides dsl-cli.py + dsl-v5.py).
wolf-howl
HOWL v2 — Hunt, Optimize, Win, Learn. Nightly self-improvement loop for the WOLF autonomous trading strategy. Runs once per day (via cron) to review all trades from the last 24 hours, compute win rates, analyze signal quality correlation, evaluate DSL tier performance, identify missed opportunities, and produce concrete improvement suggestions for the wolf-strategy skill. v2 adds fee drag ratio (FDR) analysis, holding period bucketing, LONG vs SHORT regime detection, rotation cost tracking, cumulative drift detection, and gross vs net profit factor separation. Use when setting up daily trade review automation, analyzing trading performance, or improving an autonomous trading strategy through data-driven feedback loops. Requires Senpi MCP connection, mcporter CLI, and OpenClaw cron system.
whale-index
Auto-mirror top Discovery traders on Hyperliquid. Scans top 50 traders, scores on PnL rank (35%), win rate (25%), consistency (20%), hold time (10%), drawdown (10%). Creates 2-5 mirror strategies with overlap checks. Daily rebalance with 2-day watch period before swaps. Use when setting up trader mirroring, copy trading, or portfolio auto-rebalancing based on Discovery leaderboard performance.
vixen-strategy
VIXEN v1.0 — Dual-mode emerging movers scanner. Built from FOX v1.6 live trading data (+34.5% ROI). Two entry modes: STALKER (steady accumulation, score 6+, catches the SM buildup BEFORE the explosion) and STRIKER (violent FIRST_JUMP, score 9+, with raw volume confirmation to filter blow-off tops). 2-hour per-asset cooldown after Phase 1 exits. DSL High Water Mode mandatory.