Agent Skills › HammerGPT/Hyper-Alpha-Arena › trader-diagnosis

trader-diagnosis

GitHub

用于诊断AI交易员未触发、未执行或行为异常的问题。通过检查环境、数据基础、运行状态及配置，定位根因并提供修复方案或操作指引。

backend/skills/trader-diagnosis/SKILL.md HammerGPT/Hyper-Alpha-Arena

Trigger Scenarios

用户报告交易员不工作策略未执行没有发生交易交易员未触发为什么没有信号

Install

npx skills add HammerGPT/Hyper-Alpha-Arena --skill trader-diagnosis -g -y

More Options

Non-standard path

npx skills add https://github.com/HammerGPT/Hyper-Alpha-Arena/tree/main/backend/skills/trader-diagnosis -g -y

Use without installing

npx skills use HammerGPT/Hyper-Alpha-Arena@trader-diagnosis

指定 Agent (Claude Code)

npx skills add HammerGPT/Hyper-Alpha-Arena --skill trader-diagnosis -a claude-code -g -y

安装 repo 全部 skill

npx skills add HammerGPT/Hyper-Alpha-Arena --all -g -y

预览 repo 内 skill

npx skills add HammerGPT/Hyper-Alpha-Arena --list

SKILL.md

Frontmatter

{
    "name": "trader-diagnosis",
    "shortcut": "diagnose",
    "description": "This skill should be used when the user reports that their AI Trader is not triggering, not executing trades, or behaving unexpectedly. Trigger phrases include \"why isn't my trader working\", \"strategy not executing\", \"no trades happening\", \"trader not triggering\", \"why no signals\".",
    "description_zh": "当用户反馈 AI Trader 没有触发、没有执行交易或行为异常时使用此技能进行系统化诊断。"
}

Trader Diagnosis

Systematically diagnose why an AI Trader is not triggering or executing trades as expected. Follow a structured checklist approach.

Pre-requisites (MUST confirm before proceeding)

Identify WHICH trader has the issue (if user doesn't specify, list_traders and ask)
Confirm the exchange and environment the trader is on

Workflow

Phase 0: Environment and Data Foundation Check

Before diagnosing the trader itself, verify the system foundation:

Check trading environment: get_trading_environment()
- Confirm it matches what user expects (testnet vs mainnet)
- If mismatch, guide user to switch via UI top-right mode switcher
Check watchlist: get_watchlist()
- Verify the symbol(s) the trader's signal pools monitor are in the watchlist
- If using default config (only BTC), warn user that other symbols have no data
- If target symbol missing, use update_watchlist() to add it (after user confirmation)

→ [CHECKPOINT] Report environment and watchlist status. If issues found here, they are ROOT CAUSES — fix them first before proceeding.

Phase 1: Full Diagnostic Scan

Run comprehensive diagnosis: diagnose_trader_issues(trader_id)

This checks:

Trader enabled status
Wallet binding and balance
Strategy binding (prompt or program)
Signal pool configuration and recent triggers
Cooldown timer status
Recent error logs

→ [CHECKPOINT] Present diagnostic results in plain language. Categorize issues found.

Phase 2: Issue Resolution

Based on diagnostic results, address each issue:

Configuration Issues (you can fix):

Signal pool not bound → help bind one (confirm exchange match)
Trigger interval too long → suggest adjustment
Signal thresholds too strict → delegate to Signal AI for recalibration

Manual Operations (guide user to UI):

Wallet not bound → AI Trader → click trader → bind wallet (Hyperliquid: API Wallet key + master address; Binance: API key + secret)
Trading not started → AI Trader → click trader → "Start Trading"
Program binding not active → Programs → "Program Bindings" → Edit → activation switch

Data Issues (investigate further):

API errors → check get_system_logs for details
Balance insufficient → advise deposit to the master wallet address (for Hyperliquid API Wallet)
Exchange connectivity → verify API key / API Wallet status

→ [CHECKPOINT] Summarize fixes applied and remaining manual steps. Wait for user to complete manual actions.

Phase 3: Verification

After user completes manual steps:

Re-run diagnose_trader_issues(trader_id) to verify all clear
Check list_traders(trader_id) for current status
Confirm the trader is now ready to execute

Key Rules

Always run the full diagnostic first, don't guess
Signal pool exchange must match trader's wallet exchange
Be specific about UI paths for manual operations
If the issue is "signal triggered but AI decided HOLD", explain this is normal behavior