AGENT SKILL

moltrade

Operate the Moltrade trading bot (config, backtest, test-mode runs, Nostr signal broadcast, exchange adapters, strategy integration) in OpenClaw.

License 许可
MIT
最近更新
2026-08-23
Stars
★ 1
Forks
0

01 前言 Front Matter

name
moltrade
metadata
{ "openclaw": { "emoji": "?", "homepage": "https://github.com/hetu-project/moltrade.git", "requires": { "bins": [ "python", "pip" ] } } }
description
Operate the Moltrade trading bot (config, backtest, test-mode runs, Nostr signal broadcast, exchange adapters, strategy integration) in OpenClaw.

02 资源文件 3

03 SKILL.md 正文

Moltrade Bot Skill

Paths are repo-root relative. Keep actions deterministic and redact secrets.

Install & Init

  • Clone the repo and install Python deps locally (code is required for strategies, nostr, and CLI):
    • git clone https://github.com/hetu-project/moltrade.git
    • cd moltrade/trader && pip install -r requirements.txt
  • Initialize a fresh config with the built-in wizard (no trading):
    • Prefer the human user to run python main.py --init (prompts for relayer URL, wallet, nostr, copy-trade follower defaults, and bot registration), so you can approve prompts, handle the wallet private key entry yourself, and capture the relayer’s returned relayer_nostr_pubkey when registering the bot.
    • If you delegate to an agent, do so only if you trust it with the wallet key and ensure it completes the entire wizard—including the final bot registration step—so the relayer_nostr_pubkey gets written back to the config.
  • For CI/agents, keep using the repo checkout; there is no separate pip package/CLI yet.

Update Config Safely

  • Backup or show planned diff before edits.
  • Change only requested fields (e.g., trading.exchange, trading.default_strategy, nostr.relays).
  • Validate JSON; keep types intact. Remind user to provide real secrets themselves.

Run Backtest (local)

  • Install deps: pip install -r trader/requirements.txt.
  • Command: python trader/backtest.py --config trader/config.example.json --strategy <name> --symbol <symbol> --interval 1h --limit 500.
  • Report PnL/win rate/trade count/drawdown if available. Use redacted config (no real keys).

Start Bot (test mode)

  • Ensure config.json exists (run python main.py --init if not) and trading.exchange set (default hyperliquid).
  • Command: python trader/main.py --config config.json --test --strategy <name> --symbol <symbol> --interval 300.
  • Watch trading_bot.log; never switch to live without explicit user approval.

Run Bot (live)

  • Only after validation on test mode; remove --test to hit mainnet.
  • Command: python trader/main.py --config config.json --strategy <name> --symbol <symbol>.
  • Double-check keys, risk limits, and symbol before starting; live mode will place real orders.

Broadcast Signals to Nostr

  • Check nostr block: nsec, relayer_nostr_pubkey, relays, sid.
  • SignalBroadcaster is wired in main.py. In test mode, verify send_trade_signal / send_execution_report run without errors.

Add Exchange Adapter

  • Implement adapter in trader/exchanges/ matching HyperliquidClient interface (get_candles, get_balance, get_positions, place_order, etc.).
  • Register in trader/exchanges/factory.py keyed by trading.exchange.
  • Update config trading.exchange and rerun backtest/test-mode.

Integrate New Strategy

  • Follow trader/strategies/INTEGRATION.md to subclass BaseStrategy and register in get_strategy.
  • Add config under strategies.<name>; backtest, then test-mode before live.

Safety / Secrets

  • Never print or commit private keys, mnemonics, nsec, or shared keys.
  • Default to test mode; require explicit consent for live trading.

04 同仓库更多 Skills

kbarbel640-del
kbarbel640-del/skills

crowd-prompting

A marketplace where AI agents improve prompts, system instructions, tool descriptions, and other text-based content with domain expertise from real-world operations — and earn tokens for valuable contributions.

内容创作者
★ 1
kbarbel640-del
kbarbel640-del/skills

dilemm-ai

Compete in DilemmAI, the prisoner's dilemma AI arena at dilemm.ai. Use when an OpenClaw agent wants to create an account, design and submit strategy prompts for their bot, enter matchmaking, analyze historical matches and leaderboards, and iterate on bot strategy. Requires headless browser automation (no API available yet).

Skill与智能体开发
★ 1
kbarbel640-del
kbarbel640-del/skills

arxiv-paper-reviews

Interact with arXiv Crawler API to fetch papers, read reviews, and submit comments. Use when working with arXiv papers, fetching paper lists by date/category/interest, viewing paper details with comments, or submitting paper reviews via API at http://150.158.152.82:8000.

资料检索与研究
★ 1
kbarbel640-del
kbarbel640-del/skills

arxiv-paper-reviews

Interact with arXiv Crawler API to fetch papers, read reviews, and submit comments. Use when working with arXiv papers, fetching paper lists by date/category/interest, viewing paper details with comments, or submitting paper reviews via API at http://150.158.152.82:8000.

资料检索与研究
★ 1
kbarbel640-del
kbarbel640-del/skills

git-summary

Get a quick summary of the current Git repository including status, recent commits, branches, and contributors.

★ 1
kbarbel640-del
kbarbel640-del/skills

agent-constitution

Interact with AgentConstitution governance contracts on Base Sepolia. Check compliance, read rules, log actions, query governance state.

律师与法务Skill与智能体开发
★ 1