← 返回 Skills 市场
Quantumexecute
作者
quantumexecute
· GitHub ↗
· v1.0.1
· MIT-0
67
总下载
0
收藏
0
当前安装
2
版本数
在 OpenClaw 中安装
/install quantumexecute-skill
功能描述
QuantumExecute (QE) unified skill for AI-driven cryptocurrency algorithmic trading across Binance, OKX, LTP, Deribit, Hyperliquid, and other supported exchan...
安全使用建议
This skill appears coherent for its stated purpose, but it controls high‑privilege trading actions. Before installing or running it: 1) Treat QE_API_KEY/QE_API_SECRET as powerful credentials — create and use restricted keys (minimal permissions) and enable IP whitelisting and audit logs on the QuantumExecute platform. 2) Review and install dependencies in an isolated environment (virtualenv/container) and verify the 'qe-connector' package source & version before installing. 3) Confirm the agent enforces the skill's P0 hard rules (explicit confirmation required for create/update/cancel operations). 4) Test read-only scripts first (list_exchange_apis, get_market_data, balance/positions) using non-production accounts or read-only keys. 5) Note small inconsistencies: SKILL.md references QE_WORKSPACE (not read by scripts) and QE_BASE_URL (supported), and there is no automatic install step — ensure you know where reports will be written (default ~/workspace) and supply QE_BASE_URL only to trusted endpoints. If you need higher assurance, request the upstream package repository (qe-connector) and compare its API usage to the code here, or run the skill in an isolated sandbox before granting production credentials.
功能分析
Type: OpenClaw Skill
Name: quantumexecute-skill
Version: 1.0.1
The QuantumExecute skill bundle is a legitimate integration for cryptocurrency algorithmic trading across multiple exchanges (Binance, OKX, etc.). The scripts facilitate order management, balance queries, and TCA reporting using the `qe-connector` library. The bundle includes extensive safety documentation (references/safety.md) and anti-fabrication rules (references/errors.md) to guide the AI agent's behavior. While scripts like `notify_order_complete.py` possess high-privilege capabilities such as webhook notifications and local command execution, these are documented features intended for order status updates and do not exhibit signs of malicious intent or unauthorized data exfiltration.
能力标签
能力评估
Purpose & Capability
Name/description (multi-exchange trading, order lifecycle, TCA/export) matches the included scripts and references. Requested binaries (python3) and required env vars (QE_API_KEY, QE_API_SECRET) are expected for a Python-based trading client. The set of scripts map directly to claimed features (balances, orders, TCA, exports).
Instruction Scope
SKILL.md directs running the provided scripts and keeping secrets in env vars only; scripts source QE_API_KEY/SECRET via shared _client.get_client, and the skill's hard rules require explicit confirmation before write operations. Scripts write reports to ~/workspace by default and produce JSON outputs. Minor inconsistencies: SKILL.md mentions optional QE_WORKSPACE and QE_BASE_URL, but most scripts do not read QE_WORKSPACE (they default to Path.home()/"workspace"); QE_BASE_URL is read by _client if set. Overall instructions do not attempt to read unrelated system files or exfiltrate secrets.
Install Mechanism
No install spec provided (instruction-only install), but a requirements.txt is included listing 'qe-connector', 'requests', 'pandas', 'openpyxl'. Because there is no automatic install step, operator must install dependencies manually; this is lower risk than an arbitrary download, but operators should verify package sources (especially 'qe-connector') and pin versions before installing.
Credentials
Only QE_API_KEY and QE_API_SECRET are required (declared primaryEnv QE_API_KEY). Those are exactly the high‑privilege credentials needed for the described trading capabilities. SKILL.md also documents QE_BASE_URL and QE_WORKSPACE as optional; QE_BASE_URL is supported by code, QE_WORKSPACE is documented but not consistently used by scripts (they default to ~/workspace). No unrelated credentials or system config paths are requested.
Persistence & Privilege
always:false (no forced inclusion). The skill does not request system-wide changes, does not modify other skills, and does not require persistent elevated privileges. Autonomous invocation is allowed (platform default); the skill's hard rules require explicit confirmations for write operations which mitigates autonomous write risk if enforced by the agent.
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install quantumexecute-skill - 安装完成后，直接呼叫该 Skill 的名称或使用
/quantumexecute-skill触发 - 根据 Skill 的参数说明提供必要输入，即可获得结构化输出
版本历史
v1.0.1
- License updated from MIT to MIT-0.
- Added metadata for OpenClaw environment, including env/binary requirements and homepage link.
- New section "Capability and Credential Declaration" describing credential usage and security recommendations.
- Clarified requirements for QE_API_KEY and QE_API_SECRET, with additional notes on privilege and endpoint safety.
- No functional or script changes; documentation and metadata only.
v1.0.0
Initial release of the quantumexecute unified skill for AI-driven crypto trading workflow.
- Supports algorithmic execution across Binance, OKX, LTP, Deribit, Hyperliquid, and more via scripts.
- Provides smart order splitting, AI algorithms for large-order/quantitative trading, and multi-exchange balance/position visibility.
- Includes order lifecycle management, customizable trading strategies, plug-and-play workflows, and transaction cost analysis (TCA) tools.
- Offers clear onboarding steps for new users and detailed environment setup instructions.
- Enforces strict safety, error reporting, risk checks, and anti-fabrication rules in all workflows.
元数据
常见问题
Quantumexecute 是什么？
QuantumExecute (QE) unified skill for AI-driven cryptocurrency algorithmic trading across Binance, OKX, LTP, Deribit, Hyperliquid, and other supported exchan... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 67 次。
如何安装 Quantumexecute？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install quantumexecute-skill」即可一键安装，无需额外配置。
Quantumexecute 是免费的吗？
是的，Quantumexecute 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Quantumexecute 支持哪些平台？
Quantumexecute 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。
谁开发了 Quantumexecute？
由 quantumexecute（@l-kai890）开发并维护，当前版本 v1.0.1。
推荐 Skills