← Back to Skills Marketplace
448
Downloads
1
Stars
0
Active Installs
4
Versions
Install in OpenClaw
/install ptrade
Description
Ptrade 恒生量化交易平台 - 策略运行在券商服务器上，低延迟执行，支持A股、期货、融资融券等中国证券市场。
Usage Guidance
This skill appears to be a documentation/instruction bundle for writing strategies that run on a broker-managed Ptrade server and does not itself request credentials or install remote code. Before installing: (1) verify the homepage/maintainer (there are minor metadata inconsistencies across files — e.g., version and ownerId differ between manifests), (2) understand that trading actions happen only when code runs in your broker's environment—do not supply broker API keys to this skill unless you intentionally want the agent to act on your account, (3) if you run the demo locally, only install packages from trusted sources (requirements.txt lists pandas/numpy), and (4) review and test strategies in a sandbox/backtest environment to avoid financial loss. If you need stronger assurance, ask the maintainer for a canonical repository/source and a signed release or compare the skill contents with official Ptrade documentation.
Capability Analysis
Type: OpenClaw Skill
Name: ptrade
Version: 1.0.3
The ptrade skill bundle is a comprehensive documentation and reference guide for the HengSheng Ptrade quantitative trading platform. It provides detailed API descriptions, strategy lifecycle events (initialize, handle_data), and numerous code examples for trading A-shares, futures, and convertible bonds. The bundle contains no evidence of malicious intent, data exfiltration, or harmful prompt injection; it explicitly notes that the target environment (broker intranet) lacks external network access. All functions, including notification utilities like send_email and persistence via pickle, are standard features for the described trading platform.
Capability Assessment
Purpose & Capability
Name/description (Ptrade trading platform) align with the provided SKILL.md, API examples, and demo. Required binary is only python3, which is expected for a Python-based strategy framework. The skill does not request unrelated cloud or system credentials.
Instruction Scope
SKILL.md contains event-driven strategy instructions and API usage for placing orders, scheduling, and querying market/portfolio state. It does not instruct the agent to read arbitrary system files, extract environment variables, or transmit data to unexpected external endpoints. It repeatedly states strategies run on the broker's server (no external pip install in that environment).
Install Mechanism
There is no install spec (instruction-only) which minimizes risk. A requirements.txt is included for local demo/development (pandas, numpy) — this is reasonable for local analysis but not used by the broker runtime. No downloads from untrusted URLs or archive extraction are present.
Credentials
The skill declares no required environment variables, credentials, or config paths. That is coherent because the strategy is intended to run on broker servers where the broker client provides access. There is no unexplained request for secrets.
Persistence & Privilege
The skill is not always-enabled and is user-invocable (normal). It does not request modifications to other skills or system-wide agent settings.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install ptrade - After installation, invoke the skill by name or use
/ptrade - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.3
- Added a new demo project directory with a sample README.md and demo.py script.
- No changes to main functionality; only new example/demo files introduced.
v1.0.2
Version 1.2.0 includes a major documentation language update.
- Documentation in SKILL.md switched from English to Simplified Chinese for all main sections and function explanations.
- Content structure, examples, and technical usage remain unchanged.
- No code or functional interface changes; only documentation has been localized.
v1.0.1
Ptrade 1.1.0 (from 1.0.0)
- Added documentation files: QUICK_REFERENCE.md, README.md, requirements.txt, _meta.json, metadata.json
- Improved and reorganized documentation for better clarity and conciseness
- Updated terminology and field descriptions for consistency (e.g. "Hundsun Quantitative Trading Platform," section headings)
- Enhanced code comments and renamed some fields/variables for readability
- No functional/platform changes; documentation and metadata only
v1.0.0
Initial release of ptrade quantitative trading platform skill:
- Enables cloud-hosted Python strategy deployment for Chinese securities with broker-grade, low-latency execution (requires Ptrade broker account).
- Provides event-driven strategy framework with rich lifecycle hooks (e.g., initialize, handle_data, tick_data, on_order_response).
- Supports backtesting and live trading in stocks, convertible bonds, ETFs, LOFs, futures, and more.
- Offers built-in methods for universe/benchmark setup, commissions, slippage, scheduled tasks, and position management.
- Includes comprehensive data access functions: get_history, get_price, get_snapshot, with strict precision/tick requirements.
Metadata
Frequently Asked Questions
What is Ptrade?
Ptrade 恒生量化交易平台 - 策略运行在券商服务器上，低延迟执行，支持A股、期货、融资融券等中国证券市场。 It is an AI Agent Skill for Claude Code / OpenClaw, with 448 downloads so far.
How do I install Ptrade?
Run "/install ptrade" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Ptrade free?
Yes, Ptrade is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Ptrade support?
Ptrade is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Ptrade?
It is built and maintained by coderwpf (@coderwpf); the current version is v1.0.3.
More Skills