Skip to content

algoderiv/agent-skills

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

7 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Agent Skills

A collection of skills for AI coding agents. Skills are packaged instructions and reference documentation that extend agent capabilities for quantitative trading development, video generation APIs, and payment platform integrations.

Skills follow the Agent Skills format.

Available Skills

Skill Description Refs Source
ctp-api CTP 6.7.8 综合交易平台 API — 期货期权交易接口、行情订阅、委托执行 52 Official PDF (2466p)
rice-quant 米筐 RQData Python API — A股、港股、期货、期权、基金、债券数据 9 RQData Docs
nautilus-trader NautilusTrader 开发者指南 — 源码构建、Rust/Python 集成、测试 6 GitHub Docs
wtpy WonderTrader/wtpy 量化交易综合指南 — CTA/HFT/UFT 引擎、策略回测、实盘运维 13 Official + Community
tqsdk 天勤 TqSdk Python 量化交易框架 — 期货/期权/股票策略开发、回测与实盘,含 50+ 策略示例 47 Official Docs + Strategies
seedance1.5-api Seedance 1.5/1.0 视频生成 API(火山方舟) — 文生视频、图生视频、有声视频、SDK 示例、提示词技巧 9 Volcengine Docs
stripe Stripe 支付平台集成指南 — API 认证、Webhooks、错误处理、测试卡号、Connect 平台、SDK 7 Stripe Docs

ctp-api

CTP (Comprehensive Transaction Platform) 6.7.8 API documentation for Chinese futures and options trading. Contains 52 reference files extracted from the official 2466-page PDF.

Use when:

  • Developing with CTP API for futures/options trading
  • Implementing market data subscription and processing
  • Handling order submission, cancellation, and queries
  • Understanding CTP callback mechanisms and event handling
  • Debugging CTP connection, authentication, and trading issues

Categories covered:

  • Trading interface (CThostFtdcTraderApi) (Critical)
  • Market data interface (CThostFtdcMdApi) (Critical)
  • Callback handlers (TraderSpi, MdSpi) (High)
  • Data structures and field definitions (High)
  • Error codes and troubleshooting (Medium)
  • Regulatory monitoring (Medium)

rice-quant

RiceQuant RQData Python API for accessing Chinese financial market data — A-shares, Hong Kong stocks, futures, options, funds, bonds, and macro data.

Use when:

  • Fetching Chinese market data with rqdatac Python package
  • Building quantitative analysis pipelines for Chinese markets
  • Accessing derivatives, fixed income, or alternative data
  • Working with RiceQuant data feeds in backtesting or live trading

Categories covered:

  • Getting started and authentication (Critical)
  • Stock and equity data (High)
  • Derivatives (futures, options) data (High)
  • Fixed income and bond data (Medium)
  • Fund and index data (Medium)
  • Alternative and macro data (Low-Medium)

nautilus-trader

Developer guide for NautilusTrader, a high-performance algorithmic trading platform with Python and Rust.

Use when:

  • Building NautilusTrader from source
  • Setting up Rust/Python development environment
  • Writing and running tests for NautilusTrader
  • Contributing to the NautilusTrader project
  • Understanding Rust/Python integration via PyO3

Categories covered:

  • Development environment setup (Critical)
  • Building from source (Critical)
  • Testing practices and CI (High)
  • Rust crate structure and conventions (High)
  • Contribution guidelines (Medium)

wtpy

WonderTrader/wtpy 量化交易开发综合指南。整合官方文档(Read the Docs)、社区学习笔记和非官方整理文档三大来源,覆盖 155 页文档。

Use when:

  • 使用 wtpy 开发 CTA/SEL/HFT/UFT 策略
  • 进行策略回测、仿真交易或实盘交易
  • 配置 WonderTrader 引擎和交易/行情接口(CTP/openctp/XTP)
  • 处理行情数据(DSB/CSV 转换、数据录制、WtDataHelper)
  • 开发自定义执行器(ExtExecuter)或行情解析器(ExtParser)
  • 使用 WtMonSvr 监控服务或 WtStudio 工作台
  • WonderTrader C++ 核心开发和编译

Categories covered:

  • Getting started and architecture (Critical)
  • Strategy development and API reference (Critical)
  • Configuration files (High)
  • Data tools and management (High)
  • Advanced development — ExtParser, ExtExecuter, C++ (Medium)
  • Web console, WtStudio, and operations (Medium)
  • FAQ and troubleshooting (Medium)
  • Source code analysis (Low)

tqsdk

天勤量化 TqSdk — 信易科技开发的开源 Python 量化交易框架,基于快期交易及行情服务器,支持期货、期权、股票的行情获取、策略开发、回测与实盘交易。包含 50+ 完整策略示例(趋势、套利、均值回归等)。

Use when:

  • 使用 tqsdk 编写量化交易策略
  • 获取期货/期权/股票实时行情、K线、Tick 数据
  • 进行策略回测(Tick 级和 K 线级)
  • 实盘下单交易(直连 CTP 或快期通道)
  • 使用 tqsdk 算法模块(TWAP/VWAP)、技术指标计算(ta/tafunc)
  • 使用 TargetPosTask 目标持仓交易
  • 策略图形化界面开发 (Web GUI)

Categories covered:

  • Getting started and quick start (Critical)
  • TqApi core framework and business objects (Critical)
  • Algorithm module — TWAP, VWAP, TargetPosScheduler (High)
  • Technical indicators — tafunc / ta (High)
  • Backtesting and parameter optimization (High)
  • Strategy examples — trend, arbitrage, mean reversion (High)
  • CTP direct connection (tqctp) and multi-account (Medium)
  • Risk control module (Medium)
  • GUI, IDE integration (Cursor/Trae), and EDB data service (Low)

seedance1.5-api

Seedance 1.5/1.0 系列视频生成 API 文档(火山方舟平台)。覆盖 Seedance 1.5 pro、1.0 pro、1.0 pro fast、1.0 lite 全系列模型,含 9 个参考文件。

Use when:

  • 调用 Seedance API 生成视频(文生视频、图生视频)
  • 使用火山方舟平台的视频生成 API
  • 配置视频参数(分辨率、宽高比、时长、有声视频、样片模式)
  • 编写 Seedance 视频生成提示词(运镜、连续动作、风格关键词)
  • 使用 Python/Go/Java SDK 或 OpenAI 兼容 SDK 调用
  • 处理视频任务的创建、查询、取消和错误排查

Categories covered:

  • Video generation API (create/query/list/delete tasks) (Critical)
  • Authentication and API Key configuration (Critical)
  • SDK examples — Python, Go, Java (High)
  • Prompt engineering guide — camera, style, audio (High)
  • Resolution/ratio pixel mapping tables (Medium)
  • OpenAI SDK compatibility (Medium)
  • Error codes reference (Medium)

stripe

Stripe payment platform integration guide covering the core development workflows: API authentication, payment processing, webhook handling, error management, testing, and multi-party platforms with Stripe Connect.

Use when:

  • Integrating Stripe payments (checkout, subscriptions, invoices)
  • Setting up API keys and authentication
  • Implementing webhook endpoints and signature verification
  • Handling Stripe errors and decline codes
  • Testing with sandbox/test cards
  • Building platforms or marketplaces with Stripe Connect
  • Managing connected accounts, charges, transfers, and payouts
  • Using Stripe SDKs (Node.js, Python, Ruby, Go, Java, PHP, .NET)

Categories covered:

  • API keys and authentication (Critical)
  • Error handling — 9 error types with multi-language examples (Critical)
  • Webhooks — setup, signature verification, CLI testing, best practices (Critical)
  • Testing — 60+ test card numbers by brand/country, PaymentMethods (High)
  • Connect — account types, charge types, onboarding, payouts (High)
  • Development essentials — SDKs, expand, pagination, idempotency, metadata (High)

Installation

# Install a single skill
git clone https://github.com/algoderiv/agent-skills.git /tmp/agent-skills
cp -r /tmp/agent-skills/skills/ctp-api ~/.claude/skills/

# Install all skills
cp -r /tmp/agent-skills/skills/*/ ~/.claude/skills/

Or upload the .zip file from skills/ to claude.ai project knowledge.

Usage

Skills are automatically available once installed. The agent will use them when relevant tasks are detected.

Examples:

Help me implement CTP market data subscription with OnRtnDepthMarketData callback
Fetch CZCE futures daily data using rqdatac
How do I build NautilusTrader from source with Rust?
用 wtpy 写一个 DualThrust CTA 策略并回测
用 tqsdk 订阅螺纹钢主力合约实时行情并下单
用 Seedance 1.5 pro API 生成一段 10 秒有声视频,首帧为上传的图片
Help me set up Stripe webhooks with signature verification in Node.js

Skill Structure

Each skill contains:

  • SKILL.md - Instructions for the agent
  • references/ - Supporting documentation (loaded on demand)
  • scripts/ - Helper scripts for automation (optional)
  • {skill-name}.zip - Packaged for claude.ai upload

Sources

License

MIT

About

Claude Code skills for China quant trading

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors