{"id":"ce3f68cd-fdbb-4d5c-a6d8-3b58173828df","shortId":"4SLLAx","kind":"skill","title":"akquant","tagline":"生成 akquant 框架的可执行量化策略代码，涵盖数据接口、事件驱动、风控与优化；当用户需要开发 akquant量化策略 时使用","description":"# AKQuant 量化策略开发指南\n\n## 任务目标\n\n本 Skill 用于辅助 AI 编程智能体生成符合 akquant 框架规范的可执行量化策略代码。能力包括策略设计、回测配置、订单管理、风控规则、参数优化与横截面策略实现。\n\n## 核心能力清单\n\n- 策略类生成：继承 Strategy 基类，实现生命周期钩子\n- 数据接口配置：准备 DataFrame 数据、设置预热期、访问历史数据\n- 事件驱动机制：on_bar/on_tick/on_order/on_trade 等回调\n- 订单管理：市价单/限价单/目标仓位/OCO/Bracket/Trailing Stop\n- 风控规则：持仓限制/回撤熔断/止损阈值/行业集中度\n- 参数优化：网格搜索与滚动优化（Walk-Forward）\n- 多策略编排：slot 映射与策略级风控\n\n## 触发条件\n\n当用户表达以下意图时触发：\n- 开发量化交易策略\n- 配置回测环境与参数\n- 设置风险控制规则\n- 进行参数优化与调优\n- 实现横截面或轮动策略\n- 排查策略运行错误\n\n## 策略开发工作流\n\n### 阶段一：理解需求\n\n1. 识别策略类型：趋势跟踪、均值回归、横截面轮动、套利等\n2. 确定数据需求：时间周期、标的范围、字段要求\n3. 明确风控约束：持仓上限、止损止盈、回撤限制\n\n### 阶段二：设计策略结构\n\n参考 [strategy-patterns.md](references/strategy-patterns.md) 选择范式：\n- 类风格（推荐）：继承 Strategy，封装状态与逻辑\n- 函数风格：initialize + on_bar，快速原型\n\n关键决策点：\n- 预热期设置：根据指标窗口长度计算\n- 历史数据访问：get_history (numpy) 或 get_history_df (DataFrame)\n- 执行模式：NextOpen（下一 Bar 开盘）或 CurrentClose（当前 Bar 收盘）\n\n### 阶段三：编写策略代码\n\n使用 [assets/strategy-template.py](assets/strategy-template.py) 作为起点：\n\n```python\nfrom akquant import Strategy, Bar\n\nclass MyStrategy(Strategy):\n    warmup_period = 20  # 预热数据长度\n\n    def __init__(self, param1=10):\n        self.param1 = param1\n\n    def on_start(self):\n        self.subscribe(\"600000\")\n\n    def on_bar(self, bar: Bar):\n        # 核心交易逻辑\n        history = self.get_history(self.param1, bar.symbol, \"close\")\n        if len(history) < self.param1:\n            return\n\n        import numpy as np\n        ma = np.mean(history)\n        pos = self.get_position(bar.symbol)\n\n        if bar.close > ma and pos == 0:\n            self.buy(bar.symbol, 100)\n        elif bar.close < ma and pos > 0:\n            self.sell(bar.symbol, 100)\n```\n\n### 阶段四：配置回测环境\n\n参考 [api-reference.md](references/api-reference.md) 设置参数：\n\n```python\nfrom akquant import run_backtest\n\nresult = run_backtest(\n    strategy=MyStrategy,\n    data=df,\n    symbol=\"600000\",\n    initial_cash=500_000.0,\n    commission_rate=0.0003,\n    stamp_tax_rate=0.001,\n    t_plus_one=True,  # A 股 T+1 规则\n    warmup_period=20,\n    execution_mode=\"NextOpen\",\n)\n```\n\n### 阶段五：设置风控规则\n\n参考 [risk-management.md](references/risk-management.md) 配置：\n\n```python\nfrom akquant.config import RiskConfig\n\nresult = run_backtest(\n    ...,\n    risk_config=RiskConfig(\n        max_position_pct=0.10,  # 单标的持仓不超过 10%\n        max_account_drawdown=0.20,  # 最大回撤 20%\n        max_daily_loss=0.05,  # 单日亏损 5%\n    ),\n)\n```\n\n### 阶段六：参数优化\n\n参考 [optimization.md](references/optimization.md) 执行：\n\n```python\nfrom akquant import run_grid_search, run_walk_forward\n\n# 网格搜索\nresults = run_grid_search(\n    strategy=MyStrategy,\n    param_grid={\"param1\": [10, 20, 30]},\n    data=df,\n    sort_by=\"sharpe_ratio\",\n)\n\n# 滚动优化（推荐）\nwfo_results = run_walk_forward(\n    strategy=MyStrategy,\n    param_grid={\"param1\": [10, 20, 30]},\n    data=df,\n    train_period=250,\n    test_period=60,\n    metric=\"sharpe_ratio\",\n)\n```\n\n## 横截面策略开发\n\n参考 [cross-section-guide.md](references/cross-section-guide.md) 实现多标的轮动：\n\n**推荐范式**：使用 on_timer 统一触发调仓\n\n```python\nclass CrossSectionStrategy(Strategy):\n    def __init__(self):\n        self.universe = [\"sh600519\", \"sz000858\", \"sh601318\"]\n\n    def on_start(self):\n        self.add_daily_timer(\"14:55:00\", \"rebalance\")\n\n    def on_timer(self, payload):\n        if payload != \"rebalance\":\n            return\n\n        # 计算所有标分数\n        scores = {}\n        for symbol in self.universe:\n            history = self.get_history(20, symbol, \"close\")\n            scores[symbol] = (history[-1] - history[0]) / history[0]\n\n        # 选出最佳标的并调仓\n        best = max(scores, key=scores.get)\n        self.order_target_percent(0.95, symbol=best)\n```\n\n## 资源索引\n\n| 资源 | 用途 | 何时读取 |\n|------|------|----------|\n| [api-reference.md](references/api-reference.md) | API 速查 | 查询函数签名与参数 |\n| [strategy-patterns.md](references/strategy-patterns.md) | 策略范式 | 设计策略结构 |\n| [risk-management.md](references/risk-management.md) | 风控配置 | 设置风控规则 |\n| [optimization.md](references/optimization.md) | 参数优化 | 调优策略参数 |\n| [cross-section-guide.md](references/cross-section-guide.md) | 横截面策略 | 实现多标的轮动 |\n| [strategy-template.py](assets/strategy-template.py) | 策略模板 | 快速生成代码骨架 |\n## 环境准备与依赖管理\n\n### 使用 uv 管理项目环境（推荐）\n\n由于 akquant 依赖 `pandas>=3.0.0`，全局安装可能与现有项目存在版本冲突。推荐使用 **uv** 创建隔离环境：\n\n#### 1. 安装 uv：若已安装则跳过\n\n```bash\n# macOS\nbrew install uv\n\n# Linux\ncurl -LsSf https://astral.sh/uv/install.sh | sh\n\n# Windows (PowerShell)\npowershell -ExecutionPolicy ByPass -c \"irm https://astral.sh/uv/install.ps1 | iex\"\n```\n\n#### 2. 创建项目并初始化环境\n\n```bash\n# 创建项目目录\nmkdir my-quant-strategy\ncd my-quant-strategy\n\n# 初始化项目（创建 pyproject.toml）\nuv init\n\n# 创建虚拟环境并安装依赖\nuv venv\nuv add akquant pandas numpy\n```\n\n#### 3. 运行策略脚本\n\n```bash\n# 方式一：使用 uv run（推荐）\nuv run python my_strategy.py\n\n# 方式二：激活虚拟环境后运行\nsource .venv/bin/activate  # macOS/Linux\n# .venv\\Scripts\\activate   # Windows\npython my_strategy.py\n```\n\n#### 4. 依赖版本锁定\n\nuv 会自动生成 `uv.lock` 文件，确保团队依赖一致：\n\n```bash\n# 安装精确版本（从 lock 文件）\nuv sync\n\n# 添加新依赖\nuv add scipy  # 自动更新 lock 文件\n```\n\n#### 5. 项目结构建议\n\n```\nmy-quant-strategy/\n├── .venv/              # 虚拟环境（uv 自动创建）\n├── pyproject.toml      # 项目配置\n├── uv.lock             # 依赖锁定文件\n├── strategies/         # 策略脚本\n│   ├── ma_strategy.py\n│   └── cross_section.py\n└── data/               # 数据文件\n    └── stock_data.csv\n```\n\n### 快速启动命令\n\n```bash\n# 一键创建并运行策略项目\nmkdir quant-project && cd quant-project\nuv init\nuv venv\nuv add akquant pandas numpy\n\n# 创建策略文件（使用模板）\ncat > strategy.py << 'EOF'\nfrom akquant import Strategy, Bar, run_backtest\nimport pandas as pd\nimport numpy as np\n\nclass MyStrategy(Strategy):\n    warmup_period = 20\n\n    def on_bar(self, bar: Bar):\n        closes = self.get_history(20, bar.symbol, \"close\")\n        if len(closes) < 20:\n            return\n        ma = np.mean(closes)\n        pos = self.get_position(bar.symbol)\n        if bar.close > ma and pos == 0:\n            self.buy(bar.symbol, 100)\n        elif bar.close < ma and pos > 0:\n            self.sell(bar.symbol, pos)\n\n# 准备数据并运行回测\n# result = run_backtest(strategy=MyStrategy, data=df, symbol=\"600000\")\nEOF\n\n# 运行策略\nuv run python strategy.py\n```\n\n## 注意事项\n\n1. **预热期计算**：确保 warmup_period >= 指标所需的最大窗口长度\n2. **T+1 规则**：A 股策略需设置 t_plus_one=True，并区分总持仓与可用持仓\n3. **风控优先级**：显式参数 > 配置对象 > 默认值\n4. **数据格式**：DataFrame 必须包含 date/open/high/low/close/volume/symbol 字段\n5. **横截面触发**：优先使用 on_timer，无固定时点再考虑 timestamp 收齐方案\n6. **优化风险**：网格搜索易过拟合，推荐使用滚动优化验证稳健性\n\n## 使用示例\n\n### 示例 1：双均线策略\n\n```python\nfrom akquant import Strategy, Bar\nimport numpy as np\n\nclass DualMAStrategy(Strategy):\n    warmup_period = 30\n\n    def __init__(self, fast=10, slow=20):\n        self.fast = fast\n        self.slow = slow\n        self.warmup_period = slow + 1\n\n    def on_bar(self, bar: Bar):\n        fast_ma = np.mean(self.get_history(self.fast, bar.symbol, \"close\"))\n        slow_ma = np.mean(self.get_history(self.slow, bar.symbol, \"close\"))\n\n        pos = self.get_position(bar.symbol)\n        if fast_ma > slow_ma and pos == 0:\n            self.buy(bar.symbol, 100)\n        elif fast_ma < slow_ma and pos > 0:\n            self.sell(bar.symbol, pos)\n```\n\n### 示例 2：带风控的趋势策略\n\n```python\nfrom akquant import Strategy, Bar, run_backtest\nfrom akquant.config import RiskConfig\nimport numpy as np\n\nclass TrendStrategy(Strategy):\n    warmup_period = 20\n\n    def __init__(self, ma_window=20, stop_loss=0.05):\n        self.ma_window = ma_window\n        self.stop_loss = stop_loss\n\n    def on_bar(self, bar: Bar):\n        ma = np.mean(self.get_history(self.ma_window, bar.symbol, \"close\"))\n        pos = self.get_position(bar.symbol)\n\n        if bar.close > ma * 1.02 and pos == 0:\n            self.buy(bar.symbol, 100)\n        elif bar.close < ma * 0.98 and pos > 0:\n            self.sell(bar.symbol, pos)\n\n# 运行回测\nresult = run_backtest(\n    strategy=TrendStrategy,\n    data=df,\n    symbol=\"600000\",\n    initial_cash=1_000_000.0,\n    risk_config=RiskConfig(\n        max_position_pct=0.20,\n        max_account_drawdown=0.15,\n        stop_loss_threshold=0.85,\n    ),\n)\n```\n\n### 示例 3：横截面动量轮动\n\n```python\nfrom akquant import Strategy, run_backtest\nimport numpy as np\n\nclass MomentumRotation(Strategy):\n    def __init__(self, lookback=20):\n        self.lookback = lookback\n        self.universe = [\"sh600519\", \"sz000858\", \"sh601318\"]\n        self.warmup_period = lookback + 1\n\n    def on_start(self):\n        for symbol in self.universe:\n            self.subscribe(symbol)\n        self.add_daily_timer(\"14:55:00\", \"rebalance\")\n\n    def on_timer(self, payload):\n        if payload != \"rebalance\":\n            return\n\n        scores = {}\n        for symbol in self.universe:\n            closes = self.get_history(self.lookback, symbol, \"close\")\n            if len(closes) < self.lookback:\n                return\n            scores[symbol] = (closes[-1] - closes[0]) / closes[0]\n\n        # 选出最佳标的，持仓 95%\n        best = max(scores, key=scores.get)\n        self.order_target_percent(0.95, symbol=best)\n```","tags":["akquant","finance","quant","skills","lzwme","agent-skills"],"capabilities":["skill","source-lzwme","skill-akquant","topic-agent-skills","topic-skills"],"categories":["finance-quant-skills"],"synonyms":[],"warnings":[],"endpointUrl":"https://skills.sh/lzwme/finance-quant-skills/akquant","protocol":"skill","transport":"skills-sh","auth":{"type":"none","details":{"cli":"npx skills add lzwme/finance-quant-skills","source_repo":"https://github.com/lzwme/finance-quant-skills","install_from":"skills.sh"}},"qualityScore":"0.469","qualityRationale":"deterministic score 0.47 from registry signals: · indexed on github topic:agent-skills · 39 github stars · SKILL.md body (8,733 chars)","verified":false,"liveness":"unknown","lastLivenessCheck":null,"agentReviews":{"count":0,"score_avg":null,"cost_usd_avg":null,"success_rate":null,"latency_p50_ms":null,"narrative_summary":null,"summary_updated_at":null},"enrichmentModel":"deterministic:skill-github:v1","enrichmentVersion":1,"enrichedAt":"2026-05-18T18:58:28.273Z","embedding":null,"createdAt":"2026-05-09T01:05:17.815Z","updatedAt":"2026-05-18T18:58:28.273Z","lastSeenAt":"2026-05-18T18:58:28.273Z","tsv":"'+1':244,696 '-1':404,998 '/oco/bracket/trailing':46 '/uv/install.ps1':489 '/uv/install.sh':478 '0':192,201,406,408,658,667,796,807,877,887,1000,1002 '0.0003':232 '0.001':236 '0.05':284,844 '0.10':272 '0.15':916 '0.20':278,912 '0.85':920 '0.95':418,1014 '0.98':884 '00':378,968 '000':904 '000.0':229,905 '1':72,464,688,730,762,903,952 '1.02':874 '10':149,274,313,334,752 '100':195,204,661,799,880 '14':376,966 '2':78,491,694,812 '20':143,248,280,314,335,398,628,638,644,754,835,841,942 '250':341 '3':83,518,705,922 '3.0.0':459 '30':315,336,747 '4':541,710 '5':286,562,716 '500':228 '55':377,967 '6':724 '60':344 '600000':157,225,680,900 '95':1005 'account':276,914 'activ':537 'add':514,557,599 'ai':17 'akquant':1,3,11,19,134,213,295,456,515,600,609,734,816,926 'akquant.config':260,823 'akquant量化策略':9 'api':427 'api-reference.md':208,425 'assets/strategy-template.py':129,130,447 'astral.sh':477,488 'astral.sh/uv/install.ps1':487 'astral.sh/uv/install.sh':476 'backtest':216,219,265,614,674,821,894,930 'bar':102,119,124,137,160,162,163,612,631,633,634,737,765,767,768,819,855,857,858 'bar.close':188,197,654,663,872,882 'bar.symbol':169,186,194,203,639,652,660,669,775,783,788,798,809,865,870,879,889 'bar/on_tick/on_order/on_trade':40 'bash':468,493,520,548,584 'best':410,420,1006,1016 'brew':470 'bypass':484 'c':485 'cash':227,902 'cat':605 'cd':500,590 'class':138,359,623,742,830,935 'close':170,400,635,640,643,648,776,784,866,984,989,992,997,999,1001 'commiss':230 'config':267,907 'cross-section-guide.md':350,442 'cross_section.py':579 'crosssectionstrategi':360 'curl':474 'currentclos':122 'daili':282,374,964 'data':222,316,337,580,677,897 'datafram':34,115,712 'date/open/high/low/close/volume/symbol':714 'def':145,152,158,362,369,380,629,748,763,836,853,938,953,970 'df':114,223,317,338,678,898 'drawdown':277,915 'dualmastrategi':743 'elif':196,662,800,881 'eof':607,681 'execut':249 'executionpolici':483 'fast':751,756,769,790,801 'forward':57,302,328 'get':108,112 'grid':298,306,311,332 'histori':109,113,165,167,173,182,395,397,403,405,407,637,773,781,862,986 'iex':490 'import':135,176,214,261,296,610,615,619,735,738,817,824,826,927,931 'init':146,363,509,595,749,837,939 'initi':100,226,901 'instal':471 'irm':486 'key':413,1009 'len':172,642,991 'linux':473 'lock':551,560 'lookback':941,944,951 'loss':283,843,850,852,918 'lssf':475 'ma':180,189,198,646,655,664,770,778,791,793,802,804,839,847,859,873,883 'ma_strategy.py':578 'maco':469 'macos/linux':534 'max':269,275,281,411,909,913,1007 'metric':345 'mkdir':495,586 'mode':250 'momentumrot':936 'my-quant-strategi':496,501,564 'my_strategy.py':529,540 'mystrategi':139,221,309,330,624,676 'nextopen':117,251 'np':179,622,741,829,934 'np.mean':181,647,771,779,860 'numpi':110,177,517,602,620,739,827,932 'one':239,702 'optimization.md':290,438 'panda':458,516,601,616 'param':310,331 'param1':148,151,312,333 'payload':384,386,974,976 'pct':271,911 'pd':618 'percent':417,1013 'period':142,247,340,343,627,692,746,760,834,950 'plus':238,701 'pos':183,191,200,649,657,666,670,785,795,806,810,867,876,886,890 'posit':185,270,651,787,869,910 'powershel':481,482 'project':589,593 'pyproject.toml':507,572 'python':132,211,258,293,358,528,539,685,732,814,924 'quant':498,503,566,588,592 'quant-project':587,591 'rate':231,235 'ratio':321,347 'rebal':379,387,969,977 'references/api-reference.md':209,426 'references/cross-section-guide.md':351,443 'references/optimization.md':291,439 'references/risk-management.md':256,435 'references/strategy-patterns.md':92,431 'result':217,263,304,325,672,892 'return':175,388,645,978,994 'risk':266,906 'risk-management.md':255,434 'riskconfig':262,268,825,908 'run':215,218,264,297,300,305,326,524,527,613,673,684,820,893,929 'scipi':558 'score':390,401,412,979,995,1008 'scores.get':414,1010 'script':536 'search':299,307 'self':147,155,161,364,372,383,632,750,766,838,856,940,956,973 'self.add':373,963 'self.buy':193,659,797,878 'self.fast':755,774 'self.get':166,184,396,636,650,772,780,786,861,868,985 'self.lookback':943,987,993 'self.ma':845,863 'self.order':415,1011 'self.param1':150,168,174 'self.sell':202,668,808,888 'self.slow':757,782 'self.stop':849 'self.subscribe':156,961 'self.universe':365,394,945,960,983 'self.warmup':759,949 'sh':479 'sh600519':366,946 'sh601318':368,948 'sharp':320,346 'skill':15 'skill-akquant' 'slot':59 'slow':753,758,761,777,792,803 'sort':318 'sourc':532 'source-lzwme' 'stamp':233 'start':154,371,955 'stock_data.csv':582 'stop':47,842,851,917 'strategi':29,97,136,140,220,308,329,361,499,504,567,576,611,625,675,736,744,818,832,895,928,937 'strategy-patterns.md':91,430 'strategy-template.py':446 'strategy.py':606,686 'symbol':224,392,399,402,419,679,899,958,962,981,988,996,1015 'sync':554 'sz000858':367,947 'target':416,1012 'tax':234 'test':342 'threshold':919 'timer':356,375,382,720,965,972 'timestamp':722 'topic-agent-skills' 'topic-skills' 'train':339 'trendstrategi':831,896 'true':240,703 'uv':452,462,466,472,508,511,513,523,526,543,553,556,570,594,596,598,683 'uv.lock':545,574 'venv':512,535,568,597 'venv/bin/activate':533 'walk':56,301,327 'walk-forward':55 'warmup':141,246,626,691,745,833 'wfo':324 'window':480,538,840,846,848,864 '一键创建并运行策略项目':585 '下一':118 '事件驱动':6 '事件驱动机制':38 '从':550 '任务目标':13 '优先使用':718 '优化风险':725 '会自动生成':544 '何时读取':424 '作为起点':131 '使用':128,354,451,522 '使用模板':604 '使用示例':728 '依赖':457 '依赖版本锁定':542 '依赖锁定文件':575 '全局安装可能与现有项目存在版本冲突':460 '关键决策点':104 '准备':33 '准备数据并运行回测':671 '函数风格':99 '创建':506 '创建策略文件':603 '创建虚拟环境并安装依赖':510 '创建隔离环境':463 '创建项目并初始化环境':492 '创建项目目录':494 '初始化项目':505 '单日亏损':285 '单标的持仓不超过':273 '历史数据访问':107 '参数优化':53,288,440 '参数优化与横截面策略实现':25 '参考':90,207,254,289,349 '双均线策略':731 '回撤熔断':50 '回撤限制':87 '回测配置':22 '均值回归':75 '基类':30 '多策略编排':58 '套利等':77 '字段':715 '字段要求':82 '安装':465 '安装精确版本':549 '实现多标的轮动':352,445 '实现横截面或轮动策略':67 '实现生命周期钩子':31 '封装状态与逻辑':98 '市价单':43 '带风控的趋势策略':813 '并区分总持仓与可用持仓':704 '开发量化交易策略':63 '开盘':120 '当前':123 '当用户表达以下意图时触发':62 '当用户需要开发':8 '必须包含':713 '快速原型':103 '快速启动命令':583 '快速生成代码骨架':449 '或':111,121 '执行':292 '执行模式':116 '持仓':1004 '持仓上限':85 '持仓限制':49 '指标所需的最大窗口长度':693 '排查策略运行错误':68 '推荐':95,323,454,525 '推荐使用':461 '推荐使用滚动优化验证稳健性':727 '推荐范式':353 '收盘':125 '收齐方案':723 '数据':35 '数据接口配置':32 '数据文件':581 '数据格式':711 '文件':546,552,561 '方式一':521 '方式二':530 '无固定时点再考虑':721 '时使用':10 '时间周期':80 '明确风控约束':84 '映射与策略级风控':60 '显式参数':707 '最大回撤':279 '本':14 '查询函数签名与参数':429 '标的范围':81 '核心交易逻辑':164 '核心能力清单':26 '根据指标窗口长度计算':106 '框架的可执行量化策略代码':4 '框架规范的可执行量化策略代码':20 '横截面动量轮动':923 '横截面策略':444 '横截面策略开发':348 '横截面触发':717 '横截面轮动':76 '止损止盈':86 '止损阈值':51 '注意事项':687 '涵盖数据接口':5 '添加新依赖':555 '滚动优化':322 '激活虚拟环境后运行':531 '环境准备与依赖管理':450 '理解需求':71 '生成':2 '用于辅助':16 '用途':423 '由于':455 '目标仓位':45 '确保':690 '确保团队依赖一致':547 '确定数据需求':79 '示例':729,811,921 '等回调':41 '策略开发工作流':69 '策略模板':448 '策略类生成':27 '策略脚本':577 '策略范式':432 '管理项目环境':453 '类风格':94 '统一触发调仓':357 '继承':28,96 '编写策略代码':127 '编程智能体生成符合':18 '网格搜索':303 '网格搜索与滚动优化':54 '网格搜索易过拟合':726 '股':242 '股策略需设置':699 '能力包括策略设计':21 '自动创建':571 '自动更新':559 '若已安装则跳过':467 '虚拟环境':569 '行业集中度':52 '规则':245,697 '触发条件':61 '计算所有标分数':389 '订单管理':23,42 '设置参数':210 '设置预热期':36 '设置风控规则':253,437 '设置风险控制规则':65 '设计策略结构':89,433 '访问历史数据':37 '识别策略类型':73 '调优策略参数':441 '资源':422 '资源索引':421 '趋势跟踪':74 '运行回测':891 '运行策略':682 '运行策略脚本':519 '进行参数优化与调优':66 '选出最佳标的':1003 '选出最佳标的并调仓':409 '选择范式':93 '速查':428 '配置':257 '配置回测环境':206 '配置回测环境与参数':64 '配置对象':708 '量化策略开发指南':12 '阶段一':70 '阶段三':126 '阶段二':88 '阶段五':252 '阶段六':287 '阶段四':205 '限价单':44 '项目结构建议':563 '项目配置':573 '预热数据长度':144 '预热期计算':689 '预热期设置':105 '风控与优化':7 '风控优先级':706 '风控规则':24,48 '风控配置':436 '默认值':709","prices":[{"id":"c585cd95-c1b1-433d-92e4-cfc18a1a0126","listingId":"ce3f68cd-fdbb-4d5c-a6d8-3b58173828df","amountUsd":"0","unit":"free","nativeCurrency":null,"nativeAmount":null,"chain":null,"payTo":null,"paymentMethod":"skill-free","isPrimary":true,"details":{"org":"lzwme","category":"finance-quant-skills","install_from":"skills.sh"},"createdAt":"2026-05-09T01:05:17.815Z"}],"sources":[{"listingId":"ce3f68cd-fdbb-4d5c-a6d8-3b58173828df","source":"github","sourceId":"lzwme/finance-quant-skills/akquant","sourceUrl":"https://github.com/lzwme/finance-quant-skills/tree/main/skills/akquant","isPrimary":false,"firstSeenAt":"2026-05-09T01:05:17.815Z","lastSeenAt":"2026-05-18T18:58:28.273Z"}],"details":{"listingId":"ce3f68cd-fdbb-4d5c-a6d8-3b58173828df","quickStartSnippet":null,"exampleRequest":null,"exampleResponse":null,"schema":null,"openapiUrl":null,"agentsTxtUrl":null,"citations":[],"useCases":[],"bestFor":[],"notFor":[],"kindDetails":{"org":"lzwme","slug":"akquant","github":{"repo":"lzwme/finance-quant-skills","stars":39,"topics":["agent-skills","skills"],"license":null,"html_url":"https://github.com/lzwme/finance-quant-skills","pushed_at":"2026-05-18T07:34:02Z","description":"一个面向金融量化交易领域的 Agent Skills 技能维护仓库，主要聚焦A股量化交易。","skill_md_sha":"7fc3735b11cbd873fa09258314f78abf63f71402","skill_md_path":"skills/akquant/SKILL.md","default_branch":"main","skill_tree_url":"https://github.com/lzwme/finance-quant-skills/tree/main/skills/akquant"},"layout":"multi","source":"github","category":"finance-quant-skills","frontmatter":{"name":"akquant","description":"生成 akquant 框架的可执行量化策略代码，涵盖数据接口、事件驱动、风控与优化；当用户需要开发 akquant量化策略 时使用"},"skills_sh_url":"https://skills.sh/lzwme/finance-quant-skills/akquant"},"updatedAt":"2026-05-18T18:58:28.273Z"}}