Nautilus Algo Trading

使用 NautilusTrader 配置驱动的 BacktestNode 运行高性能多市场回测，支持 Parquet 数据目录和外部 CSV 数据导入，策略可直接过渡到实盘交易。。

Security Scan

The skill is an instruction-only backtesting helper that asks about markets/data/strategy and references a large set of documentation; its requested actions and checks are coherent with a backtesting workflow and do not ask for unrelated credentials or hidden installs, but it may instruct the agent to run package installs and create/read local data directories (expected for this domain).

Purpose & Capability

Name/description (Nautilus backtesting) match the included materials (BacktestNode, Parquet catalogs, ZVT-like concerns). The files and referenced preconditions are relevant to running backtests and data ingestion; there are no unrelated environment variables, binaries, or config paths required.

Instruction Scope

The SKILL.md and seed.yaml direct the agent to run precondition checks that execute Python commands (e.g., import zvt, get_kdata, create ~/.zvt) and to prompt the user for backtest parameters. Those checks and directory write tests are proportionate to preparing a backtest environment, but they do cause the agent to run shell/python commands on the host if executed. No instructions request secrets or data exfiltration, but the agent may install packages or write/read local data directories as part of setup.

Install Mechanism

There is no formal install spec (instruction-only). The runtime preconditions suggest using pip (python3 -m pip install zvt) if zvt is missing — a common, expected pattern for a Python backtest skill. This implies network package installation at the user's discretion; no unusual download URLs or archive extraction are present in the skill bundle itself.

Credentials

The skill declares no required env vars, secrets, or config paths. It references ZVT_HOME and will check/create ~/.zvt for data directory readiness; this is proportional to the skill's function (data storage for backtests). No credentials or unrelated tokens are requested.

Persistence & Privilege

always:false and default autonomous invocation are retained. The skill does not declare elevated privileges, does not request permanent presence, and does not attempt to modify other skills or system-wide agent settings. Its workspace path hints (host_workspace) are normal for a local backtesting tool.

Assessment

This skill is internally consistent with a backtesting helper: it will ask you for market/data/strategy details and may instruct the agent to run small host-side checks (python -c import zvt, test write to ~/.zvt) and to install missing Python packages (pip install zvt). Before running those commands: 1) confirm you trust installing the referenced Python packages from PyPI (they will run network downloads); 2) be aware the skill may create/write to a data directory (default ~/.zvt) — change ZVT_HOME if you prefer a different location; 3) no secrets or cloud credentials are requested by the skill, so there is no obvious secret-exfiltration vector in the provided files. If you want extra caution, run the precondition commands yourself in a controlled environment (virtualenv) rather than letting the agent run them automatically.

Nautilus 算法回测 (nautilus-algo-trading)

使用 NautilusTrader 配置驱动的 BacktestNode 运行高性能多市场回测，支持 Parquet 数据目录和外部 CSV 数据导入，策略可直接过渡到实盘交易。

Pipeline

data_collection -> data_storage -> factor_computation -> target_selection -> trading_execution -> visualization

Top Use Cases (73 total)

High-Level Backtest with Parquet Data Catalog (UC-101)

Users need to run backtests using a config-driven approach with the BacktestNode, enabling reproducible production workflows that can transition to li Triggers: backtest, BacktestNode, Parquet catalog

Low-Level Backtest with Direct Component Access (UC-102)

Users need fine-grained control over backtesting components including custom execution algorithms like TWAP for sophisticated order execution simulati Triggers: BacktestEngine, TWAP, direct control

Quickstart EMA Crossover Backtest (UC-103)

New users need a minimal example to run their first backtest quickly, demonstrating the core strategy and market data handling pattern Triggers: quickstart, first backtest, EMA crossover

For all 73 use cases, see references/USE_CASES.md.

Execute trigger: When user intent matches intent_router.uc_entries[].positive_terms AND user uses action verb (run/execute/跑/执行/backtest/fetch/collect)

What I'll Ask You

• Target market: A-share (default), HK, or crypto? (US stocks in ZVT are half-baked — stockus_nasdaq_AAPL exists but coverage is thin)
• Data source / provider: eastmoney (free, no account), joinquant (account+paid), baostock (free, good history), akshare, or qmt (broker)?
• Strategy type: MACD golden-cross, MA crossover, volume breakout, fundamental screen, or custom factor?
• Time range: start_timestamp and end_timestamp for backtest period
• Target entity IDs: specific stocks (stock_sh_600000) or index components (SZ1000)?

Semantic Locks (Fatal)

ID	Rule	On Violation
SL-01	Execute sell orders before buy orders in every trading cycle	halt
SL-02	Trading signals MUST use next-bar execution (no look-ahead)	halt
SL-03	Entity IDs MUST follow format entity_type_exchange_code	halt
SL-04	DataFrame index MUST be MultiIndex (entity_id, timestamp)	halt
SL-05	TradingSignal MUST have EXACTLY ONE of: position_pct, order_money, order_amount	halt
SL-06	filter_result column semantics: True=BUY, False=SELL, None/NaN=NO ACTION	halt
SL-07	Transformer MUST run BEFORE Accumulator in factor pipeline	halt
SL-08	MACD parameters locked: fast=12, slow=26, signal=9	halt

Full lock definitions: references/LOCKS.md

Top Anti-Patterns (25 total)

• AP-ZVT-183: 除权因子为 inf/NaN 时直接参与乘法导致复权静默失败
• AP-ZVT-179: 第三方数据接口超限后异常被吞噬，数据静默缺失
• AP-ZVT-183B: HFQ（后复权）与 QFQ（前复权）K 线表使用错误导致因子计算漂移

All 25 anti-patterns: references/ANTI_PATTERNS.md

Evidence Quality Notice

[QUALITY NOTICE] This crystal was compiled from blueprint finance-bp-098. Evidence verify ratio = 33.3% and audit fail total = 20. Generated results may have uncaptured requirement gaps. Verify critical decisions against source files (LATEST.yaml / LATEST.jsonl).

Reference Files

File	Contents	When to Load
references/seed.yaml	V6+ 全量权威 (source-of-truth)	有行为/决策争议时必读
references/ANTI_PATTERNS.md	25 条跨项目反模式	开始实现前
references/WISDOM.md	跨项目精华借鉴	架构决策时
references/CONSTRAINTS.md	domain + fatal 约束	规则冲突时
references/USE_CASES.md	全量 KUC-* 业务场景	需要完整示例时
references/LOCKS.md	SL-* + preconditions + hints	生成回测/交易代码前
references/COMPONENTS.md	AST 组件地图（按 module 拆分）	查 API 时

---

Compiled by Doramagic crystal-compilation-v6.1 from finance-bp-098 blueprint at 2026-04-22T13:00:43.978060+00:00. See human_summary.md for non-technical overview.