Ml4t Book Notebooks

基于《Machine Learning for Trading》第二版配套 notebooks 实现量化交易策略开发与回测，涵盖多市场金融数据的时间序列机器学习分析。

Security Scan

The skill's purpose (ML trading notebooks) aligns with the bundled documentation, but there are small inconsistencies and implicit runtime actions (environment checks, package installs, filesystem writes) that are not declared in the registry metadata — review before installing or running.

Purpose & Capability

Name/description, included references (components, locks, anti-patterns) and the SKILL.md all consistently describe a Machine Learning for Trading notebook/workflow (zvt-based pipeline). Nothing in the files claims unrelated capabilities (no unexpected cloud or secret access requested).

Instruction Scope

SKILL.md and seed.yaml contain explicit runtime preconditions that expect the host to run Python checks and pip actions (e.g., `python3 -c 'import zvt'`, instructions to `pip install zvt`, and `python3 -m zvt.init_dirs`) and to create/check ~/.zvt. Those actions will read/write local filesystem and may install packages — reasonable for this kind of notebook, but the registry metadata declared no required binaries or install steps, so the runtime scope is under-declared.

Install Mechanism

This is an instruction-only skill (no install spec, no code files). That's lower risk. However SKILL.md claims "Requires Python 3.12+ with uv package manager" while the manifest lists no required binaries and no install recipe — mismatch. Also 'uv package manager' is not a common/default manager on many hosts and may indicate a typo or an undocumented dependency; the skill does advise installing zvt via pip in preconditions but does not ship a formal install recipe.

Credentials

The skill declares no required environment variables or credentials and the files do not ask for tokens or secrets. It does reference ZVT_HOME and suggests creating/checking ~/.zvt via precondition checks — appropriate for a local data directory but worth noting that the skill will access home-directory paths.

Persistence & Privilege

Default invocation/persistence flags are normal (always:false, agent-autonomy allowed). The seed.yaml execution_protocol instructs the host/agent to re-read seed.yaml and run preconditions before executing — this gives the skill broad runtime control over checking/installing local packages and directories, which is expected for a heavy notebook workflow but is not reflected in the registry install metadata.

What to consider before installing

This package appears to be a collection of docs/blueprints for zvt-based quant notebooks and is internally consistent with that purpose. Before you run it: 1) Confirm your environment has Python 3.12+ (the manifest did not declare it) and clarify what the author means by the "uv" package manager (may be a typo or an undocumented requirement). 2) Inspect seed.yaml and the LOCKS.md/PC-* preconditions — the skill expects to run local Python checks, may call pip to install zvt and will create/check ~/.zvt (writes to your home). If you don't want packages installed or dirs created automatically, do not run the automated preconditions; instead manually review and run only the commands you trust. 3) There are no requested credentials, but verify there is no hidden code (this is instruction-only here, but if you later download code or run commands that fetch code, inspect those sources). 4) If you plan to use real brokers or execution paths, audit any later code that performs network or order execution — this skill enforces strong semantic locks (e.g., no look-ahead, T+1 rules) that will affect trading behavior. If any of the above is unclear, treat the skill as untrusted until you can verify the exact runtime commands and dependencies.

ML4T 交易教程 (ml4t-book-notebooks)

基于《Machine Learning for Trading》第二版配套 notebooks 实现量化交易策略开发与回测，涵盖多市场金融数据的时间序列机器学习分析。

Pipeline

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

Top Use Cases (0 total)

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 (15 total)

• AP-TIME-SERIES-ML-001: TimeSeries values array dimensionality mismatch
• AP-TIME-SERIES-ML-002: Non-floating-point dtype in TimeSeries values
• AP-TIME-SERIES-ML-003: Irregular or non-monotonic time index

All 15 anti-patterns: references/ANTI_PATTERNS.md

Evidence Quality Notice

[QUALITY NOTICE] This crystal was compiled from blueprint finance-bp-121. Evidence verify ratio = 31.8% and audit fail total = 35. 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	15 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-121 blueprint at 2026-04-22T13:00:59.543591+00:00. See human_summary.md for non-technical overview.