Rqalpha Cn Backtest

基于20日价格动量在沪深300、沪深500与国债之间自动轮转配置，通过RQAlpha框架执行完整回测并评估组合绩效。

Security Scan

The skill's description (RQAlpha A-share backtest) is plausible, but there are clear mismatches between what the SKILL.md/seed.yaml instructs and the declared metadata (no install spec, no required env vars) — the skill will ask the agent to run installs, write to ~/.zvt, and rely on provider accounts without declaring them.

Purpose & Capability

Purpose (backtesting A‑share rotation strategy via RQAlpha/ZVT) matches the included content (lots of domain constraints, anti‑patterns, pipelines). However SKILL.md claims runtime requirements (Python 3.12+ and 'uv' package manager, Doramagic-host compatibility, compiled blueprint v6.1) that are not reflected in the registry metadata (no required binaries, no install spec). Also the skill references third‑party data providers (joinquant, eastmoney, akshare) where credentials may be needed but no credentials are declared.

Instruction Scope

SKILL.md and seed.yaml instruct the agent to run precondition commands (python checks, pip install zvt if missing), to re-read seed.yaml at runtime, and to create/write files under ZVT_HOME (~/.zvt). They also direct the agent to choose data providers that can require account tokens. These are normal for a backtest helper, but the instructions give the agent permission to run shell/Python commands and write to disk; the skill does not clearly limit or declare those operations in the manifest.

Install Mechanism

The registry shows no install spec and no code files (instruction‑only), but seed.yaml's execution_protocol and SKILL.md expect the host/agent to perform installs/verification (pip installs, verify packages). That mismatch (no declared install steps vs instructions that perform installs) is an inconsistency and increases risk because the agent could be instructed to run arbitrary package installs at runtime.

Credentials

The skill declares no required environment variables, yet the SKILL.md and preconditions reference ZVT_HOME and data providers (joinquant, brokers) that commonly require credentials. The skill may prompt the agent/user for provider credentials at runtime but has not declared or scoped them in the manifest; this is a proportionality/visibility gap.

Persistence & Privilege

always:false and disable-model-invocation:false (normal). The skill does instruct the agent to re-read seed.yaml on every behavioral decision and to write into ~/.zvt as part of precondition checks; this gives the skill runtime persistence (filesystem writes) but not elevated platform privileges or forced always-on inclusion. No evidence it modifies other skills or global agent settings.

What to consider before installing

This skill appears coherent with a backtesting helper, but there are important mismatches you should address before installing or granting the agent permission to run it: - Expect the agent to run Python commands and possibly pip install packages (e.g., zvt). Run these installs yourself in a controlled environment (virtualenv/container) rather than letting the agent run them automatically. - The skill references ZVT_HOME (~/.zvt) and will attempt to write there as a precondition test; ensure you are comfortable with that path and permissions or set ZVT_HOME to an isolated directory. - If you plan to use paid data providers (joinquant, brokers), the skill will need credentials — the package manifest does not declare these env vars. Do not paste credentials into a conversation; prefer configuring them in the host's secret store and only grant minimal-scoped tokens. - Ask the publisher for a concrete install spec and a list of required env vars (jqdatasdk tokens, broker API keys) and for a minimal reproducible example of the exact shell/python commands the agent will run. The current mismatch (no install spec vs SKILL.md that runs installs) is the main reason this is flagged suspicious. If you want to proceed safely: run the skill locally inside an isolated VM/container, manually perform and inspect the prereq installs, set ZVT_HOME to a disposable directory, and do not provide credentials until you confirm what calls will be made and where data will be transmitted.

RQAlpha A 股回测 (rqalpha-cn-backtest)

基于20日价格动量在沪深300、沪深500与国债之间自动轮转配置，通过RQAlpha框架执行完整回测并评估组合绩效。

Pipeline

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

Top Use Cases (2 total)

Index Futures Momentum Rotation Strategy (UC-101)

Implements a momentum-based rotation strategy between equity indices (CSI 300, CSI 500) and government bonds, automatically rebalancing to the best-pe Triggers: momentum rotation, index futures, equity bond allocation

Sphinx Documentation Configuration (UC-102)

Configuration file for building rqalpha project documentation using Sphinx, setting up autodoc, autosummary, and other documentation extensions Triggers: documentation, sphinx, configuration

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-089. Evidence verify ratio = 44.4% and audit fail total = 12. 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-089 blueprint at 2026-04-22T13:00:37.233732+00:00. See human_summary.md for non-technical overview.