feat(cli): composable analyze run command with --modules flag

.claude/skills/haoinvest/SKILL.md

@@ -12,15 +12,15 @@ All-in-one investment management via CLI + Claude Code agent. CLI does data + co
 
 ### Workflow 1: "帮我分析 XXX" — Analyze a stock
 
-1. Run comprehensive report:
+1. Run composable analysis (single call, all modules including peer):
    ```bash
-   uv run haoinvest analyze report <symbol>
+   uv run haoinvest analyze run <symbol>
    ```
-2. For A-shares, also run peer comparison:
+   Or select specific modules:
    ```bash
-   uv run haoinvest analyze peer <symbol>
+   uv run haoinvest analyze run <symbol> --modules fundamental,risk,peer
    ```
-3. Interpret ALL sections in Chinese:
+2. Interpret ALL sections in Chinese:
    - 估值: Is it cheap or expensive? Compare PE/PB to peers.
    - 财务健康: Is the company profitable and growing?
    - 风险: How volatile is it? What's the worst drawdown?
@@ -51,9 +51,9 @@ All-in-one investment management via CLI + Claude Code agent. CLI does data + co
 
 **触发条件**: 用户表达买入/卖出意图时。
 
-1. Run comprehensive report + guardrails pre-trade data (2 calls):
+1. Run composable analysis + guardrails pre-trade data (2 calls):
    ```bash
-   uv run haoinvest analyze report <symbol> --json
+   uv run haoinvest analyze run <symbol> --json
    uv run haoinvest guardrails pre-trade-data <symbol> <buy/sell> <qty> -m <type> --json
    ```
    If user hasn't specified quantity, ask first. If price not known, the command auto-fetches.
@@ -129,15 +129,11 @@ All-in-one investment management via CLI + Claude Code agent. CLI does data + co
 
 ### Workflow 4: "对比 A 和 B" — Compare stocks
 
-1. Batch fundamental comparison:
+1. Batch composable analysis (single call):
    ```bash
-   uv run haoinvest analyze fundamental <A>,<B> --verbose
+   uv run haoinvest analyze run <A>,<B> --modules fundamental,risk,signals
    ```
-2. Batch technical comparison:
-   ```bash
-   uv run haoinvest analyze technical <A>,<B>
-   ```
-3. Summarize: who's better on what dimension, and overall recommendation
+2. Summarize: who's better on what dimension, and overall recommendation
 
 ### Workflow 5: "定期体检" — Portfolio checkup
 
@@ -179,6 +175,13 @@ uv run haoinvest market sector <name>                          # Sector constitu
 
 ### Analysis
 ```bash
+# Composable analysis (preferred — single call, choose modules)
+uv run haoinvest analyze run <symbol(s)>                       # All modules (fundamental,technical,risk,volume,signals,peer,checklist)
+uv run haoinvest analyze run <symbol> --modules fundamental,risk,peer  # Selective modules
+uv run haoinvest analyze run <symbol> --json                   # JSON output for structured parsing
+uv run haoinvest analyze run <A>,<B> --modules fundamental     # Batch comparison
+
+# Individual commands (still available)
 uv run haoinvest analyze report <symbol>                       # Full report + buy-readiness checklist
 uv run haoinvest analyze fundamental <symbol(s)> [--verbose]   # Valuation + financial health (batch OK)
 uv run haoinvest analyze technical <symbol(s)>                 # MA/MACD/RSI/BB (batch OK)

CLAUDE.md

@@ -35,8 +35,10 @@ haoinvest/
 ├── journal.py         # Investment journal with emotion/decision tagging
 ├── cli/               # Typer CLI — entry point: `uv run haoinvest`
 │   ├── __init__.py    # App + subcommand registration
-│   ├── formatters.py  # Output formatting (text/JSON)
-│   ├── analyze.py     # analyze subcommand
+│   ├── _shared.py     # Shared CLI utilities (init_db, fetch_current_prices)
+│   ├── formatters.py  # Output formatting (text/JSON/section headers)
+│   ├── analyze.py     # analyze subcommand (includes composable `run` command)
+│   ├── guardrails.py  # guardrails subcommand
 │   ├── journal.py     # journal subcommand
 │   ├── market.py      # market subcommand
 │   ├── portfolio.py   # portfolio subcommand
@@ -48,6 +50,8 @@ haoinvest/
 │   └── optimization_engine.py # Portfolio optimization (HRP, min vol, max Sharpe)
 ├── portfolio/         # Trade recording, position tracking, returns (TWR)
 ├── analysis/          # Thin adapters over engine
+│   ├── cache.py       # Price data caching (ensure_prices_cached)
+│   ├── registry.py    # Module registry for composable `analyze run`
 │   ├── fundamental.py # Valuation assessment (PE/PB/ROE), financial health
 │   ├── technical.py   # Technical indicator adapter
 │   ├── risk.py        # Risk metrics adapter

README.md

@@ -13,6 +13,7 @@ Built for a beginner investor in China covering A-shares, US stocks, HK stocks,
 - **Fundamental Analysis** — PE/PB/ROE valuation assessment with financial health scoring; batch support for multi-symbol comparison
 - **Peer Comparison** — Find and compare same-sector stocks by valuation and performance
 - **Sector Browsing** — Browse A-share industry sectors and their constituent stocks
+- **Composable Analysis** — `analyze run` command with `--modules` flag to compose any combination of fundamental, technical, risk, volume, signals, peer, and checklist in a single call
 - **Comprehensive Report** — Full stock report with buy-readiness checklist combining fundamental, technical, and risk analysis
 - **Risk Metrics** — Annualized volatility, max drawdown, Sharpe ratio, Sortino ratio (powered by QuantStats)
 - **Technical Analysis** — MA, MACD, RSI, Bollinger Bands with Chinese explanations (powered by pandas-ta)
@@ -47,7 +48,12 @@ uv run haoinvest portfolio list                   # View holdings
 uv run haoinvest portfolio add-trade 600519 buy 100 1800.50
 uv run haoinvest portfolio returns                # P&L summary
 
-# Analysis
+# Composable analysis (preferred — single call, choose modules)
+uv run haoinvest analyze run 600519               # All modules
+uv run haoinvest analyze run 600519 --modules fundamental,risk,peer  # Selective
+uv run haoinvest analyze run 600519,000858 --modules fundamental     # Batch
+
+# Individual analysis commands
 uv run haoinvest analyze fundamental 600519       # PE/PB valuation
 uv run haoinvest analyze fundamental 600519,000858 # Batch comparison
 uv run haoinvest analyze risk --symbol NVDA       # Volatility, Sharpe, drawdown

haoinvest/analysis/cache.py

@@ -0,0 +1,31 @@
+"""Price data caching — ensure price history is available before analysis."""
+
+from datetime import date, timedelta
+
+from ..db import Database
+from ..market import get_provider
+from ..models import MarketType
+
+
+def ensure_prices_cached(
+    db: Database, symbol: str, market_type: MarketType, start: date, end: date
+) -> None:
+    """Fetch and cache price history if not already present.
+
+    Includes gap-fill: if cached data doesn't cover the requested start date,
+    fetches the missing earlier portion.
+    """
+    existing = db.get_prices(symbol, market_type, start, end)
+    if len(existing) > 10:
+        earliest_cached = min(b.trade_date for b in existing)
+        if earliest_cached <= start + timedelta(days=7):
+            return
+        provider = get_provider(market_type)
+        bars = provider.get_price_history(symbol, start, earliest_cached)
+        if bars:
+            db.save_prices(bars)
+        return
+    provider = get_provider(market_type)
+    bars = provider.get_price_history(symbol, start, end)
+    if bars:
+        db.save_prices(bars)