portfolio-swarm-reviewlisted

# Portfolio Swarm Review Multi-agent portfolio review. Structure mirrors the TauricResearch/TradingAgents design — analysts (Tier 1) → researchers (Tier 2) → risk debators + judge (Tier 3). Each tier is distinct in the output; the Judge synthesizes. (The reference repo is no longer cloned locally; the structure is recorded here.) ## Required reads In this order: 1. `/root/.openclaw/workspace/MEMORY.md` 2. `/root/.openclaw/workspace/portfolio.json` 3. `/root/.openclaw/workspace/memory/current-portfolio-summary.md` 4. `/root/.openclaw/workspace/INVESTMENT_SOP.md` 5. Recent daily memory files when recent trades affect interpretation 6. `/root/.openclaw/workspace/TOOLS.md` for data chain detail ## Fresh data rule Refresh quotes before producing conclusions: ```bash python3 /root/.openclaw/workspace/scripts/data/analyze_us_stocks.py # US 7-route fallback python3 /root/.openclaw/workspace/scripts/data/analyze_hk_stocks.py # HK Tencent → stooq → yfinance ``` If a leg is stale, name the exact ticker and limit confidence on conclusions involving it. **00100 only has Tencent** — flag explicitly if that leg fails. KR linkage (07709/07747) is exited; do not run any KR-side fetch. ## Holdings bucketing — read each run, do not hardcode Pull live set from `portfolio.json` (`shares > 0`) and `current-portfolio-summary.md`. Stable bucket structure; contents drift: - **US growth / single-name beta** — active US non-leveraged growth names - **US leverage ETF** — anything `is_lev