portfolio-rebalancerlisted

# portfolio-rebalancer You hand over a positions map plus book value, a per-name variance- share cap, a per-name weight cap, and a max churn per rebalance. The skill returns a specific trade-ticket list with dollar amounts, weight deltas, and before/after variance-share readings. risk-report tells you which name is driving the risk. This skill tells you what to trim, by how much, and where to redistribute. ## When to invoke - After a risk-report run flags a name with variance share far disproportionate to its weight - Portfolio-review workflow, decision-support step - The user says "rebalance", "trim my winners", "cut variance", "what should I sell", "reduce concentration" - Any time the operator wants an actionable answer, not a report ## What you need - `MASSIVE_API_KEY` — Stocks Starter for daily aggs on every name + benchmark ## What you get back **Layer 1 JSON** matching [`output-schema.json`](./output-schema.json). Per-name trade tickets sorted by absolute dollar amount, plus portfolio-level before/after summary (vol, top-3 variance share, Herfindahl, max variance share), constraint-satisfaction status. **Layer 2 rendered table**. Before/after summary block, then a table of trades, then a status line. See [`references/rendering.md`](./references/rendering.md). ## How it works 1. **Parse positions** from either a comma-separated string (`TICKER=WEIGHT`) or a book JSON file. Same shape as risk-report for consistency. 2. **Pull daily aggs** for every