Update to Analyzera

{
  "description": "Build, create or set up an investment portfolio or strategy with Analyzera, such as \"build a portfolio with AI stocks\", \"make me a dividend portfolio\", \"I want a Swedish tech portfolio\" or \"a 60/40 with funds\". Use before calling any Analyzera screening, weighting or build tool when the client has not said how they want to build it.",
  "included_files": [],
  "name": "analyzera-build-portfolio",
  "skill_md_contents": "---\r\nname: analyzera-build-portfolio\r\ndescription: Build, create or set up an investment portfolio or strategy with Analyzera, such as \"build a portfolio with AI stocks\", \"make me a dividend portfolio\", \"I want a Swedish tech portfolio\" or \"a 60/40 with funds\". Use before calling any Analyzera screening, weighting or build tool when the client has not said how they want to build it.\r\n---\r\n\r\n# Building a portfolio with Analyzera\r\n\r\n\"Build a portfolio\" can mean several things, and each ends differently. Find out\r\nwhich one the client means before calling a screening, weighting or build tool.\r\n\r\n## 1. Check the catalogue\r\n\r\nFor a theme, sector or style (AI, defence, energy, dividends), call\r\n`find_portfolios` with it as `query`, for example `query='AI'`. It is a read and\r\nsaves nothing. Its result tells you whether a ready-made strategy exists.\r\n\r\n## 2. Ask which route\r\n\r\nAsk one question, offering the routes that apply:\r\n\r\n1. **Use a ready-made strategy**: only when step 1 found one. Name it.\r\n2. **Build their own strategy now**: saved, rebalances automatically, can be\r\n   backtested. `build_equity_strategy` for single stocks,\r\n   `build_allocation_strategy` for funds, ETFs or other strategies.\r\n3. **Explore by hand first**: screen stocks and try weightings, nothing saved.\r\n   `screen_equities`, then `solve_equity_weights`.\r\n\r\nPresent the routes neutrally and recommend none. The client chooses.\r\n\r\n### Example\r\n\r\n> **Client:** I want to build a portfolio with AI stocks\r\n>\r\n> `find_portfolios(query='AI')` finds \"Artificial Intelligence Theme\".\r\n>\r\n> **Ask:** \"Analyzera has a ready-made Artificial Intelligence Theme strategy.\r\n> Would you like to look at it, build your own AI strategy straight away, or\r\n> explore AI stocks by hand with the screener and weightings first?\"\r\n\r\n## 3. Follow the answer\r\n\r\n- **Ready-made**: show it with `open_dashboard` on its id, or answer from the\r\n  `describe_portfolio_*` reads.\r\n- **Build now**: ask for the construction choices the client has not stated\r\n  (markets, weighting, roughly how many holdings), say that building saves a\r\n  strategy, then build.\r\n- **Explore**: screen with `screen_equities`, weight with `solve_equity_weights`,\r\n  and measure with `describe_holdings`. Stay unsaved until the client asks to\r\n  save. Then save the screen: pass the result's `request` unchanged, plus a name,\r\n  to `build_equity_strategy`. Never copy the ids from a screen or weighting into\r\n  `asset_ids`.\r\n\r\n## A screen is saved as a screen, not as its result\r\n\r\nInvesting with a stock screener over time means running the screen again at\r\nevery rebalance, with only what was known on that date. That is what\r\n`build_equity_strategy` does with screen arguments, and it is what makes a\r\nbacktest of the approach honest.\r\n\r\nThe companies a screen or weighting lists are today's answer. Saved as\r\n`asset_ids` they are held through the whole history, including the years before\r\nanyone could have picked them. That backtest shows hindsight bias, not how the\r\napproach would have done. The same holds for a list assembled from today's\r\nscreen results, today's holdings, or both together.\r\n\r\nUse `asset_ids` only when the client brings their own list from outside the\r\nscreener (names they chose, a list they follow) and says they want it held as\r\nit is. If a backtest should match exactly the mix you just showed, explain that\r\nthe saved screen will hold different names in earlier years, and that this is\r\nthe point.\r\n\r\n## When not to ask\r\n\r\nA request that names its route goes ahead on it:\r\n\r\n- \"Screen Swedish dividend stocks\": screen.\r\n- \"Build me a 60/40 with these funds\": an allocation.\r\n- \"Record what I hold in my ISK\": `sync_external_account`.\r\n\r\n## Funds and ETFs\r\n\r\nThe client chooses the funds. Resolve the names they give with `find_assets`\r\n(kinds `fund` or `etf`). There is no fund screener, so never pick funds on the\r\nclient's behalf.\r\n\r\n## Factors and styles\r\n\r\nThe catalogue holds factor strategies (value, quality, momentum and others) side\r\nby side. Analyzera does not recommend either direction of any factor tilt, and\r\nhaving a strategy on the shelf is no endorsement. You may make a case for one from\r\nyour own sources and test it with the builders and backtests, but present it as\r\nyours, not Analyzera's. The client chooses.\r\n"
}