Fetching metadata from the HF Docker repository...
Sep 21 10:29 Sector-gap sourcing six deep, gap-driven second batch, gap-covering shortlist, per-sector search record, narrator share rules 09:56 Manager desk: conviction-ranked funding with cash residual, watch tier, min-size exits, tax-lot plan, watchlist with entry conditions, catalysts, post-trade risk, 12m scenarios, rule-of-40 valuation for loss-makers, specific tripwires 09:19 47-name report review fixes + build hash in health for deploy verification 09:18 47-name report review: CIO fallback + lenient JSON, bank model only for book-value financials, depressed-EPS blend, reinvestment cap, size-trim to cap, band-relative tiny rule, one add per correlated pair, distress disqualifies candidates, travel sector map fix 08:32 Pre-split warning wire code renamed so stale cached clients prompt instead of auto-rewriting 07:06 Front-end cache-busting: versioned asset URLs and no-cache revalidation so a redeploy is never masked by a stale app.js Sep 20 12:47 Pre-split entries ask the user (adjust both, price only, keep) instead of a silent rewrite; data-layer canary so renamed tickers (SQ→XYZ) resolve while real outages read as retry 12:35 Resolver hardening: data-source failures are retryable (LookupUnavailable), per-row endpoint never 500s, client retries once then click-to-retry Sep 7 12:09 Random-book fixes: staged-trim state reset, sector ETFs are sector bets, cautious bar for flagged names and high-beta adds, sub-2% trims exit or hold, broad-fund harvest stand-ins, coverage-weighted valuation, static SPY sector fallback, every new buy explained 11:22 Battery fixes: standard-tier thinking headroom, forced Trim orders, starter ceilings, staged trims, second committee batch, bound-respecting normalisation 10:16 Tail cleanup exits band Trims but keeps asymmetric small bets; triage agrees with the trade list; sector fill; weight-aware rationales 09:44 Wider candidate funnel with two-draw committee, alternatives floor of two stocks, damage tilts before add gates, Trim always reduces 09:26 Order plan never adds to hurdle-gated Keeps; sub-1% rule as final pass; overvalued candidates need a strong committee; narrative at medium effort 09:02 Stock-picker plans never hold an index fund unless the profile is cautious or a loss tolerance is stated 08:57 CIO reply must carry its verdict (truncation was silently accepted); bounded CIO reasoning; weak-trend names never strong adds; breadth card scales with book size 08:43 Two-lens dissent restricted to cyclicals (sector or industry): secular winners are not trimmed early 08:38 Verdict coherence: oversized tilt at 1.25x cap, two-lens dissent, no adds in the overweight sector, strong-only alternatives, triage discount wording 07:57 Report tuning: rationale writer split from CIO, value-gap text, verdict-consistent order bounds, size-scaled diversification target, sector-aware narrative 07:28 Cheap tiers: memos on GPT-5.6 Luna, tiny lookups on Gemini 3.1 Flash Lite, model override + provider guard 07:03 Token cost strategy: usage accounting, effort levels, thinking off for extraction, Sonnet for mechanics, slim CIO payload, one sample from 20 names 06:44 Split-aware import with auto-correction, table import without LLM, parallel strategist+candidates, stock-picker instrument rules, steeper concentration 06:07 Speed: Sharadar price layer, batched prefetch, pre-warm on import, batched resolve, split CIO, time budget, street targets Sep 6 13:18 Six-pillar health score with visible breakdown; prospects pillar; profile-weighted drawdown efficiency 12:57 Stock-picker alternatives, starter sizing, beta floor, tiny-trim rule, share-class Sharadar fallback, PDF fixes 10:59 Random-battery fixes: plan coherence, sector-overweight gate, ballast protection, ADR per-share, loss-tolerance constraint + de-risk scenario 10:15 US practitioner layer: forensic scores, owner economics, valuation history, ownership, asymmetry hurdle, memos, harvesting 09:45 Lot averaging: several purchases per ticker, sales netted, per-lot tax periods 09:13 European institutional layer: factor panel, sector UCITS, UK/EU duties, wrapper 07:52 company name fallback from Symbol Search when the venue has no overview 07:50 ADR bridge: licensed Sharadar ratio fundamentals for non-US holdings via their NYSE/NASDAQ ADR; local EPS/BVPS derived from price x P/E, P/B 07:39 resolver: exact-symbol rule applies only to direct ticker searches; name-expanded searches use whole-word matching 07:31 resolver: exact base-symbol match for ticker queries, whole-word name matching for name queries 07:28 resolver: secondary-venue demotion, US-first for US-shaped tickers, cached-search fix, demerged-ticker recovery 07:25 emerging markets: full Alpha Vantage venue map (India via BSE line, China, Frankfurt, Lisbon, Brussels, TSX-V), symbol-search resolver, EM benchmarks, India instruments, demerged-ticker recovery 07:15 deterministic ISIN resolution via OpenFIGI (primary listing by country), ISIN-aware fallback 07:12 UK/EU support: STOXX 600 benchmark for mixed European books, UCITS instruments, multi-exchange local ideas, jurisdiction-neutral tax notes, non-US split fallback 07:02 construction layer (risk decomposition, joint selection, constrained scenarios, post-trade proof, instruments, guards, exclusions) + Alpha Vantage premium as primary data source + policy backtest Sep 3 11:46 candidate guards (peak-cycle, extended, informed money leaving) + analyst committee screen on prospective additions; switches/target need a clean committee verdict 11:14 portfolio-aware recommendation engine: diagnosed needs, marginal-fit scoring (correlation/effective-N/beta/stress/yield/DCF), fit-eligible candidates only, overweight-aware switches 10:48 institutional layer: Sharadar-primary data + sanity gates, valuation ranges/reverse DCF/sector models, risk analytics (stress, VaR, look-through, tilts, liquidity), domicile benchmark, lens sampling + audit trail, tax/cost-aware plan, suitability, regression suite 09:40 quality round 3: adaptive name cap, REIT/FFO + GBp fixes, context-fit candidates, lens retry, percent-safe narrative, yield units 09:40 quality round 3: adaptive name cap, REIT/FFO + GBp fixes, context-fit candidates, lens retry, percent-safe narrative, yield units 09:40 quality round 3: adaptive name cap, REIT/FFO + GBp fixes, context-fit candidates, lens retry, percent-safe narrative, yield units 09:40 quality round 3: adaptive name cap, REIT/FFO + GBp fixes, context-fit candidates, lens retry, percent-safe narrative, yield units 09:40 quality round 3: adaptive name cap, REIT/FFO + GBp fixes, context-fit candidates, lens retry, percent-safe narrative, yield units 09:17 feature: entry anomaly validation — split detection, wrong-listing LLM check, one-click corrections 09:17 feature: entry anomaly validation — split detection, wrong-listing LLM check, one-click corrections 09:17 feature: entry anomaly validation — split detection, wrong-listing LLM check, one-click corrections 09:17 feature: entry anomaly validation — split detection, wrong-listing LLM check, one-click corrections 09:17 feature: entry anomaly validation — split detection, wrong-listing LLM check, one-click corrections 
