How a recommendation is born

When you invoke /rx-finance (or fitness, or learning), eleven steps fire in a fixed order. The first two reconcile yesterday's state. The middle steps compose the new rec. The last steps persist it. Every step is deterministic — the LLM does narrative, not arithmetic.

The 11 canonical steps

1. 0.5 · Auto-revive snoozed recs D-032

   Update the rec store: any row with snoozed_until ≤ NOW() flips back to status=open. Lazy — runs only when you invoke a rx command, not on a cron.

2. 0.7 · Phase W reconciler D-040

   Query the rec store for recent rows. For each, check the on-disk markdown frontmatter. If the DB says status=acted but the file still says open, patch the file. Idempotent — re-running is a no-op once synced.

3. 1 · Read drift signals from canonical source

   Fitness: SELECT * FROM v_drift_signals (7-component SQL view). Finance: ~6 SQL queries against hypotheses/opportunities/drift_alerts/positions. Learning: ganesh_state_snapshot.py (markdown helper). Numbers are copied verbatim (D-011) — no LLM math.

4. 2 · Read auxiliary signals

   Fitness: body_recomp_snapshot.py (10% second layer). Finance: watchlist context, position concentration. Learning: bundle helper output. Each door has one or two extras the SQL view doesn't cover.

5. 3 · Compose composite drift

   Weighted average. Fitness uses two layers: composite = 0.90 × view.drift + 0.10 × body_recomp_sub (D-022). Finance and learning are flat single-layer composites.

6. 4 · Auto-dedup check D-012

   If a prior open rec exists, was created < 48 hours ago, has the same driving signal, and a drift score within ±0.05 — and no acute signal has fired — skip writing a new rec. Instead, write a short "no-change confirmation" artifact and stop. Prevents inbox bloat from same-state re-runs.

7. 5 · Pick the driving signal

   Of the components that fed into the composite, the one with the highest weighted contribution becomes the driving signal. The rec body addresses this signal specifically — not the whole drift surface.

8. 6 · Pull evidence from the vault

   Call the vault indexer's /search endpoint with a query keyed off the driving signal. Top-K results (typically 3–5) become source citations. See Deep tech for retrieval mechanics.

9. 7 · Compose the rec artifact D-008

   YAML frontmatter (id, created_at, drift_score, drift_breakdown, confidence + breakdown, status=open, signals_fired, source_refs, prior_open_recs).
   Body in fixed sections: TL;DR, What I'm seeing, Recommendation, Why, Counter-thesis (mandatory), Sources.

10. 8 · Write markdown to disk

    File path <Door>/rx/rx-YYYY-MM-DD-NN.md, where NN is the next sequence number for that date.

11. 9 · INSERT into rec store

    Door-specific:
    · Fitness/nutrition → Supabase recommendations
    · Finance → TradingV postgres recommendations via POST /v1/rx/recs with X-RX-Ingest-Token
    · Learning → skipped (markdown is the only store)

12. 10 · Patch frontmatter with store ID

    After insert, the returned id is written back into the markdown's frontmatter (verocity_rx_id for fitness, tradingv_rec_id for finance). Learning skips this step.

13. 11 · Return TL;DR to chat

    Operator gets the artifact path + the one-line TL;DR. The full rec is then read either in the editor (markdown) or in the UI surface (Lovable / FastAPI panel).

Three load-bearing rules

The snooze + auto-revive lifecycle

A rec spends its life in one of these states:

Where the LLM is allowed to think