feat(profile+pricing): KI-269 — D1+D2 bundle: chip merge + panel pre-fill + copay + family history

D1 (frontend, page.tsx) — chip merge + panel pre-fill:
- Deleted the redundant "ESTIMATE Annual premium" CTA button.
- Made the "EST. PREMIUM ₹X-Y/yr" band chip itself clickable: same onClick
as the old CTA (toggles PremiumCalculatorPanel + closes sibling panels).
Added hover state (ring + brightness + cursor-pointer), chevron-edit
SVG, tooltip "Tap to refine premium with sliders" (en + hi).
- PremiumCalculatorPanel now reads `initialProfile` prop (passed from page
as `profileCompleteness?.profile`). Three pure derivation helpers
pre-populate the 5 sliders from the live profile:
Age → profile.age (fallback 35)
Sum insured → existing_cover_inr (fallback 10L; will widen to
desired_sum_insured_inr once UserProfile schema is regen'd)
Family covered → derived from profile.dependents string
Pre-existing conditions → mapped from profile.health_conditions
City tier → profile.location_tier (default metro)
- Users can still override via sliders post-open.

D2 (backend, 4 files) — copay_pct + family_medical_history slots end-to-end:

backend/needs_finder.py — Profile dataclass:
+ copay_pct: Optional[int] # 0-50, % of every claim user accepts
+ family_medical_history: list[str] # blood-family conditions

backend/brain_tools.py:
- SLOT_UNION constant extended with both new fields (now 15 slots total).
- _ACCEPTED_FIELDS adds copay_pct + family_medical_history.
- _coerce_copay_pct(value): int parse, accepts "20"/"20%"/20, clamps [0,50].
- _coerce_family_medical_history(value): list[str] lowercase, aliases
(BP → hypertension, sugar → diabetes), "none"/"no family history" → [].
- save_profile_field routes both new fields to coercers.
- Top docstring slot-→-consumer matrix updated.

backend/single_brain.py SYSTEM_PROMPT:
- RULE 2.5 extended with two new asks:
* "OK with co-pay 10-30% to lower premium?" → save copay_pct
* "Major conditions in your blood family — cancer/diabetes/heart?"
→ save family_medical_history
- RULE 2 retrieve_policies query construction now includes family-history
boost terms: if family has cancer → include "critical illness rider
cancer cover"; diabetes → "diabetes short waiting period"; heart →
"cardiac care rider".

backend/premium_calculator.py:
- _copay_discount(copay_pct): 0%→1.00× / 10%→0.95× / 20%→0.88× / 30%→0.80×
+ smooth interpolation for in-between values.
- _family_history_loading(family_medical_history):
* empty/["none"] → 1.00×
* 2+ family conditions → 1.10×
* cancer / heart → 1.05×
* other single → 1.03×
- estimate() folds copay_mult + fam_mult into the multiplicative chain
(already wired by D2 before stall).
- bulk_estimate() — finished D2's wiring: reads copay_pct + family_medical_history
from profile dict, computes copay_mult + fam_mult once, multiplies into
both the curated-anchor branch (via estimate() pass-through kwargs) AND
the flat-fallback branch. Surfaces both in the breakdown dict when ≠ 1.0×
(copay_discount_x / copay_discount_reason / family_history_loading_x /
family_history_loading_reason).

Live smoke (HDFC Optima Secure, age 40, metro, self+spouse):
- baseline: ₹14,400/year
- copay 20% + family cancer: ₹13,310/year (-7.6% net, matches model)
- breakdown shows copay_discount_x=0.88, family_history_loading_x=1.05

Verification:
- python -m py_compile clean on all 4 backend files
- npx tsc --noEmit clean
- bulk_estimate smoke produces expected loaded/discounted values

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

backend/brain_tools.py

@@ -46,7 +46,11 @@ Slot → consumer matrix:
     parents_age_max       → pricing (parents age loading 1.0× / 1.4× / 1.8×)
     parents_has_ped       → pricing (PED loading inflation for parents)
 
-Total: 13 slots. `gender` is tolerated by save_profile_field for forward
+  D2 ADDITIONS (2026-05-15 — copay + family medical history)
+    copay_pct             → pricing (copay discount 1.0× / 0.95× / 0.88× / 0.80×)
+    family_medical_history → pricing (family-history loading) + retrieval boost
+
+Total: 15 slots. `gender` is tolerated by save_profile_field for forward
 compat but is NOT on the Profile dataclass today; it does not appear in
 SLOT_UNION because no consumer reads it.
 """
@@ -83,6 +87,9 @@ _ACCEPTED_FIELDS = {
     "parents_to_insure",
     "parents_age_max",
     "parents_has_ped",
+    # D2 (2026-05-15) — coupled additions: co-pay tolerance + family medical history
+    "copay_pct",
+    "family_medical_history",
     "gender",  # tolerated; not persisted unless Profile gains the field
 }
 
@@ -128,6 +135,9 @@ SLOT_UNION: tuple[str, ...] = (
     "parents_to_insure",
     "parents_age_max",
     "parents_has_ped",
+    # D2 additions (2026-05-15)
+    "copay_pct",
+    "family_medical_history",
 )
 
 # Invariant: every SLOT_UNION field must be accepted by save_profile_field
@@ -218,6 +228,10 @@ def save_profile_field(session, field: str, value: Any) -> dict:
             normalized = _coerce_bool(value)
         elif fld == "parents_age_max":
             normalized = _coerce_age(value)
+        elif fld == "copay_pct":
+            normalized = _coerce_copay_pct(value)
+        elif fld == "family_medical_history":
+            normalized = _coerce_family_medical_history(value)
         elif fld == "name":
             normalized = (str(value).strip() if value is not None else None) or None
         elif fld == "gender":
@@ -804,6 +818,143 @@ def _coerce_health_conditions(value: Any) -> Optional[list[str]]:
     return real
 
 
+# ---------------------------------------------------------------------------
+# D2 (2026-05-15) — copay_pct + family_medical_history coercers
+# ---------------------------------------------------------------------------
+
+# Word-number map for "twenty", "ten" etc. (RULE 2.5 asks the user in
+# multiples of 10; Gemini sometimes echoes the user's word verbatim).
+_COPAY_WORD_TO_INT: dict[str, int] = {
+    "zero": 0, "none": 0, "no": 0,
+    "ten": 10, "fifteen": 15, "twenty": 20,
+    "twenty five": 25, "twenty-five": 25,
+    "thirty": 30, "forty": 40, "fifty": 50,
+}
+
+
+def _coerce_copay_pct(value: Any) -> Optional[int]:
+    """Parse a co-pay tolerance percent, clamped to [0, 50].
+
+    Accepts:
+      - int / float  → int + clamp
+      - "20", "20%", "  20 % ", "20 percent" → 20
+      - "no copay" / "zero" / "none" → 0
+      - word numbers like "twenty" → 20
+      - bool → blocked (KI-091 null-overwrite caution: bool is an int subclass)
+
+    Returns None for unrecognised input so the null-overwrite guard in
+    save_profile_field can refuse to clobber a previously-captured slot.
+    """
+    if value is None:
+        return None
+    if isinstance(value, bool):
+        return None
+    if isinstance(value, (int, float)):
+        n = int(value)
+        return max(0, min(50, n))
+    s = str(value).strip().lower()
+    if not s:
+        return None
+    # Explicit zero phrasings.
+    if s in ("no", "none", "nil", "zero", "no copay", "no co-pay", "no co pay"):
+        return 0
+    # Word-number lookup (exact match).
+    if s in _COPAY_WORD_TO_INT:
+        return _COPAY_WORD_TO_INT[s]
+    # Strip "%" + "percent" + "pct".
+    cleaned = (
+        s.replace("%", " ")
+         .replace("percent", " ")
+         .replace("pct", " ")
+         .replace("copay", " ")
+         .replace("co-pay", " ")
+         .replace("co pay", " ")
+    )
+    # Digit run.
+    import re as _re
+    m = _re.search(r"\d+(?:\.\d+)?", cleaned)
+    if m:
+        try:
+            n = int(float(m.group(0)))
+            return max(0, min(50, n))
+        except ValueError:
+            return None
+    # Word-number fall-through (substring on cleaned text).
+    for word, num in _COPAY_WORD_TO_INT.items():
+        if word in cleaned.split():
+            return num
+    return None
+
+
+# Alias map for family medical history — same canonicalisation logic as
+# health_conditions but kept inline so this slot stays self-contained.
+_FAMILY_HISTORY_ALIASES: dict[str, str] = {
+    "bp": "hypertension",
+    "high bp": "hypertension",
+    "high-bp": "hypertension",
+    "hi-bp": "hypertension",
+    "high blood pressure": "hypertension",
+    "blood pressure": "hypertension",
+    "sugar": "diabetes",
+    "diabetic": "diabetes",
+    "type 2 diabetes": "diabetes",
+    "type 1 diabetes": "diabetes",
+    "heart attack": "heart",
+    "heart disease": "heart",
+    "cardiac": "heart",
+    "cardiac disease": "heart",
+    "stroke": "heart",
+    "tumor": "cancer",
+    "tumour": "cancer",
+    "carcinoma": "cancer",
+}
+
+_FAMILY_HISTORY_NEGATION = {
+    "none", "no", "n/a", "na", "nil", "nothing", "healthy",
+    "no family history", "no history", "no medical history",
+}
+
+
+def _coerce_family_medical_history(value: Any) -> Optional[list[str]]:
+    """Return list[str] lowercase canonical conditions running in BLOOD family.
+
+    Accepts:
+      - list / tuple of strings
+      - comma-joined string ("cancer, diabetes")
+      - "none" / "no family history" → []
+
+    Alias map collapses BP/sugar/cardiac/tumor → hypertension/diabetes/heart/
+    cancer respectively (same family as _coerce_health_conditions). Negation
+    sentinels return `[]` since downstream pricing & retrieval BOTH treat an
+    empty list as the "no family history" branch (different from health_
+    conditions where the explicit `["none"]` sentinel is needed for the
+    profile-completeness gate).
+    """
+    if value is None:
+        return None
+    if isinstance(value, str):
+        items = [t.strip() for t in value.split(",")]
+    elif isinstance(value, (list, tuple)):
+        items = [str(t).strip() for t in value]
+    else:
+        items = [str(value).strip()]
+    cleaned = [t.lower() for t in items if t]
+    # Full-string negation collapses to [].
+    if cleaned and all(t in _FAMILY_HISTORY_NEGATION for t in cleaned):
+        return []
+    # Drop negation noise from mixed input ("cancer, none").
+    cleaned = [t for t in cleaned if t not in _FAMILY_HISTORY_NEGATION]
+    # Canonicalise via alias map.
+    canonical: list[str] = []
+    seen: set[str] = set()
+    for t in cleaned:
+        c = _FAMILY_HISTORY_ALIASES.get(t, t)
+        if c and c not in seen:
+            seen.add(c)
+            canonical.append(c)
+    return canonical
+
+
 __all__ = [
     "save_profile_field",
     "retrieve_policies",

backend/needs_finder.py

@@ -42,6 +42,12 @@ class Profile:
     budget_band: Optional[str] = None  # "under_15k", "15k_30k", "30k_60k", "60k+"
     desired_sum_insured_inr: Optional[int] = None  # SOFT pricing input (post-recap)
     health_conditions: Optional[list[str]] = field(default_factory=list)  # ["diabetes", "hypertension", ...]
+    # D2 (2026-05-15) — co-pay tolerance + family medical history. Coupled
+    # SLOT_UNION additions captured via RULE 2.5 post-recap, both flow into
+    # premium_calculator (copay discount + family-history loading) and
+    # retrieval (family-history rider boost keywords).
+    copay_pct: Optional[int] = None  # 0-50, % of every claim user accepts
+    family_medical_history: list[str] = field(default_factory=list)  # blood-family conditions
     asked: list[str] = field(default_factory=list)  # question IDs / field names already asked
     free_form_session: bool = False  # True = user asks free questions, not driven by us
     # KI-063 (2026-05-15) — per-user policy interaction log so the bot

backend/premium_calculator.py

@@ -29,6 +29,9 @@ location / family_size that B2 already handles):
   parents_age_max         → parents_loading 1.0× / 1.4× / 1.8×
                             (only when `dependents` mentions "parents")
   parents_has_ped         → adds +0.10× on top of parents_loading
+  copay_pct (D2)          → copay_discount 1.0× / 0.95× / 0.88× / 0.80×
+  family_medical_history  → family_history_loading 1.0× / 1.03× / 1.05× / 1.10×
+  (D2)                      (cancer/heart +5%, 2+ conditions +10%, other +3%)
 
 Slots that are profile-only (no pricing effect): name, primary_goal,
 income_band, budget_band (matched against output, not folded into the
@@ -189,6 +192,95 @@ def _parents_loading(dependents, parents_age_max, parents_has_ped=None) -> tuple
     return base, label
 
 
+# ───────────────────────────────────────────────────────────────────────────
+# D2 (2026-05-15) — copay_pct + family_medical_history loadings
+# ───────────────────────────────────────────────────────────────────────────
+
+def _copay_discount(copay_pct) -> tuple[float, str]:
+    """Return (multiplier, label) for SLOT_UNION's `copay_pct` slot.
+
+    Distinct from the legacy `_copay_multiplier` (formula-based, used by the
+    `copayment_pct` arg on estimate()). This is a profile-driven step-discount
+    grid keyed to the 4 buckets RULE 2.5 asks the user about (0/10/20/30):
+
+      0%  → 1.00× ("no copay")          — insurer pays it all (highest premium)
+      10% → 0.95× ("10% copay")          — mild tier
+      20% → 0.88× ("20% copay")          — typical
+      30% → 0.80× ("30% copay")          — aggressive
+      other → linear interpolate between the two nearest buckets, clamped to [0,50]
+    """
+    if copay_pct is None:
+        return 1.0, "no_copay"
+    try:
+        pct = int(copay_pct)
+    except (TypeError, ValueError):
+        return 1.0, "no_copay"
+    if pct <= 0:
+        return 1.0, "no_copay"
+    # Clamp to [0, 50] to match _coerce_copay_pct.
+    pct = min(50, pct)
+    # Step grid (exact buckets).
+    if pct == 10:
+        return 0.95, "10_pct_copay"
+    if pct == 20:
+        return 0.88, "20_pct_copay"
+    if pct == 30:
+        return 0.80, "30_pct_copay"
+    # Linear interpolation for off-grid values (e.g. 15, 25, 40).
+    grid = [(0, 1.00), (10, 0.95), (20, 0.88), (30, 0.80), (50, 0.70)]
+    for i in range(len(grid) - 1):
+        p0, m0 = grid[i]
+        p1, m1 = grid[i + 1]
+        if p0 <= pct <= p1:
+            t = (pct - p0) / (p1 - p0) if p1 != p0 else 0
+            mult = m0 + (m1 - m0) * t
+            return round(mult, 3), f"{pct}_pct_copay"
+    return 1.0, "no_copay"
+
+
+# Family medical history canonical condition keywords. Matches the canonical
+# tokens emitted by brain_tools._coerce_family_medical_history (cancer /
+# diabetes / heart / hypertension).
+_FAM_CANCER_KEYWORDS = {"cancer"}
+_FAM_HEART_KEYWORDS = {"heart"}
+
+
+def _family_history_loading(family_medical_history) -> tuple[float, str]:
+    """Return (multiplier, label) for blood-family medical history.
+
+    Logic (D2 spec):
+      • empty list / None / ["none"]  → (1.00, "no_family_history")
+      • 2+ family conditions          → (1.10, "multi_family_history")
+                                        (highest — compounded genetic risk)
+      • contains "cancer"             → (1.05, "family_cancer")
+      • contains "heart"              → (1.05, "family_heart")
+      • other single condition (e.g. diabetes / hypertension) → (1.03, "family_history")
+
+    Order: 2+ check FIRST so a profile with both cancer + diabetes lands on
+    the multi-family multiplier (not the cancer-only +5%).
+    """
+    if not family_medical_history:
+        return 1.0, "no_family_history"
+    if isinstance(family_medical_history, str):
+        items = [t.strip().lower() for t in family_medical_history.split(",") if t.strip()]
+    else:
+        items = [str(t).strip().lower() for t in family_medical_history if str(t).strip()]
+    # Drop the "none" sentinel if a caller passed it (defensive).
+    items = [t for t in items if t != "none"]
+    if not items:
+        return 1.0, "no_family_history"
+    # 2+ conditions wins — compounded genetic risk loading.
+    if len(items) >= 2:
+        return 1.10, "multi_family_history"
+    # Single condition — bucket by keyword.
+    single = items[0]
+    if any(k in single for k in _FAM_CANCER_KEYWORDS):
+        return 1.05, "family_cancer"
+    if any(k in single for k in _FAM_HEART_KEYWORDS):
+        return 1.05, "family_heart"
+    return 1.03, "family_history_single"
+
+
 # Co-pay reduces premium. Industry norm (PolicyBazaar/Acko): each 10 pct
 # points of co-pay yields ~7% premium reduction, capped at 40% co-pay.
 def _copay_multiplier(pct: float) -> float:
@@ -256,6 +348,9 @@ def estimate(
     dependents: Optional[str] = None,
     parents_age_max: Optional[int] = None,
     parents_has_ped: Optional[bool] = None,
+    # D2 additions (2026-05-15) — copay_pct + family_medical_history.
+    copay_pct: Optional[int] = None,
+    family_medical_history: Optional[list] = None,
 ) -> PremiumEstimate:
     data = _load_data()
     base_premiums = data.get("base_premiums", {})
@@ -318,6 +413,14 @@ def estimate(
     )
     base *= parents_mult
 
+    # D2 — copay_pct discount + family_medical_history loading. Each is 1.0×
+    # when the corresponding SLOT_UNION field is None / empty, so legacy
+    # callers see no change.
+    copay_mult, copay_label = _copay_discount(copay_pct)
+    base *= copay_mult
+    fam_mult, fam_label = _family_history_loading(family_medical_history)
+    base *= fam_mult
+
     point = int(round(base / 100) * 100)  # round to nearest ₹100
     return PremiumEstimate(
         policy_id=policy_id or "generic",
@@ -488,6 +591,9 @@ def bulk_estimate(
     dependents = profile.get("dependents")
     parents_age_max = profile.get("parents_age_max")
     parents_has_ped = profile.get("parents_has_ped")
+    # D2 — copay_pct + family_medical_history (same read pattern).
+    copay_pct = profile.get("copay_pct")
+    family_medical_history = profile.get("family_medical_history")
     # desired_sum_insured_inr — when present, becomes the default SI for
     # any policy without an explicit overrides entry (per-policy override
     # still wins, since this is the DEFAULT).
@@ -504,6 +610,11 @@ def bulk_estimate(
     parents_mult, parents_label = _parents_loading(
         dependents, parents_age_max, parents_has_ped
     )
+    # D2 — copay_pct discount + family_medical_history loading. Each is 1.0×
+    # when the corresponding SLOT_UNION field is None / empty, so legacy
+    # callers see no change.
+    copay_mult, copay_label = _copay_discount(copay_pct)
+    fam_mult, fam_label = _family_history_loading(family_medical_history)
 
     out: dict[str, BulkPolicyPremium] = {}
     for pid in policy_ids:
@@ -547,6 +658,9 @@ def bulk_estimate(
                     dependents=dependents,
                     parents_age_max=parents_age_max,
                     parents_has_ped=parents_has_ped,
+                    # D2 — copay + family-history threaded through too
+                    copay_pct=copay_pct,
+                    family_medical_history=family_medical_history,
                 )
                 # estimate() already folded age/location/family AND the B6
                 # loadings — unwind so the widget can display the same
@@ -585,6 +699,8 @@ def bulk_estimate(
                 * health_mult
                 * ec_mult
                 * parents_mult
+                * copay_mult
+                * fam_mult
                 * tenure_mult
                 * ded_mult
             )
@@ -617,6 +733,12 @@ def bulk_estimate(
         if parents_mult != 1.0:
             breakdown["parents_loading_x"] = round(parents_mult, 3)
             breakdown["parents_loading_reason"] = parents_label
+        if copay_mult != 1.0:
+            breakdown["copay_discount_x"] = round(copay_mult, 3)
+            breakdown["copay_discount_reason"] = copay_label
+        if fam_mult != 1.0:
+            breakdown["family_history_loading_x"] = round(fam_mult, 3)
+            breakdown["family_history_loading_reason"] = fam_label
         if desired_si and not ov.get("sum_insured_inr"):
             breakdown["desired_si_default_inr"] = int(desired_si)
 

backend/single_brain.py

@@ -160,7 +160,13 @@ Required ingredients:
   health-condition keywords — every captured condition by name ("diabetes", "hypertension", "heart disease") OR the literal "no PED" when health_conditions == ["none"],
   primary goal keyword,
   existing cover signal — when existing_cover_inr > 0 add "top-up over existing X lakh cover"; when 0 add "fresh base policy",
-  parents-cover signal — when dependents mentions parents add "parents age ~XX" using parents_age_max (if captured).
+  parents-cover signal — when dependents mentions parents add "parents age ~XX" using parents_age_max (if captured),
+  family-history rider boost — if family_medical_history is non-empty, INCLUDE keywords in the query that bias retrieval toward policies with relevant coverage:
+    - "cancer" → "critical illness rider cancer cover"
+    - "diabetes" → "diabetes short waiting period reduced PED wait"
+    - "heart" → "cardiac care rider heart cover"
+    - "hypertension" → "hypertension short waiting period"
+    Multiple family conditions → concatenate the relevant phrases.
 
 Worked example A (no PED, no existing cover). Profile = {age=34, location_tier=metro, income_band=10L-25L, dependents=spouse+1 kid, primary_goal=first_buy, health_conditions=["none"], desired_sum_insured_inr=1500000, existing_cover_inr=0}:
   retrieve_policies(query="family floater plan metro sum insured 15 lakh adult 30-40 with spouse and one child no PED fresh base policy first-time buyer", top_k=8)
@@ -178,12 +184,16 @@ After all 7 slots are saved AND the user has confirmed the recap (RULE 4 implici
    1. How much sum insured? (e.g., ₹5L / ₹10L / ₹25L / ₹1Cr)
    2. Premium budget? (e.g., ₹10–15K/year, or ₹50K+ for premium covers)
    3. Any existing health cover from work or otherwise? (e.g., '5L through employer' or 'no')  [SKIP if existing_cover_inr already captured]
-   4. Approximate age of the eldest parent you'd cover?  [ASK ONLY IF dependents mentions parents AND parents_age_max not yet captured]"
+   4. Co-pay tolerance: Are you OK with a co-pay — sharing 10-30% of every claim — to lower the premium? Or do you want zero co-pay (insurer pays it all)?
+   5. Family medical history: Any major conditions running in your blood family (parents/siblings) — cancer / diabetes / heart disease / hypertension?
+   6. Approximate age of the eldest parent you'd cover?  [ASK ONLY IF dependents mentions parents AND parents_age_max not yet captured]"
 
 When the user answers, call save_profile_field once per provided value:
   save_profile_field(field="desired_sum_insured_inr", value="1000000")  # ₹10L
   save_profile_field(field="budget_band",            value="10K-20K")
   save_profile_field(field="existing_cover_inr",     value="500000")    # 5L corporate top-up; 'no' / 'none' → value="0"
+  save_profile_field(field="copay_pct",              value="0" or "10" or "20" or "30")  # 0 = no co-pay (higher premium), 10-30 = typical tiers
+  save_profile_field(field="family_medical_history", value="cancer, diabetes" or "none")  # blood family only (parents/siblings)
   save_profile_field(field="parents_age_max",        value="68")        # eldest parent's age, only if covering parents
 
 Gender hint: if the user mentions gender, keep it for conversational context only — Profile has no `gender` slot. Do NOT call save_profile_field(field="gender", ...) — it returns `field_not_on_profile_dataclass` and wastes a tool-call iteration.

frontend/src/app/page.tsx

@@ -1367,24 +1367,10 @@ export default function Page() {
                 )}
               </div>
             </button>
-            <button
-              onClick={() => { setShowPremium(!showPremium); setShowMarketplace(false); setShowCoverage(false); setShowProfile(false); setShowAdmin(false); }}
-              className={`group relative overflow-hidden rounded-xl transition-all shadow-sm hover:shadow-md ${
-                showPremium ? "ring-2 ring-[var(--primary)]" : ""
-              }`}
-              title={t("header.annual_premium")}
-            >
-              <div className="absolute inset-0 bg-gradient-to-br from-amber-500 via-orange-500 to-rose-500" />
-              <div className="relative flex items-stretch text-white">
-                <div className="flex items-center justify-center px-3 py-2 bg-black/15">
-                  <RupeeIcon />
-                </div>
-                <div className="px-3 py-2 text-left">
-                  <div className="text-[10px] uppercase tracking-wider opacity-85 leading-none">{t("header.annual_premium_kicker")}</div>
-                  <div className="text-xs font-bold leading-tight whitespace-nowrap">{t("header.annual_premium")}</div>
-                </div>
-              </div>
-            </button>
+            {/* KI (2026-05-15) — old "ESTIMATE Annual premium" CTA chip
+                removed. The premium-band chip below is now itself the
+                clickable surface to open the PremiumCalculatorPanel, so
+                two redundant premium UI elements collapsed into one. */}
             <button
               onClick={() => { setShowProfile(!showProfile); setShowMarketplace(false); setShowPremium(false); setShowCoverage(false); setShowAdmin(false); }}
               className={`group relative overflow-hidden rounded-xl transition-all shadow-sm hover:shadow-md ${
@@ -1417,9 +1403,13 @@ export default function Page() {
               profileCompleteness.completeness_pct >= 50 &&
               premiumBand &&
               premiumBand.sample_size > 0 && (
-              <div
-                className="group relative overflow-hidden rounded-xl shadow-sm"
-                title={`Estimate across ${premiumBand.sample_size} polic${premiumBand.sample_size === 1 ? "y" : "ies"}. Refresh as your profile fills in.`}
+              <button
+                type="button"
+                onClick={() => { setShowPremium(!showPremium); setShowMarketplace(false); setShowCoverage(false); setShowProfile(false); setShowAdmin(false); }}
+                className={`group relative overflow-hidden rounded-xl shadow-sm transition-all hover:shadow-md hover:brightness-110 cursor-pointer ${
+                  showPremium ? "ring-2 ring-[var(--primary)]" : ""
+                }`}
+                title={uiLang === "hi" ? "Premium को sliders से refine करने के लिए tap करें" : "Tap to refine premium with sliders"}
               >
                 <div className="absolute inset-0 bg-gradient-to-br from-amber-500 via-orange-500 to-amber-600" />
                 <div className="relative flex items-stretch text-white">
@@ -1434,8 +1424,16 @@ export default function Page() {
                       ₹{premiumBand.min_inr.toLocaleString("en-IN")}–₹{premiumBand.max_inr.toLocaleString("en-IN")}/yr
                     </div>
                   </div>
+                  {/* Subtle "edit" affordance — pencil-on-slider icon hints
+                      that tapping the chip opens the slider panel. */}
+                  <div className="flex items-center justify-center px-2 py-2 bg-white/15 border-l border-white/20 transition-transform group-hover:translate-x-0.5">
+                    <svg width="14" height="14" viewBox="0 0 24 24" fill="none" stroke="currentColor" strokeWidth="2.2" strokeLinecap="round" strokeLinejoin="round" aria-hidden="true">
+                      <path d="M4 21v-4l11-11 4 4-11 11H4z" />
+                      <path d="M14 6l4 4" />
+                    </svg>
+                  </div>
                 </div>
-              </div>
+              </button>
             )}
             {/* Admin access — opens the LLM control panel in an embedded view.
                 Backend admin API is password-gated (KI-097); enter the admin
@@ -1779,7 +1777,12 @@ export default function Page() {
                 isPersonalized={profileCompleteness?.is_personalized === true}
               />
             )}
-            {showPremium && <PremiumCalculatorPanel onClose={() => setShowPremium(false)} />}
+            {showPremium && (
+              <PremiumCalculatorPanel
+                onClose={() => setShowPremium(false)}
+                initialProfile={profileCompleteness?.profile}
+              />
+            )}
             {showProfile && (
               <ProfileBuilderPanel
                 sessionId={sessionId}
@@ -2089,13 +2092,72 @@ function ProfileBuilderPanel({
   );
 }
 
-function PremiumCalculatorPanel({ onClose }: { onClose: () => void }) {
-  const [age, setAge] = useState(35);
-  const [sumInsured, setSumInsured] = useState(1000000);
-  const [cityTier, setCityTier] = useState<"metro" | "tier1" | "tier2">("metro");
+function PremiumCalculatorPanel({
+  onClose,
+  initialProfile,
+}: {
+  onClose: () => void;
+  initialProfile?: UserProfile;
+}) {
+  // KI (2026-05-15) — Fix B. The panel previously opened with static
+  // defaults (Age 35 / SI 10L / Self only / None / metro) which felt
+  // disconnected from the user's already-captured profile. We now seed
+  // each slider from initialProfile (forwarded by page.tsx from
+  // profileCompleteness.profile) and fall back to the legacy default
+  // only when a slot is missing. User can still slide to override.
+  //
+  // The UserProfile schema (api.ts) does not yet carry a
+  // `desired_sum_insured_inr` slot, so we fall back to
+  // `existing_cover_inr` as the closest available signal; if that's
+  // also missing we land on the legacy 10L default.
+  const deriveFamilySize = (dep?: string | null): number => {
+    if (!dep) return 0;
+    const d = dep.toLowerCase();
+    if (d === "self" || d === "self only" || d === "self_only") return 0;
+    if (d.includes("parents") && d.includes("spouse")) return 4; // self+spouse+2 parents
+    if (d.includes("parents")) return 2; // self+parents
+    if (d.includes("kids") || d.includes("children")) return 3; // self+spouse+kids -> floater
+    if (d.includes("spouse")) return 1; // self+spouse
+    return 0;
+  };
+  const derivePed = (
+    conds?: string[] | null,
+  ): "none" | "diabetes_or_hypertension" | "heart_disease" | "multiple" => {
+    if (!conds || conds.length === 0) return "none";
+    const lower = conds.map((c) => (c || "").toLowerCase());
+    if (lower.every((c) => !c || c === "none")) return "none";
+    if (lower.length >= 2 && lower.some((c) => c !== "none")) {
+      const distinct = lower.filter((c) => c && c !== "none");
+      if (distinct.length >= 2) return "multiple";
+    }
+    if (lower.some((c) => c.includes("heart"))) return "heart_disease";
+    if (lower.some((c) => c.includes("diabetes") || c.includes("hypertension") || c.includes("bp")))
+      return "diabetes_or_hypertension";
+    return "diabetes_or_hypertension"; // any single non-none condition lands on the closest model bucket
+  };
+  const deriveCityTier = (loc?: string | null): "metro" | "tier1" | "tier2" => {
+    const l = (loc || "metro").toLowerCase();
+    if (l === "metro") return "metro";
+    if (l === "tier1" || l === "tier_1" || l === "tier-1") return "tier1";
+    if (l === "tier2" || l === "tier_2" || l === "tier-2") return "tier2";
+    // tier3 / unknown — fold down to tier2 (closest supported bucket)
+    return "tier2";
+  };
+
+  const [age, setAge] = useState<number>(initialProfile?.age ?? 35);
+  const [sumInsured, setSumInsured] = useState<number>(
+    initialProfile?.existing_cover_inr && initialProfile.existing_cover_inr > 0
+      ? initialProfile.existing_cover_inr
+      : 1000000,
+  );
+  const [cityTier, setCityTier] = useState<"metro" | "tier1" | "tier2">(
+    deriveCityTier(initialProfile?.location_tier),
+  );
   const [smoker, setSmoker] = useState(false);
-  const [familySize, setFamilySize] = useState(0);
-  const [ped, setPed] = useState<"none" | "diabetes_or_hypertension" | "heart_disease" | "multiple">("none");
+  const [familySize, setFamilySize] = useState<number>(deriveFamilySize(initialProfile?.dependents));
+  const [ped, setPed] = useState<"none" | "diabetes_or_hypertension" | "heart_disease" | "multiple">(
+    derivePed(initialProfile?.health_conditions),
+  );
   const [copay, setCopay] = useState(0);
   const [estimate, setEstimate] = useState<PremiumEstimateResponse | null>(null);
   const [busy, setBusy] = useState(false);