fix(samples): pad all 4 sample initiatives above the 200-word minimum

Reported live: clicking a sample then Diagnose hit "need at least 200
words" (one was 185, one 171, one 188, one 175). The samples were below
the validator threshold the LLM needs for meaningful scoring.

Padded each with substantive operational detail (not filler) — extra
context that makes the case clearer rather than just adding words:

compounder 185 → 231 (added: premium size, decline + referral
rates, more claim types, "agents favor
faster turnaround" punchline)
one-shot win 171 → 228 (added: collateral fields, annual loan
count, vendor's deterministic-task framing,
chief credit officer's flag about LGD)
wrong thing 188 → 230 (added: 20K orders/day network volume,
specific 6-10pm dock window, labor cost
up 8% because of idle hours)
roman candle 175 → 217 (added: $480M revenue, "shared inference
fleet" with competitor brands, list of
actual unexamined constraints — drive-thru,
menu, breakfast — and the Q4 board deadline)

Synchronized between both locations:
gradio-apps/compounding-test/app.py — the Space's gr.Examples
src/components/CompoundingTestAI.tsx — the site's React component

(The duplication is intentional for now; consolidating to a shared JSON
is a fair cleanup if these samples need to change often enough that
sync drift becomes painful.)

Verified word counts via str.split() / split(/\s+/) which match between
Python and JS. Astro build + 31 pytest tests pass.

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

app.py

@@ -401,83 +401,100 @@ BUDGETS = ["<$100K", "$100K–$1M", "$1M–$10M", ">$10M"]
 _SAMPLE_COMPOUNDER = (
     "We're a regional commercial insurance carrier specializing in restaurant "
     "general liability. We write about 8,000 policies a year across the "
-    "Midwest. Underwriting is the bottleneck of our business — agents wait "
-    "36 to 48 hours for a quote because our underwriters manually pull "
-    "industry codes, loss runs, and prior-carrier history from three "
-    "different systems and decide whether to bind, decline, or refer. "
-    "We're deploying an LLM-powered underwriting assistant that pulls the "
-    "data automatically, flags risk factors based on patterns in our "
-    "12-year claims database, and proposes a base rate. The underwriter "
-    "reviews and approves. Every policy we write generates new claim "
-    "outcomes — fires, slip-and-falls, liquor-liability claims — and those "
-    "outcomes feed back into the next quarter's model retraining. Our "
-    "competitors mostly use Verisk's industry-standard models, which we "
-    "don't share data with, so our model gets better on our book of "
-    "business while theirs reflects the industry average. Internal goal: "
-    "cut time-to-quote from 36 hours to 4 hours, increase win rate on "
-    "profitable risks by 15%, and progressively shift the loss ratio by "
-    "1–2 points per year as the model learns from each renewal cycle."
+    "Midwest, with average annual premium around $4,500. Underwriting is "
+    "the bottleneck of our business — independent agents wait 36 to 48 "
+    "hours for a quote because our underwriters manually pull industry "
+    "codes, loss runs, and prior-carrier history from three different "
+    "systems and then decide whether to bind, decline, or refer. Roughly "
+    "30% of submissions get declined and another 15% are referred to "
+    "senior underwriters, which adds another day. We're deploying an "
+    "LLM-powered underwriting assistant that pulls the data automatically, "
+    "flags risk factors based on patterns in our 12-year claims database, "
+    "and proposes a base rate with an explanation. The underwriter "
+    "reviews, adjusts, and approves. Every policy we write generates new "
+    "claim outcomes — fires, slip-and-falls, liquor-liability claims, "
+    "food-poisoning suits — and those outcomes feed back into the next "
+    "quarter's model retraining. Our competitors mostly use Verisk's "
+    "industry-standard rating models, which we don't share data with, so "
+    "our model gets better on our specific book of business while theirs "
+    "reflects the industry average. Internal goal: cut time-to-quote from "
+    "36 hours to 4 hours, increase the win rate on profitable risks by "
+    "15%, and progressively shift the loss ratio by 1–2 points per year "
+    "as the model learns from each renewal cycle. Independent agents have "
+    "already started favoring carriers with faster quote turnaround."
 )
 
 _SAMPLE_ONE_SHOT_WIN = (
-    "We're a community bank with $4B in assets, 38 branches across two "
+    "We're a community bank with $4B in assets and 38 branches across two "
     "states. Loan officers spend about 6 hours per commercial loan "
     "reviewing financial statements, tax returns, and corporate documents "
     "before they can write the credit memo. We're deploying GPT-4 to "
     "extract key fields — revenue, EBITDA, debt service coverage ratio, "
-    "ownership structure, related-party transactions — from these "
-    "documents into a structured form. The loan officer reviews the "
-    "extraction and writes the credit memo by hand. We expect to cut "
-    "document review time from 6 hours to about 90 minutes per loan. "
-    "The vendor provides the model and the document templates and is "
-    "selling the same system to four of our peer community banks in the "
-    "region. The AI doesn't learn from the outcome of the loan: defaults, "
-    "prepayments, modifications, restructurings all go into our loan "
-    "servicing system, which doesn't connect back to the extraction "
-    "model. The vendor's roadmap doesn't include any feedback loop. "
-    "We're funding the project from the operations budget; the credit "
-    "team is excited about the time savings."
+    "ownership structure, related-party transactions, collateral "
+    "descriptions — from these documents into a structured form. The loan "
+    "officer reviews the extraction and writes the credit memo by hand. "
+    "We expect to cut document review time from 6 hours to about 90 "
+    "minutes per loan, processing roughly 2,400 commercial loans a year. "
+    "The vendor provides the model, the document templates, and the "
+    "extraction prompts, and is selling the same system to four of our "
+    "peer community banks in the region under identical contracts. The AI "
+    "doesn't learn from the outcome of the loan: defaults, prepayments, "
+    "modifications, restructurings all go into our separate loan "
+    "servicing system, which has never connected back to the extraction "
+    "model. The vendor's three-year roadmap doesn't include any feedback "
+    "loop between loan performance and the model — they treat extraction "
+    "as a deterministic task. We're funding the project from the "
+    "operations budget; the credit team is excited about the time savings "
+    "but the chief credit officer has flagged that the productivity gain "
+    "will be one-time and won't show up in the loss-given-default rate "
+    "over time."
 )
 
 _SAMPLE_WRONG_THING = (
-    "We're a third-party logistics provider with 8 warehouses on the "
-    "East Coast. We're investing in computer vision software to optimize "
-    "order picking routes — the AI looks at the warehouse layout, "
-    "current orders, and worker positions and suggests optimized pick "
-    "paths in real time. Pilot results show a 12% reduction in steps "
-    "per order on the test floor. Our operations team has been excited "
-    "about this for 18 months and we just signed a multi-year contract "
-    "with the vendor. Some context on the operation: our warehouses run "
-    "2 shifts. Order volume in shift 1 is around 14,000 orders/day; "
-    "shift 2 is around 6,000. The pick wave finishes by 2pm on shift 1, "
-    "then workers wait 4 to 5 hours for shift 2 trucks to arrive at the "
-    "loading docks. The trucks are scheduled by the customer (a major "
-    "national retailer) and arrive in unpredictable windows. We don't "
-    "control the truck schedule. The CFO has been asking us why total "
-    "throughput per warehouse hasn't moved much in three years; our "
-    "answer has been that the legacy WMS is the constraint, which is "
-    "why we're investing in better picking AI."
+    "We're a third-party logistics provider with 8 warehouses on the East "
+    "Coast handling about 20,000 orders a day across the network. We're "
+    "investing in computer vision software to optimize order picking "
+    "routes — the AI looks at the warehouse layout, current orders, and "
+    "worker positions and suggests optimized pick paths in real time. "
+    "Pilot results show a 12% reduction in steps per order on the test "
+    "floor. Our operations team has been excited about this for 18 months "
+    "and we just signed a multi-year contract with the vendor. Some "
+    "context on the operation: our warehouses run 2 shifts. Order volume "
+    "in shift 1 is around 14,000 orders per day; shift 2 is around 6,000. "
+    "The pick wave finishes by 2pm on shift 1, then workers wait 4 to 5 "
+    "hours for shift 2 trucks to arrive at the loading docks. The trucks "
+    "are scheduled by the customer (a major national retailer) and arrive "
+    "in unpredictable windows between 6pm and 10pm. We don't control the "
+    "truck schedule and the customer won't share their advance schedule "
+    "with us. The CFO has been asking us why total throughput per "
+    "warehouse hasn't moved much in three years; our answer has been that "
+    "the legacy warehouse management system is the constraint, which is "
+    "why we're investing in better picking AI. Same-store labor cost is "
+    "up 8% year-over-year because workers are paid through the idle hours."
 )
 
 _SAMPLE_ROMAN_CANDLE = (
-    "We run a chain of 220 quick-service restaurants across the "
-    "Southeast. Our gross margin has been under pressure from rising "
-    "ingredient costs and we're rolling out an AI-powered personalized "
-    "marketing platform that sends customized email and SMS offers "
-    "based on customer purchase history, location, and weather. The "
-    "platform is from a major QSR-tech vendor used by several of our "
-    "competitors. Our customer data — names, emails, phone numbers, "
-    "purchase frequency — lives in our point-of-sale provider's cloud, "
-    "which the marketing platform pulls from via the POS provider's "
-    "standard integration. Our purchase data and the modeling are both "
-    "the vendor's stack; we don't see the underlying model and our data "
-    "is commingled with other QSR brands the vendor serves. We expect "
-    "to lift email click-through by 8–12% based on the vendor's "
-    "benchmark studies of similar brands. The marketing team is "
-    "running the rollout; finance signed off on the multi-year "
-    "subscription. We have not measured what's actually constraining "
-    "same-store sales growth — we just know revenue has been flat for "
-    "two years and the board wants action."
+    "We run a chain of 220 quick-service restaurants across the Southeast "
+    "doing about $480M in annual revenue. Our gross margin has been under "
+    "pressure from rising ingredient costs and we're rolling out an "
+    "AI-powered personalized marketing platform that sends customized "
+    "email and SMS offers based on customer purchase history, location, "
+    "and local weather. The platform is from a major QSR-tech vendor used "
+    "by several of our direct competitors in the same markets we operate "
+    "in. Our customer data — names, emails, phone numbers, purchase "
+    "frequency, average ticket size — lives in our point-of-sale "
+    "provider's cloud, which the marketing platform pulls from via the "
+    "POS provider's standard integration. Both the purchase data feed and "
+    "the modeling are the vendor's stack; we don't see the underlying "
+    "model and our data is commingled with other QSR brands the vendor "
+    "serves on a shared inference fleet. We expect to lift email "
+    "click-through by 8–12% based on the vendor's benchmark studies of "
+    "similar brands. The marketing team is running the rollout; finance "
+    "signed off on the multi-year subscription. We have not measured what "
+    "is actually constraining same-store sales growth — drive-thru wait "
+    "times, menu pricing relative to local competitors, or breakfast "
+    "daypart penetration — we just know revenue has been flat for two "
+    "years and the board wants visible action by Q4."
 )