Heatmap variant with the T2 bounty row outlined

examples/ddd-architectural-challenges/verdict_heatmap.py

@@ -207,7 +207,7 @@ def collect() -> tuple[dict[tuple[str, str], list[dict]], list[str]]:
          for g, ramp in GROUP_RAMP.items()}
 
 
-def heatmap(per_arm: dict, lang: str) -> None:
+def heatmap(per_arm: dict, lang: str, highlight: str | None = None) -> None:
     t = TEXT[lang]
     grid = [[sum(VAL[evals[c]] for evals in per_arm[arm]) / len(per_arm[arm])
              for arm in ARM_ORDER] for c in CHECKS]
@@ -248,9 +248,20 @@ def heatmap(per_arm: dict, lang: str) -> None:
                     xycoords="axes fraction", ha="left", va="center", fontsize=8.6,
                     color=GROUP_RAMP[group][1], rotation=270)
 
+    if highlight:
+        yi = CHECKS.index(highlight)
+        ax.add_patch(plt.Rectangle((-0.5, yi - 0.5), len(ARM_ORDER), 1.0,
+                                   fill=False, edgecolor=INK, linewidth=2.4,
+                                   zorder=6, clip_on=False))
+        ax.get_yticklabels()[yi].set_fontweight("bold")
+        ax.get_yticklabels()[yi].set_color(INK)
+
     fig.text(0.5, 0.015, t["scale"], ha="center", fontsize=8.2, color="#777")
     fig.tight_layout(rect=(0, 0.035, 0.97, 0.905))
-    out = HERE / "assets" / t["out"]
+    name = t["out"]
+    if highlight:
+        name = name.replace(".png", f"_{highlight.lower()}.png")
+    out = HERE / "assets" / name
     fig.savefig(out, dpi=160, facecolor="white")
     print("saved:", out)
 
@@ -263,3 +274,6 @@ def heatmap(per_arm: dict, lang: str) -> None:
         print("FLAG", fl)
     for lang in ("pl", "en"):
         heatmap(per_arm, lang)
+        # article/social variant: the bounty row outlined (T2 - the check
+        # no model met for two generations until Opus 5)
+        heatmap(per_arm, lang, highlight="T2")

src/nasde_toolkit/pricing.toml

@@ -15,8 +15,10 @@
 # per-step cost accounting on the 2026-07 grid. OpenAI bills no write premium,
 # so gpt entries omit the field (writes fall back to the base input rate).
 #
-# Verified 2026-07-13 (Claude models re-checked against the official pricing page
-# before publishing) — CONFIRM before publishing any cost figures.
+# Claude models re-verified 2026-08-24 against the official pricing page (Fable 5,
+# Opus 5, Opus 4.8, Sonnet 4.6 unchanged since 2026-07-13; Sonnet 5 added) —
+# CONFIRM before publishing any cost figures. gpt entries not re-checked since
+# 2026-06-08.
 
 [models."gpt-5.5"]
 input_per_1m = 5.0
@@ -37,29 +39,37 @@ input_per_1m = 5.0
 output_per_1m = 25.0
 cached_input_per_1m = 0.50
 cache_write_per_1m = 10.0
-as_of = "2026-07-24"
+as_of = "2026-08-24"
 source = "https://platform.claude.com/docs/en/about-claude/pricing"
 
 [models."claude-opus-4-8"]
 input_per_1m = 5.0
 output_per_1m = 25.0
 cached_input_per_1m = 0.50
 cache_write_per_1m = 10.0
-as_of = "2026-07-24"
+as_of = "2026-08-24"
 source = "https://platform.claude.com/docs/en/about-claude/pricing"
 
 [models."claude-fable-5"]
 input_per_1m = 10.0
 output_per_1m = 50.0
 cached_input_per_1m = 1.0
 cache_write_per_1m = 20.0
-as_of = "2026-07-24"
+as_of = "2026-08-24"
+source = "https://platform.claude.com/docs/en/about-claude/pricing"
+
+[models."claude-sonnet-5"]
+input_per_1m = 2.0
+output_per_1m = 10.0
+cached_input_per_1m = 0.20
+cache_write_per_1m = 4.0
+as_of = "2026-08-24"
 source = "https://platform.claude.com/docs/en/about-claude/pricing"
 
 [models."claude-sonnet-4-6"]
 input_per_1m = 3.0
 output_per_1m = 15.0
 cached_input_per_1m = 0.30
 cache_write_per_1m = 6.0
-as_of = "2026-06-08"
+as_of = "2026-08-24"
 source = "https://platform.claude.com/docs/en/about-claude/pricing"

tests/test_evaluator.py

@@ -282,7 +282,7 @@ def test_assessment_summary_includes_economics(tmp_path: Path) -> None:
     assert summary.token_usage["total_tokens"] == 1_060_000
     # sonnet $3/$15/$0.30 cached (ADR-014): fresh 0.2M*3 + cached 0.8M*0.30 + 0.06M*15 = 1.74
     assert summary.cost_usd == pytest.approx(1.74)
-    assert summary.pricing_as_of == "2026-06-08"
+    assert summary.pricing_as_of == "2026-08-24"
     assert not hasattr(summary, "cost_efficiency")  # removed: arbitrary zero → use Pareto front
     assert not hasattr(summary, "token_efficiency")
 

tests/test_pricing.py

@@ -171,6 +171,7 @@ def test_layered_three_layers_compose(tmp_path: Path, empty_user_layer: Path) ->
         "claude-opus-5",
         "claude-fable-5",
         "claude-opus-4-8",
+        "claude-sonnet-5",
         "claude-sonnet-4-6",
         "azure-gpt5",
         "enterprise-claude",

tests/test_results_exporter.py

@@ -174,7 +174,7 @@ def test_export_includes_token_cost_economics(job_dir: Path, tmp_path: Path) ->
     assert usage["total_tokens"] == 1_060_000
     # sonnet $3/$15/$0.30 cached (ADR-014): fresh 0.2M*3 + cached 0.8M*0.30 + 0.06M*15 = 1.74
     assert metrics["cost_usd"] == pytest.approx(1.74)
-    assert metrics["pricing_as_of"] == "2026-06-08"
+    assert metrics["pricing_as_of"] == "2026-08-24"
     assert "cost_efficiency" not in metrics  # removed: arbitrary zero → use Pareto front
     assert "token_efficiency" not in metrics
     assert metrics["reasoning_effort"] == ""  # fixture set no override