test(portfolio): cover the CLI→run_portfolio wire for --verdict-dir (§11 gap)

The gap, found by the mutation sweep of 2026-07-25: verdict_dir=args.verdict_dir
→ None in run.py's execute_portfolio call left the suite 603/603 GREEN. The
flag was wired but not guarded — the inner merge (test_portfolio_learning_
loadbearing.py), the argparse refusal (--verdict-dir without --portfolio) and
the README↔--help sync all stay green under that mutation, so none of them
covered the forwarding itself.

One load-bearing test, no production code. It authors an expert verdict into a
tmp portfolio inbox — keyed on the bundle's own codes + measure type so it ranks
into the fold, with a distinct saving so its id cannot collide with the bundle's
seed — drives main(["--portfolio", …, "--verdict-dir", X]) with the scripted
client, and asserts the verdict's id AND a marker token (present nowhere in the
bundle) reach the proposer prompt.

Detach proof (mutation restored from a COPY, never git checkout): the wire
mutated to None → 1 failed, 603 passed, and the failure is this test alone.
Restored → 604 passed, ruff format left 71 files unchanged, ruff check + mypy
--strict clean.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MQu2xxwedckjU56byu1aUG

tests/test_cli_paritet_loadbearing.py

@@ -8,7 +8,10 @@ makes them REACHABLE. A hard goal already met by the ledger stops the run
 BEFORE any model call — the goal bounds achievement where the §8 budget bounds
 spend, and a stop is structured output, never a silent one. Detach the goal
 check from the entrance → the run proceeds and spends → red. Detach the
-portfolio branch → the config's projects never run → red.
+portfolio branch → the config's projects never run → red. Stop forwarding
+``--verdict-dir`` → the expert inbox the operator named reaches no fold → red:
+the flag's refusal path and the README↔``--help`` sync leave that wire uncovered,
+so it is asserted here on the fold itself.
 
 **The README claims exactly what the CLI delivers (§1).** Every ``--flag`` the
 README documents must exist in the help of a project CLI the README names.
@@ -32,6 +35,8 @@ import pytest
 from _scripted import ScriptedClient, reply
 
 from portfolio_optimiser_claude.contracts import Contracts, FeedbackContract
+from portfolio_optimiser_claude.experience import CandidateFeatures
+from portfolio_optimiser_claude.inbox import VerdictDocument, write_verdict
 from portfolio_optimiser_claude.ir import load_validator_input
 from portfolio_optimiser_claude.ledger import SavingsLedger
 from portfolio_optimiser_claude.loop import ModelClient, ModelReply
@@ -97,6 +102,35 @@ def _ledger_file(tmp_path: Path, *, amount_nok: float) -> Path:
     return path
 
 
+# A token that lives NOWHERE in the bundle — its presence in a prompt can only
+# have come through the inbox the operator named with --verdict-dir, never from
+# the bundle context or from seeding.
+_VERDICT_DIR_MARKER = "K12-CLI-VERDICT-DIR-MARKER"
+
+
+def _portfolio_inbox(tmp_path: Path) -> tuple[Path, VerdictDocument]:
+    """An EXPERT verdict in a portfolio inbox, keyed near the bundle's candidate.
+
+    Same codes + measure type as the bundle's proposal (so it ranks into the
+    fold) with a distinct saving (so its id cannot collide with the bundle's own
+    seed). The marker rides in the rationale — the fold's learning signal.
+    """
+    candidate = CandidateFeatures.from_proposal(load_validator_input(BUNDLE))
+    verdict = VerdictDocument.from_candidate(
+        CandidateFeatures(
+            affected_codes=candidate.affected_codes,
+            measure_type=candidate.measure_type,
+            claimed_saving_nok=candidate.claimed_saving_nok + 3000.0,
+        ),
+        decision="approved",
+        rationale=f"prior portfolio verdict — realiseringskorreksjon [{_VERDICT_DIR_MARKER}]",
+        description="K12 CLI portfolio-inbox fixture (surface text, excluded from ranking)",
+    )
+    verdict_dir = tmp_path / "portfolio-inbox"
+    write_verdict(verdict_dir, verdict)
+    return verdict_dir, verdict
+
+
 def _portfolio_file(tmp_path: Path, project_ids: list[str]) -> Path:
     path = tmp_path / "portfolio.json"
     path.write_text(
@@ -263,6 +297,32 @@ class TestPortfolioOnTheEntrance:
             main([], client_factory=factory)
         assert created == []
 
+    def test_verdict_dir_reaches_every_projects_fold(self, tmp_path: Path) -> None:
+        # Detach point: stop forwarding args.verdict_dir out of main() (pass None
+        # to execute_portfolio) → the inbox the operator named never reaches a
+        # fold → RED. The merge itself is proven inside run_portfolio
+        # (test_portfolio_learning_loadbearing.py); what this keeps alive is the
+        # CLI→run_portfolio WIRE — the refusal test below and the README↔--help
+        # sync both stay green under that mutation, so neither covers it.
+        verdict_dir, inbox_verdict = _portfolio_inbox(tmp_path)
+        factory, created = _scripted_factory(runs=1)
+        code = main(
+            [
+                "--portfolio",
+                str(_portfolio_file(tmp_path, ["prosjekt-a"])),
+                "--verdict-dir",
+                str(verdict_dir),
+            ],
+            client_factory=factory,
+        )
+        assert code == 0
+        assert created
+        prompts = created[0].prompts("proposer")
+        assert prompts, "the project must have driven the loop at all"
+        assert any(_VERDICT_DIR_MARKER in p and inbox_verdict.id in p for p in prompts), (
+            "the verdict from --verdict-dir (id + marker) must reach the project's fold"
+        )
+
     def test_verdict_dir_without_portfolio_is_refused(self, tmp_path: Path) -> None:
         # --verdict-dir is the PORTFOLIO-level expert inbox; on a single run the
         # per-run inbox is --inbox. Accepting it silently would claim a wiring