docs(major2): invariantraden - eksperten svarer paa forslaget paa bordet, og svaret brukes

CLAUDE.md: en invariantrad etter verdict-gate-raden, i husets skjelett - nevneren
(grep "prior_" finner EN soem, maskinens egen), D6 og hvorfor alternativet ble
forkastet, den ledger-bevisste attempts_remaining, honoured = "hentet faktisk",
kanalvalget begrunnet av tre maalinger, den kaller-eide sinken, skriv-iff-reviewer,
de fem nektene ved navn og den hostede pre-whitelist-sjekken. Load-bearing-blokka
lister M1-M40 med roedtall, M29 staar som et FUNN (uvitnet baerer), og de to
armene som var groenne av feil grunn i steg 1-8 er skrevet ned som repoets
vakuoes-gate-klasse, sekstende og syttende gang. Aerlighets-grensene til slutt.

README.md: --proposal-review i enkeltprosjekt-flagglista og en prosablokk etter
--plan-review/--checkpoint-dir-paret, i samme form - hva operatoeren ser og
skriver, at en revise KJOEPER ett forsoek til under de eksisterende takene, at
approve ikke er en ekspertdom, og alle fem nektene navngitt (--checkpoint-dir-en
peker paa --resume, doera som VIRKER). Kundevendt vokabular.

Ny arm: test_the_readme_block_names_every_flag_the_cli_refuses_the_door_with,
Fase-3-formen (raa tekst, uttrukket blokk, kontroll paa at uttrekkeren finner noe
som finnes). Intet M-nummer - lista lukket ved M40 - saa den ble drevet ROED TO
ganger: mot README-en foer blokka fantes, og med blokka paa plass men
--checkpoint-dir omskrevet til aa beskrive nekten uten aa navngi flagget.

1364 passed / 5 skipped. Golden demo-transcript.stdout BYTE-UENDRET
(shasum -a 1 av INNHOLDET = ea8c534773acdbe41ae68f2c55724d69aaf8be4f). Node-ID-ene
er et strengt supersett: 1319 -> 1369, 0 fjernet. mypy og ruff rene.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

README.md

@@ -416,7 +416,8 @@ when the seam is detached, so the loop cannot silently degrade into theater.
   cannot exercise every flag:
   - **Single-project** — `PROJECT_ID --docs-dir <dir>`, plus optional `--bundle-dir`,
     `--verdict-dir`, `--outbox-dir` (which requires `--run-id`), `--dimension-config`,
-    `--semantic-retrieval`, `--decision`/`--rationale`, `--live-dry-run`, and
+    `--semantic-retrieval`, `--decision`/`--rationale`, `--live-dry-run`, `--proposal-review`
+    (opt-in: answer the proposal the validator just accepted, at your terminal — see below), and
     `--scripted-replies <file>` (the offline whole-loop door — see
     [Walk the whole chain offline](#walk-the-whole-chain-offline); mutually exclusive with
     `--live-dry-run`, which stops before the first model call rather than answering it), and
@@ -511,6 +512,44 @@ when the seam is detached, so the loop cannot silently degrade into theater.
   `--checkpoint-dir` are refused together (two doors onto one review), as are `--resume` and
   `--explore` (two sources of one exploration).
 
+  **Answering the proposal review (`--proposal-review`).** The plan review asks you about the
+  *plan*, before any candidate exists. This one asks you about the **candidate on the table**:
+  every time the deterministic validator accepts a proposal, the run stops and shows you the
+  measure, the cost lines it touches, the claimed saving, the validator's percentiles, the
+  checker's verdict and how many further attempts are still affordable. You type `approve` to
+  take it as it stands, or `revise <what to change>`.
+
+  A `revise` is not a note filed for later. It **buys one more generation attempt** under the
+  caps the run already had — no new loop — and your words go into that attempt's prompt verbatim,
+  where they stay standing until you next answer. The run then carries whatever the validator
+  ruled on that later attempt: the expert asks for another try, the machine still decides. Every
+  round trip is written to `{run_id}-proposal-reviews.json` with your words, which candidate you
+  were shown, and whether the attempt your `revise` bought actually ran. Input that ends without
+  an answer stops the run rather than counting as approval.
+
+  ```bash
+  uv run python -m portfolio_optimiser.run BYGG-KONTOR-NORD --docs-dir <docs> \
+    --bundle-dir <bundle> --proposal-review --outbox-dir out --run-id r1
+  ```
+
+  `approve` is **not** an expert verdict, and does not become one: it says the candidate may
+  stand, not that the measure was reviewed and judged. The verdict still arrives the way it
+  always did — `--decision`/`--rationale` on the run, or a verdict file dropped into
+  `--verdict-dir` for a later run to read.
+
+  The door belongs to single-project mode, and every other combination is refused by name rather
+  than quietly ignored: `--portfolio` (its waves share one terminal, so several projects'
+  candidates would arrive at the same prompt with nothing to tell them apart), `--report` (a
+  read-only roll-up generates nothing to review), `--live-dry-run` (it stops before the first
+  model call, so no candidate ever reaches you), `--proposals-from-mandate` (that mode settles the
+  commission deterministically and never generates a candidate), and `--checkpoint-dir` (a parked
+  exploration returns before any candidate exists — pass `--proposal-review` at `--resume`
+  instead, which is where the two doors do compose).
+
+  Like the plan review it is **synchronous**, so the hosted surface refuses it by name and points
+  at this CLI. With `--explore --plan-review --proposal-review` you answer two different doors
+  from the same terminal in sequence: first the plan, then each candidate the validator accepts.
+
 - **Deriving the cost baseline from the knowledge base** — `--derive-cost-baseline` (opt-in,
   requires `--bundle-dir`). The deterministic validator anchors a proposal's `affected_items` to
   the project's actual cost lines. Normally those come from a hand-written `cost-baseline.json` in