Commit graph

2 commits

Author SHA1 Message Date
c84e8bf6f1 feat(p19): a direction must NAME the requirement that binds it, and have READ it
Two paid rounds scored 0 of 26 fasit concepts opened -- the same number twice.
P18 closed the navigation side (a listing is a window, an invented path is
refused by name) and it did not move, which makes it a ROLE question: nothing
in the loop ever asked the model to say what requirement binds the direction it
committed to, so opening one was never on the critical path to an answer.

A PREMISE OF THE ORDER WAS FELLED BEFORE ANYTHING WAS BUILT ON IT. A1 places
the demand in _INSTRUCTIONS[HYPOTHESISER_ROLE] alone. Measured: the stress
command sends --mandate and NOT --explore, the two are refused together by
name, and none of the nine round-1/2 outboxes holds a {run_id}-exploration.json
-- the hypothesiser never runs in a stress round, so A3 would have been
unreachable in exactly the paid runs this order commissions.

A2's own sentence resolves it: the refusal goes to the model "som en tur den
kan rette (samme mekanisme som quick_validate's nekt), ikke som en raise" --
and quick_validate IS a tool. declare_requirement therefore lives in
navigator_tools, held by BOTH roles that navigate (the exploration, and since
S2c the debate). It EXISTS only when the caller offers both sinks, which keeps
every pre-P19 call site byte-identical; one sink without the other is refused
at construction. 'opened' is the SAME list ExplorationToolRecorder fills, so
the refusal reads the run's own read trace.

The marked hypothesis carries 'requirement' as a REQUIRED key: omitted is a
hard error, explicit null is legal and needs 'why_none', a half-named one is
refused. A minted approach carries it; a seed never acquires one. The proposer
prompt names it only when the field exists, and the judge counts a hit against
THIS approach's fasit concepts, never against the base.

Load-bearing measured (12 arms), four mutations all red against the whole
suite, green control 1711/5 and demo-transcript.stdout byte-unchanged.
A-iii's predicted signature was FALSIFIED: the golden stays green because the
demo runs without a mandate, so _build_messages' approach branch is never
taken there. A-iv was GREEN first -- the repo's vacuous-gate class, 24th time:
the arm drove _attributable while the hit is computed at the call site.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
2026-09-15 01:21:47 +02:00
7552c0ef73 fix(explore): a plan review never shows an expert the word None and calls it progress
[skip-docs]

PM addendum 4 to order 20260902T151931Z-250257273.

PlanReviewRequest.current_progress and ParkedExploration.current_progress were
both built with str(review.current_progress). The value is a
MagenticProgressLedger | None and is None in every run measured so far, so
str() produced the four characters None, the renderer's truthiness guard found
them non-empty, and the terminal printed a "progress so far" section whose only
content was None. The same four characters went into
{run_id}-plan-review.json -- the one thing that crosses the process boundary in
the asynchronous door, where nobody can ask what it meant.

_progress_text is the ONE conversion, beside _plan_text: absent becomes the
empty string, never "None". The data layer states absence by being absent, as
Bundle.skipped's empty tuple does. The TERMINAL states it in words -- this is
the one surface where omission is wrong, because silence at a gate somebody
signs is exactly what PlanReviewInputError already refuses for EOF.

Measured, and it changed the test: reverting both sites left ONLY a
source-inspection arm red, which is a lint and not a gate, because the first
arms construct a PlanReviewRequest themselves and never enter either site. Each
site now has a behavioural witness that drives the real door -- the terminal for
the synchronous one, the parked question FILE for the asynchronous one.

Load-bearing measured, five mutations, all red against the WHOLE suite, each
with its own signature: both sites reverted (3 red) - synchronous site alone
(2) - parked site alone (2) - the terminal goes silent again (2) - the
placeholder always fires, swallowing a real ledger (1, the control alone).
Green control 1202 passed / 5 skipped; golden demo-transcript.stdout unchanged
(ea8c534773acdbe41ae68f2c55724d69aaf8be4f).

The recorder wiring in resume_exploration is untouched, per the order.
2026-09-03 01:06:27 +02:00