docs(trading): minimal D7 with expected R1 rejection - #808

Scope

Fix J808 P2-1, P2-2 and P2-3 in the minimal D7 proposal. The record permits only the original table's 15 date-field bindings; all 57 D6 entries retain unknown scope_kind, identifiers_or_universe and access_kind unless an existing frozen source-cited rule applies. No new classifier, identifier mapping, supplementary/code source or projection is introduced.

• Lane: lane:trading; custody follows trading: R1 deviation D6 binds the critic's overnight-volume read and its reviewed class #796 with 5f.
• Rebased onto landed main 67c6b6f.
• Source/preregistration basis remains ecfa112 / ACKed b253161, with all seven kept files and both protected catalyst files byte-identical.
• Owned paths: deviations-20261006-5.json and q5-clarification-20261006.md in blueprints/us-equities/rnd-wave1-20261006; manifests/evidence.json is updated in the last commit.
• Draft only. Request the co-op's delta GPT read and the CC ACK of this exact head before the R1 rerun uses it.

SOTA sources

• Frozen D1/D4/D5/D6 at b253161a.
• Retained source records at ecfa1127: original exact file/span hashes and date-only patterns/JSON paths.
• CC RULING-Q5-MINIMAL, task-ns2604-coop-20261006T204552Z section 2; J808 full GPT verdict at 7ebf616.
• Repository hot-file protocol in docs/lanes.md, host_receipts.register_file and native validate.py publication scanner.

Evidence-class table

Local commands run

nice -n 19 python3 <private minimal date/source audit>
native=0 sanitizer=0; 15 authorized bindings; 14 full pairs match D6;
C06 is [unknown, 2026-09-21] and intentionally differs from the complete D6 pin

nice -n 19 git rebase --onto origin/main b253161a480f8369283e4661e36e737792765891
native=0 sanitizer=0

nice -n 19 git checkout origin/main -- manifests/evidence.json
nice -n 19 python3 <host_receipts.register_file invocation for the two D7 paths>
nice -n 19 python3 <private registry parity check>
native=0 sanitizer=0; preserve all main rows; D6=1, D7=2

nice -n 19 python3 scripts/validate.py --scan-file <D7 JSON> --scan-file <D7 note> --scan-file manifests/evidence.json
native=0 sanitizer=0; scanned_files=3

nice -n 19 git diff --check
native=0 sanitizer=0

These are source/document checks, not R1 A1-A8/H1 acceptance. Full repository validation and the R1 rerun remain held until the allowed slot and CC ACK.

Decision record

The dated D7 and its companion note now state the concrete expected rerun: with prerequisites met and no lower-numbered rejection, REJECTED_DERIVATION (11) follows known D6 pins that differ from required unknown derivations. Unknown itself adds no rejection rule. Unknown entries count as present for completeness and conservatively refuse their potential scope; unbounded unknown covers every case. Eligibility 0 is reported context and must be recomputed, never a typed gate. Prerequisites, native/sanitizer exits, ERROR handling and D4's lowest-numbered rejection precedence are unchanged.

C06's stale-bar lower endpoint stays unknown for the scoped adjudication requested in the GPT verdict. The original literal-pair table was not proof of a full data-read interval. No oracle value fills the absence.

Host evidence

No host, credential, acquisition or strategy changes. Attempt 1, all superseded drafts and failed attempts are retained privately and on the pre-rebase history branch.

Checklist

• P2-1 minimal source rules; P2-2 expected unknown outcome; P2-3 rebase and main registry parity.
• Frozen bytes and failed/refused evidence preserved.
• Delta GPT read and scoped C06 adjudication on this exact head.
• CC ACK of this dated record.
• Full validation / later R1 execution in an allowed slot.