From d16958bcfd14c94dce28ba4cbb69045a363d138a Mon Sep 17 00:00:00 2001 From: Jason Woltje Date: Sat, 24 Oct 2026 00:26:11 -0500 Subject: [PATCH] docs: W brief D2 lets O2 rows build before 10-27 and land only on an idle slot, Opus 5.5 review once at landing (mos-1571); register Codex seats Glomgold (row 309, W1) and Gyro (row 310, W2); doc only --- agents/glomgold/README.md | 9 +++++++++ agents/gyro/README.md | 10 ++++++++++ docs/plans/2026-10-21_q195-writing-editor.md | 10 ++++++---- 3 files changed, 25 insertions(+), 4 deletions(-) create mode 100644 agents/glomgold/README.md create mode 100644 agents/gyro/README.md diff --git a/agents/glomgold/README.md b/agents/glomgold/README.md new file mode 100644 index 000000000..d5b766b8a --- /dev/null +++ b/agents/glomgold/README.md @@ -0,0 +1,9 @@ +# Glomgold + +Builder seat, Codex gpt-6.1-sol, started by Sage under Q246 and mos-1571. Owns row 309, O2 W1: the reviewed local +thesaurus data (docs/plans/2026-10-21_q195-writing-editor.md, "W1 reviewed local thesaurus data"). + +No launcher in this directory. The seat runs as a T3 thread and works in its own worktree. It has no credentials of its +own; it pushes only its own `glomgold/*` branches through the repository credential helper (mos-0981), never `refactor`, +`main` or another seat's branch. Claude seats orchestrate and give the Opus 5.5 landing review (mos-1571). Registered by +Sage on 2026-10-24 so the queue can name the seat as row owner (store.mjs isSeatDir). diff --git a/agents/gyro/README.md b/agents/gyro/README.md new file mode 100644 index 000000000..6a510e27a --- /dev/null +++ b/agents/gyro/README.md @@ -0,0 +1,10 @@ +# Gyro + +Builder seat, Codex gpt-6.1-sol, started by Sage under Q246 and mos-1571, as the W2 brief names. Owns row 310, O2 W2: +the WordCraft source import and reproducible restricted web build (docs/plans/2026-10-21_q195-writing-editor.md, +"W2 WordCraft source import and reproducible restricted web build"). + +No launcher in this directory. The seat runs as a T3 thread and works in its own worktree. It has no credentials of its +own; it pushes only its own `gyro/*` branches through the repository credential helper (mos-0981), never `refactor`, +`main` or another seat's branch. Claude seats orchestrate and give the Opus 5.5 landing review (mos-1571). Registered by +Sage on 2026-10-24 so the queue can name the seat as row owner (store.mjs isSeatDir). diff --git a/docs/plans/2026-10-21_q195-writing-editor.md b/docs/plans/2026-10-21_q195-writing-editor.md index 550557503..8d39017c2 100644 --- a/docs/plans/2026-10-21_q195-writing-editor.md +++ b/docs/plans/2026-10-21_q195-writing-editor.md @@ -27,8 +27,9 @@ pane, persistent chat in the right, a draggable divider, document navigation, pi right-click thesaurus. It is not a reskinned row 154 editor. The first useful release, opened at W7 after the required gates, must pass writing, footnote, comment, tracked-change and page-layout save/reopen tests together. -D2. Q210/Q220 allow design and review now. O2 builds remain after the October 27 MVP work -and never take its gate slot. S0 baseline clearance is conditional on M1–M14, Fenton/Filbert +D2. Q210/Q220 allow design, review and builds ahead of the October 27 MVP work (mos-1571). +O2 rows never take the gate slot ahead of a v1 row; they land only when the slot would +otherwise sit idle. S0 baseline clearance is conditional on M1–M14, Fenton/Filbert at `origin/fenton/q195-s0` 74ec32013 (#1798, Sage PASS comment 30917). Q234 approves the exact measured frame CSP; the shell stays unchanged. mos-1363 requires W2 to remeasure zero violations with style-src 'self'; no rfd style hash is approved. These gates are satisfied @@ -585,14 +586,15 @@ Their implementation gates remain separate. No code row W1/W2/W6/W3a/W3b/W3/W4b/ before S0 PASS (recorded for this baseline, with M1–M14 still binding). Each also observes its explicit dependencies, source ownership, the CSP gate and the MVP gate-slot priority. Q177 permits chaining reviewed candidates, not overlapping owners. Sage assigns model-only seats. Each implementation reviewer is Opus 5.5 and did not -author the code. Fenton reviews this design and can review code only when not its author. +author the code. A Codex-built row gets its first reviews from a separate Codex thread (Q246); +the Opus 5.5 review runs once, on the landing candidate (mos-1571). Fenton reviews this design and can review code only when not its author. Build order (mos-1470, Vikunja 571 comment 961): W7 first supplies reviewed conversion/format routing code with production import still closed, then W8 lands the Connections view, then W7's human opening step may open the first real chapter import. W8 is after W3/W4b and W7's closed routing candidate, before that W7 opening step. This is Mos's build-order rule, not a new REQ-LIB-8 condition or an addition to import acceptance. No converted real chapter -exists without the view. O2 code remains after 10-27 and never ahead of MVP rows. +exists without the view. O2 code may be built before 10-27 (mos-1571) but never lands ahead of MVP rows. G1. Future code acceptance runs each touched test file independently and affected package suites on Sage's slot after MVP priority, plus required repository checks. Source builds need