jason.woltje
3510c3cbf8
queue: row 174 for #1672 , skill read BOM and fs errors after row 173, not started (mos-0863 D4; Sage rev 2026)
2026-10-19 16:20:16 -05:00
jason.woltje
9c80ba7f17
plans: brief for #1672 , skill reads keep a leading BOM and report fs errors as skill-refused (row 64 N2/N3, mos-0863 D4)
2026-10-19 16:18:46 -05:00
jason.woltje
46de608936
sage: window carry step for any pin past row 64, install and activate the seven role skills from the pinned copy (mos-0863 D3, #1557 )
2026-10-19 16:18:46 -05:00
jason.woltje
ebfc23c431
sage: row 64 packet, build notes, both gates, mutants and the landing manifest; BUILD.md item 5 corrected for Darkwing N1 (mos-0863 D5, #1557 )
2026-10-19 16:18:46 -05:00
jason.woltje
6b6567b8b4
harness: README names bundle-refused for the skill content refusals, skill-refused at the launcher (row 64 F1, #1557 comment 29218)
2026-10-19 16:11:42 -05:00
jason.woltje
9f00397110
roles, harness, cli: a managed session gets its role's skills as prompt text on both harnesses (row 64, #1557 )
...
Author and gate owner: Sage (mos-0855).
Role files take an optional skills list. The bundle reads each named skill
from <dataRoot>/skills-enabled, copies it into the bundle and renders every
file into prompt.md between skill-file markers, so the prompt text matches
the bundle copy byte for byte. The role's skills together are capped at
SKILL_BYTES (64 KiB): the claude adapter passes prompt.md as one argument,
and Linux caps one argument at 128 KiB. Symlinks are refused, not followed.
The launcher checks the list before a run exists: skill-not-enabled names
the activate command, anything else refuses skill-refused.
Pi keeps --no-skills; Claude Code keeps --disable-slash-commands and
--restricted. The two harness modules and MOSAIC_SKILLS are unchanged.
First lists: coder, reviewer, cto and pm, ms-unslop on each.
2026-10-19 16:11:36 -05:00
jason.woltje
1a01f63ead
queue: row 64 round 1 approvals, Filbert rev 2024 and Darkwing rev 2025 on manifest 7689cd9b ( #1557 )
2026-10-19 16:10:42 -05:00
jason.woltje
7c81d5e9c1
sessions: Filbert row 64 r1 review, approve
2026-10-19 16:03:07 -05:00
jason.woltje
12ac2c428b
sessions: Sage line for row 64 candidate, gate and in-review round 1
2026-10-19 15:58:19 -05:00
jason.woltje
e76c73f725
queue: row 64 started and in-review round 1 on c6d7d538f, reviewers darkwing and filbert ( #1557 , Sage revs 2022-2023 and the start rev)
2026-10-19 15:58:06 -05:00
jason.woltje
087d202bc3
queue: row 173 for the Seen click intent gap after row 172 ( #1671 , mos-0861 A1; Sage rev 2021)
2026-10-19 15:44:18 -05:00
jason.woltje
a0da43dab2
plans: brief for the Seen click intent gap after row 172 ( #1671 , mos-0861 A1)
2026-10-19 15:43:03 -05:00
jason.woltje
b5d4d0f225
sessions: Sage line for the row 172 landing and close-out
2026-10-19 15:38:53 -05:00
jason.woltje
c24fe5d0ff
build-log: row 172 landing, the dropped Seen click and the row 165 reload race ( #1670 )
2026-10-19 15:36:23 -05:00
jason.woltje
83db85aa3a
queue: row 172 round 1 Darkwing approve (darkwing rev 2018), landing note and done (Sage revs 2019-2020)
2026-10-19 15:35:49 -05:00
jason.woltje
be10e0c838
webui: a Seen click during a Board read waits for the read and posts, and the row 165 test loads the page instead of reloading from an evaluate (row 172, #1670 )
...
The click handler returned on busy, which refresh() holds for the whole /api/board read, so a click during a
poll did nothing and the sign-out test waited 5 s for a sign-out that was never asked for. The row 165 test's
location.reload() from an evaluate let the next wait start on the old document, and the reload killed it.
2026-10-19 15:34:00 -05:00
jason.woltje
7f38a2f549
sage: row 172 packet, base control and trace lines corrected (Darkwing D1, D2 on #1670 )
2026-10-19 15:34:00 -05:00
jason.woltje
0ed9412fca
sessions: Sage line for the 170/171/134 build-log and row 172 round 1
2026-10-19 15:21:23 -05:00
jason.woltje
9d6e47dd58
queue: row 172 brief re-pin, briefed, start and round 1 in-review on 15c3d8ca1 (Sage revs 2013-2017)
2026-10-19 15:21:04 -05:00
jason.woltje
17d5663054
sage: row 172 packet, candidate 15c3d8ca1 (cause trace, mutants, 10-run loop, gate)
2026-10-19 15:19:31 -05:00
jason.woltje
164e1f97fe
row 172 brief: the cause is a Seen click dropped during a Board read, and the reload race in the row 165 test (P19 366, #1670 )
2026-10-19 15:03:56 -05:00
jason.woltje
bab069b3c5
build-log: rows 170, 171 and 134 landings, with the row 134 range-diff evidence (mos-0859/0860)
2026-10-19 14:53:51 -05:00
jason.woltje
b42934b85c
queue: row 134 done on 3a600cf5b (Sage revs 2011-2012: note, done)
2026-10-19 14:48:47 -05:00
jason.woltje
a85e573d4f
cli README: the chief of staff reads the organization revision with organization_revision, not from the human (row 134 follow-up, Filbert #1669 29193)
2026-10-19 14:47:10 -05:00
jason.woltje
3a600cf5b2
bus, harness: a role holding organization.change reads the organization revision itself with organization.revision and the organization_revision tool (row 134, #1669 )
...
Researcher's candidate 8977d907 (patch 843fdee9), approved by Filbert in #1669 comment 29193. Built on d58c9c5cd and applied on 04155deff without fuzz; the patch's +/- lines hash the same on both bases. Row 171's one broker.mjs line (project-name-is-repo, at :2256) is the only change on a manifest path between them.
2026-10-19 14:35:31 -05:00
jason.woltje
04155deff6
queue: row 171 done on 078518e4e, plus revs 1996-2010 (Sage: 171 reopen/request/note/done, 172 add, 134 review request; Rocko 1997-1998 row 171 round 2 in-review; Researcher 2002-2004 row 134 start and round 1 in-review; Dewey 2007 row 171 approve; Filbert 2008 row 134 approve)
2026-10-19 14:34:30 -05:00
jason.woltje
078518e4e1
bus, cli, webui: a thread project named like a business-file repo refuses project-name-is-repo unless --repo matches, with a clearer CLI refusal (row 171, #1668 )
2026-10-19 14:31:36 -05:00
jason.woltje
b2d3b6c475
sessions: Filbert row 134 r1 review, approve
2026-10-19 14:28:54 -05:00
jason.woltje
ed045ac84a
dewey: SESSIONS line for the row 171 round 2 review ( #1668 )
2026-10-19 14:24:40 -05:00
jason.woltje
4d9114ab4c
sage: row 172 brief, the board-held sign-out wait under gate load (P19 366, #1670 , mos-0858); cause section open until traced
2026-10-19 14:21:35 -05:00
jason.woltje
d58c9c5cda
queue: row 170 done on cb43dcb45, plus revs 1970-1994 (Sage: 171 scope/re-pin/request/note, 134 re-pins/assign/issue/briefed, 64 and 66 re-pins, 64 start, 170 note and range-diff; Rocko 1970, 1973-1974 row 171 start and round 1; Dewey 1971, 1981-1982 row 170 claim and in-review, 1990 row 171 approve; Filbert 1986 row 170 approve)
2026-10-19 14:14:33 -05:00
dewey
cb43dcb45e
bus, cli: a tester message to a role nobody holds refuses tester-target-not-running, and talk --tester exits 3 (row 170, #1667 )
2026-10-19 13:59:15 -05:00
jason.woltje
228aeb5037
sage: rows 64/66 brief, skills reach both harnesses as prompt text (pi read is workspace-confined, Claude Code Skill tool offers its own skills; probe in agents/sage/work/queue-64/probe.md)
2026-10-19 13:58:02 -05:00
jason.woltje
c9f6e214f1
dewey: SESSIONS line for the row 171 round 1 review ( #1668 )
2026-10-19 13:52:51 -05:00
jason.woltje
2cef8cd350
sage: row 134 brief owns the three stale revision-from-the-human sentences (Researcher build report)
2026-10-19 13:46:29 -05:00
jason.woltje
9b11b87313
sessions: Filbert row 170 r1 review, approve
2026-10-19 13:42:36 -05:00
jason.woltje
2e9fe381f6
docs: register Sage's rows 171, 134, 64 and 66 brief work and the b0721a0b0 push correction
2026-10-19 13:41:54 -05:00
jason.woltje
e0bf9960dd
sage: rows 64 and 66 brief refreshed to 7319bd867 after row 65 (harness module files owned, plugins suite in the gates), row 64 candidate names Sage as author and gate owner (mos-0855)
2026-10-19 13:40:44 -05:00
jason.woltje
b0721a0b0d
dewey: SESSIONS line for the row 170 build and in-review ( #1667 )
2026-10-19 13:39:39 -05:00
jason.woltje
7319bd8673
sage: row 171 brief names cli.mjs (sage-1021, mos-0855), row 134 brief refreshed to 000730949 with owner Researcher, row 42 run log takes mos-0855's triggers ( #1668 )
2026-10-19 13:34:05 -05:00
jason.woltje
0007309492
queue: rows 170 and 171 briefed on Mos's approval, row 171 re-pinned to the mos-0854 brief (mos-0854, #1667 , #1668 ) (revs 1964-1969)
2026-10-19 13:23:54 -05:00
jason.woltje
fcfa577f20
sage: row 171 brief cites mos-0854 (D1 withdrawn, option B approved) in place of its open decision section ( #1668 )
2026-10-19 13:21:48 -05:00
jason.woltje
92160cb301
queue: rows 170 (P19 380, #1667 ) and 171 (P19 379, #1668 ) added queued, briefs 97cc7f0a3 await Mos; Darkwing's row 169 claim (revs 1961-1963)
2026-10-19 13:20:28 -05:00
jason.woltje
97cc7f0a32
sage: briefs for P19 380 (tester refusal, #1667 , row 170) and P19 379 (project names vs repos, option B, #1668 , row 171), for Mos's approval
2026-10-19 13:18:59 -05:00
jason.woltje
42817065e1
queue: row 169 after cleared, in-review waits on 166's landing by note (mos-0852 point 5, #1666 ) (revs 1959-1960)
2026-10-19 13:12:56 -05:00
jason.woltje
56534526b2
sage: Window B last section rewritten for the approved two-phase path, row 169 (mos-0848 (b), mos-0852, #1558 , #1666 )
2026-10-19 13:11:26 -05:00
jason.woltje
cbdcf1967f
queue: row 169 briefed on Mos's brief approval (mos-0852, #1666 ) (rev 1958)
2026-10-19 13:10:17 -05:00
jason.woltje
3ff2360dc3
sage: Window B P3 recorded done from Darkwing's rehearsal, with what it didn't exercise ( #1558 )
2026-10-19 13:06:33 -05:00
jason.woltje
c5b74bef20
darkwing: Window B P3 scratch rehearsal, activate --pm --migrate 3h to 3i and rollback --restore on blob 4b16e848 (sage-1004, mos-0838 D2, #1558 )
2026-10-19 13:06:33 -05:00
jason.woltje
3b5c964eb0
queue: row 169, the stack-release.sh two-phase PM start (P19 382, #1666 ), brief pinned at bfc5fa7ba3 (revs 1956-1957)
2026-10-19 13:06:06 -05:00