Why: R-platform spans core/ui/frontend/documents; also honors the
agreed housekeeping item — .env.dev must be gitignored (it holds live
API keys).
Evidence: make test full suite green on this head (exit 0); headless
walk on the real binary: avatar columns, settings shell with badges,
admin-row avatars, zero console errors.
Impact: capability layer only, no behavior changes. Merge order:
core -> ui -> frontend -> documents -> this PR.
Refs: proposal 2026-09-30-ui-redesign-and-auth.md (capability platform)
Why: R4 spans ui (token scale) and frontend (icon set + restyle); the
Dev repo advances pointers and the task record.
Evidence: make test full suite green on this head (exit 0); headless
journey on the real binary across both themes with zero console errors
and zero leftover emoji glyphs.
Impact: pure visual layer. Merge order: ui -> frontend -> this PR.
Refs: proposal 2026-09-30-ui-redesign-and-auth.md (R4)
Why: R2 spans backend (migration v10, auth/session APIs, CLI user
subcommands), frontend (login page, forced change gate, admin user
management), documents (decision record).
Evidence: make test full suite green on this head (exit 0); headless
journey on the real binary with --auth on covers admin bootstrap,
first-login forced change for created and reset passwords, non-admin
menu gating, API-key login, logout.
Impact: auth-enabled deployments gain password login and admin account
management; embedded no-auth mode unchanged. Merge order: backend ->
frontend -> documents -> this PR.
Refs: proposal 2026-09-30-ui-redesign-and-auth.md (R2 admin-issued)
Why: give the catalog's WebSearch prompt a carrier by registering
Agent-Native's Tavily tools behind settings, with library copy on top.
Evidence: backend 16 packages green incl. 4 new search tests; live
boot checks for both unconfigured and configured states.
Impact: pins backend 0d55ebb, documents b4546fe.
Refs: feature/tavily-search-tool task card.
Why: close the coverage gap against the user's take-everything-portable
bar and record the porting rule.
Evidence: backend 16 packages green at 44-module floor; documents guide
section 9 carries the full disposition table.
Impact: pins backend ed036cb, documents 60bb2ed.
Refs: feature/prompt-catalog-coverage task card.
Why: R1 implementation spans core (Conversation flags), backend
(migration v8, task list APIs, /api/me), frontend (TaskSidebar rebuild),
ui (contracts mirror), documents (proposal acceptance record); the Dev
repo advances the pointers and the task record.
Evidence:
- make test full suite green on this head (core+backend+frontend+ui+
agent-command, exit 0) before advancing gitlinks.
- Headless walk on the built binary: cross-project task open, pinned-
first ordering, archive hide/show/restore, new-task flow, account
footer identity, zero console errors.
Impact: UI restructure per approved option A; no mid-column or inbox
changes (D25/D26/D31 intact). Merge order: core -> backend -> ui ->
frontend -> documents -> this PR.
Refs: proposal docs/proposals/2026-09-30-ui-redesign-and-auth.md (option A, R1)
Why: prompts move from hardcoded strings to a layered embed library
with trigger-based assembly, user overrides and observability.
Evidence: composite make test green over this combo; runtime check on a
live mock server (40 modules, 7 layers, override round-trip, skill
command routing); task card records results and remaining items.
Impact: pins core 0cb0a40, backend 4938bd0 (with its vendored core
gitlink), frontend 4abf13d, documents 2715ed6.
Refs: feature/prompt-library task card; design guide in Documents.
Why: the archived card still had a placeholder for its main-record PR
link, and the combo pins pointed at feature-branch commits instead of
the merged submodule main tips.
Evidence: Dev PR #29 merged as c6f5368; submodule PRs Frontend #9, UI
#9, Backend #12 merged (tips 0eb90665, 6068f146, 10f400a2, content
identical to the pinned feature commits).
Impact: one-line card backfill plus three gitlink advances; no code
change.
Why: the UI redesign proposal and demos live in the Documents repo; the
Dev repo advances only the submodule pointer and the task record.
Evidence:
- modules/documents -> e601b25 (proposal + 5 HTML demos; verified headless,
see that commit); make test full suite passed before advancing the
gitlink (core+backend+frontend+deps, exit 0).
- Task card filled with outcome, verification, and the pending user
decision (direction A/B/C, auth stage-2 scope).
Impact: gitlink + .task only; no code change. Documents PR #14 carries
the content; merge order: Documents PR first, then this PR.
Refs: Documents PR https://git-home.u1680832.nyat.app:30319/weichen/Agent-System-Documents/pulls/14
Why: the shipped chat column showed nothing until a run finished, blew
up markdown spacing, and gutted long documents before the model saw
them.
Evidence: composite make test green over this combo (backend 16 pkgs
incl. new ctxasm/ingest tests, frontend 9, ui 23); headless E2E on the
mock backend confirmed live thinking/tool/text streaming and compact
list geometry.
Impact: pins frontend 616795b, ui 0689d72, backend fe4887b; task card
records results and remaining user-side acceptance on a real model.
Refs: feature/chat-live-quality task card.
Why: the chat page could not scroll once messages exceeded the viewport
and the input bar was pushed off-screen (missing min-height:0 on the
column flex chain .main->.chat->.chat-body).
Evidence: combo make test green with the advanced frontend gitlink;
headless-Chromium before/after measurements and four layout variants
recorded in the task card.
Impact: middle column only; also ignore /data-test/ local runtime data
so precheck does not report BUSY on the user's test artifacts.
Refs: Agent-System-Frontend a54022c (feature/chat-scroll-fix).
Why: T5 closes the interaction-foundation proposal (T0-T5); the Dev combo pins the verified combination.
Evidence: zero-page walkthrough 12/12 PASS on a fresh real binary; combo make test PASS at pinned heads; records at documents docs/records/2026-09-30-t5-polish-verification.md (PR#13).
Impact: gitlinks frontend/documents updated; task card backfilled; no Dev-side code change.
Refs: task card feature/polish-r1; proposal T5
Why: T3 submodule PRs merged (backend #10, ui #7, frontend #6, documents #11) and the Dev combo must pin the verified combination.
Evidence: combo make test PASS at pinned heads (backend 15 pkgs incl. cmdbuf; ui 18/18; frontend 9/9); record at documents docs/records/2026-09-30-t3-command-bus-verification.md (PR#11).
Impact: gitlinks backend=0f74026 frontend=6a2b21a ui=76f976d documents=<PR#11 head>; task card backfilled; no Dev-side code change.
Refs: task card feature/command-bus-r1; proposal T3
Why: Backend merged its consumer-adaptation PR (nested core/agent-command
pins); the Dev combo must pin that merge so the tested combination is the
publishable one.
Evidence: full make test re-run at this exact combo (see task card evidence).
Impact: gitlink only.
Refs: modules/documents docs/records/2026-09-30-t1-core-cards-commands-verification.md
Why: Provider repos merged for T1 (core PR#2, agent-command PR#2, frontend
PR#3, ui PR#3, documents PR#9); gitlinks pinned to those mains. The
task-finish guard prevents a repeat of the T0 incident where running
finish from the main checkout deleted it (scripts/task.sh treated $PWD as
the worktree to remove).
Evidence: make test green on the pinned combo — core, backend 13 pkgs,
frontend 9, ui 6, agent-command (vet + tests incl. conformance); make build
produces the binary; combo version string now includes agentcmd SHA.
Impact: combo grows its version traceability; AGENTS.md documents the
worktree-cwd requirement for finish.
Refs: modules/documents docs/records/2026-09-30-t1-core-cards-commands-verification.md
Why: Foundation task T0 of the UI/interaction base (proposal 2026-09-30,
D25-D33): add the new UI component library and command-bus ecosystem repos,
extend build/test/pin orchestration to twelve submodules.
Evidence: make test green on the pinned combo (core + backend 13 pkgs +
frontend 9 + ui 6 tests); make build produces bin/agent-system with combo
version string including ui; make contracts now generates to frontend and ui
with submodules staying clean; renderer/tokens migration verified
CSS-rule-identical (zero behavior change).
Impact: New modules/ui and deps/agent-command gitlinks; frontend consumes ui
via source alias; backend vendors agent-command (replace only, no import
yet); follows provider-first merges (ui PR#1/#2, agent-command PR#1,
documents PR#8, frontend PR#2, backend PR#6).
Refs: modules/documents docs/proposals/2026-09-30-ui-inbox-command-workspace.md
Why: the startup configuration surface (file/env/CLI layers, config
explain, instance examples) merged in both component repos; the composition
must consume the reviewed combination.
Evidence: backend PR#4 (1e62188, 12 packages pass, explain live-checked)
and documents PR#6 (84b3421); make test on this branch per the gitlink rule.
Impact: gitlink-only advance; also picks up the nested core-pin build-blocker
fix (019617e -> 4ee1979) that makes fresh backend clones materializable.
Refs: docs/proposals/2026-09-30-configuration-surface.md
Why: the library now makes precedence, memory tiers, and lesson
mechanization explicit; the composition must pin the revision and register
how its layers map onto the new rules.
Evidence: guidelines PR#3 (ed2af3e) and documents PR#4 (2205138) merged;
make test on this branch per the gitlink rule.
Impact: gitlink advances; adoption registers the hierarchy/memory rows as
manual-with-existing-artifacts. No code change.
Refs: policy/adoption.md; docs/records/2026-09-30-history-migration-sha-map.md
Why: the two 2026-09-30 lesson entries merged into documents main; the
composition should consume them for future retrieval.
Evidence: documents PR#3 merged (2ec3e50); make test on this branch before
merge per the gitlink rule.
Impact: gitlink-only advance; no product code change.
Refs: docs/records/2026-09-30-lesson-*.md
Why: the backend store List* functions encoded empty results as JSON null,
which crashed the frontend nav (.map on null) on any newly created project.
Evidence: backend go test all packages pass; store regression
TestEmptyListsSerializeAsEmptyArrays asserts [] for empty lists.
Impact: parent pins backend at the serialization fix plus the core-pin
build-blocker advance.
Refs: .task/fix-empty-list-json.md
Why: the nine submodule entry docs and the documents migration record merged
via their own PRs; the parent must pin the new combination and register the
adoption with honest enforcement levels.
Evidence: per-repo PR merges (9x entry-docs PR#1, documents PR#2) verified
via API; make test on this branch follows before merge.
Impact: gitlinks advance to post-migration, gate-adopted mains; policy/
adoption.md registers enforced (local hooks) vs advisory vs manual honestly.
Refs: ef89fc8; Development-Guidelines 0.1.0-draft.7
Details (original message):
Advance backend gitlink (intentgate integration a671d60) with its new deps pins; example plugin intent asset uses the overlay matcher format.
weichen
weichen