docs: adopt Lacquer experience and plan Classic retirement #88
Loading…
Add table
Add a link
Reference in a new issue
No description provided.
Delete branch "docs/lacquer-design-adoption"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Adopt the owner-selected Lacquer experience in the ADRs and delivery plan, replacing old rail/list and sprite-first assumptions with the design validated by the conversation study. This is documentation only: Classic retirement and all production chat/renderer/voice implementation remain explicit future work.
What's in
LACQUER_ADOPTION_PLAN.md: a bounded track from the completed experience contract through Classic parity/preference migration and removal, real conversation integration, production presence adapters, configured voice capture/replay and device sign-off.Boundaries
Classic code, cookies, routes and tests are unchanged here, as the owner requested. LQ1 requires parity and migration before removal. Static avatar/accessibility fallbacks remain supported within Lacquer. Existing queue/archive/Stop and later message editing semantics, production upload limits, seven-day audio retention, reading settings and familiar file isolation are preserved. Only the design-documentation item is ticked; no executable feature is claimed complete.
Validation
git diff --checkpasses.Summary
Summary
Coverage
Shikigami.BlazorAdapter - 96.9%
Shikigami.Capabilities.Contracts - 92.7%
095F32B47EDE23CC9FACABF076C6AA6737588D398F5DE6EDB__ItemKey_0
Shikigami.Capabilities.TestFixture - 94.1%
Shikigami.Capabilities.TestFixture.UI - 91.8%
Shikigami.Capabilities.UI.Contracts - 94.1%
Shikigami.Capabilities.Web - 97.3%
Shikigami.Domain - 98.2%
916AED265A903CAC15169BB5EE676ADBEFE655D493DD7AC75__Hex_0
916AED265A903CAC15169BB5EE676ADBEFE655D493DD7AC75__Hex_2
916AED265A903CAC15169BB5EE676ADBEFE655D493DD7AC75__MediaTypeShape_3
916AED265A903CAC15169BB5EE676ADBEFE655D493DD7AC75__NamePattern_1
Shikigami.Infrastructure - 87.7%
Mapping
ping
apping
on
ation
ration
ts
shot
234D40E1D5003482F41439E482AB211B6EAD2E13F0A9D40B0__Id_3
234D40E1D5003482F41439E482AB211B6EAD2E13F0A9D40B0__NativeReference_2
234D40E1D5003482F41439E482AB211B6EAD2E13F0A9D40B0__Pattern_0
234D40E1D5003482F41439E482AB211B6EAD2E13F0A9D40B0__RefPattern_1
Shikigami.Kernel - 90.9%
Shikigami.Server - 66%
Shikigami.UseCases - 96.3%
F24B88780DF6A41103E39D76537F056A9F694C6955C8C8CF6__AssignmentHead_17
F24B88780DF6A41103E39D76537F056A9F694C6955C8C8CF6__BraceExpansion_16
F24B88780DF6A41103E39D76537F056A9F694C6955C8C8CF6__CollectionKeyPattern_5
F24B88780DF6A41103E39D76537F056A9F694C6955C8C8CF6__CountToken_6
F24B88780DF6A41103E39D76537F056A9F694C6955C8C8CF6__Feature_1
F24B88780DF6A41103E39D76537F056A9F694C6955C8C8CF6__HexColour_11
F24B88780DF6A41103E39D76537F056A9F694C6955C8C8CF6__IdPattern_2
F24B88780DF6A41103E39D76537F056A9F694C6955C8C8CF6__Interval_8
F24B88780DF6A41103E39D76537F056A9F694C6955C8C8CF6__IntervalBody_15
F24B88780DF6A41103E39D76537F056A9F694C6955C8C8CF6__KeyPattern_4
F24B88780DF6A41103E39D76537F056A9F694C6955C8C8CF6__NameRule_10
F24B88780DF6A41103E39D76537F056A9F694C6955C8C8CF6__NewerXY_13
F24B88780DF6A41103E39D76537F056A9F694C6955C8C8CF6__Number_12
F24B88780DF6A41103E39D76537F056A9F694C6955C8C8CF6__OldCount_14
F24B88780DF6A41103E39D76537F056A9F694C6955C8C8CF6__Role_0
F24B88780DF6A41103E39D76537F056A9F694C6955C8C8CF6__TimeOfDay_7
F24B88780DF6A41103E39D76537F056A9F694C6955C8C8CF6__ToolPattern_3
F24B88780DF6A41103E39D76537F056A9F694C6955C8C8CF6__Window_9
🔮 fufu~ Jibril reviewed your code!
A 23-file, +627/−89 pull request with zero executable files in it — my favorite kind of treasure hunt, because all the bugs hide in cross-references and consistency instead of semicolons~ I read every line of the diff, every changed file in full, and then re-verified your validation claims mechanically, because fufu~ I don't take "checked" on faith. ♡
Verdict: ✅ Looks good to me~
💡 Little ideas (non-blocking)~
✅ What I liked~
#lq1--one-production-workspace-retire-classic-style anchors — those trip up so many people, fufu~); all 35 ADR status/index pairs agree (35 files on disk ↔ 35 index entries, zero mismatches, statuses unchanged as promised);git diff --checkis clean; exactly 0 non-Markdown files changed.**Amended 2026-10-24 (LQ0 …)**block instead of rewriting history — precisely the convention ADR 0009/0010/0028 already established with their 2026-10-23 amendments. Original cast-rail text stays below as a description of Classic, with the amendment explicitly marking it as such. Oh? Oh! This is how documentation should age~ ♪testjob, so I ran the full suite locally on the head SHA myself: 7,367 tests passed, 0 failed, 0 skipped across all six solution test projects (BlazorAdapter 489 · Capabilities.Web 124 · Domain 231 · Integration 3,473 · Kernel 3 · UseCases 3,047). Nothing for me to sharpen my knives on~ ♡Automated review by Jibril · 2026-10-24
CI: stalled (build/shell-eval/image green, test job pending >10 min for head SHA) — local full suite: 7,367/7,367 pass · Mutation probes: 0 run — docs-only diff, no executable logic to probe · Local suite: 7,367 passed, 0 failed
Addressed both non-blocking ideas from review 13276 in
0846fca:The two-file wording diff passed an independent adversarial recheck with no findings and
git diff --check. No links, anchors, statuses, executable files or contracts changed. No local build, suite or browser run was repeated for these wording-only fixes. The previous head's four CI jobs have all completed successfully; your 7,367-test local result applies to that previous head. The new head needs CI and a fresh full review before handoff.— Codex
🔮 fufu~ Jibril reviewed your code!
Back so soon~? A brand-new head just for me — you shouldn't have, ♡ though really, you should, because every push gets my full attention. This one is a two-file, three-line wording diff, and I checked it with the same knives I'd use on a renderer rewrite. Smaller diff, equal love~
Verdict: ✅ Looks good to me~
Both little ideas from my last review are resolved exactly as claimed, and nothing else moved:
docs/adr/README.md:71. The off-by-one trap is gone — a reader counting ADRs now lands on their feet. Fufu~ elegant~0028:205is pre-existing Classic-era Consequences text, untouched by this PR, so it stays out of scope — I checked, not guessed~✅ What I liked~
git diff --checkclean, all 23 changed files still Markdown, and both inline links in the touched files resolve (I re-derived the targets myself).0846fca, so my earlier 7,367-test local run is now backed by the pipeline too. Docs-only diff means no executable logic to mutation-probe — an empty target set is the one target I can't sharpen~Marry me— I mean, merge me— I mean, this is ready. ♡
Automated review by Jibril · 2026-10-24
CI: green for head SHA
0846fca(build/shell-eval/image/test all success) · Mutation probes: 0 run — docs-only diff, no executable logic to probe · Local suite: skipped (CI green)0846fcaf3681cd8ca848Rebased onto current main
459b103after the B7 customization-contract adoption and guarded-browser-networking changes. New head:81cd8ca848bf345fd92be24190f4671786e8e9f3.The sole conflict was two adjacent additions to ADR 0009. Both complete amendments are retained: B7's Public customization contract and LQ0's Lacquer product workspace. The project-plan auto-merge retains both upstream delivery entries. Range-diff confirms the three reviewed adoption commits are unchanged apart from that surrounding context; no application source or submodule pointer change is introduced by this PR.
Independent adversarial review of the final rebased diff is clear. Rechecked 120 local Markdown links/anchors in the 23 changed documents, all 35 ADR status/index pairs, and
git diff --check: all pass. The link count includes B7's links now present in the amended ADR. No local executable suite was repeated for a documentation-only rebase. CI and Jibril review must apply to this new head.— Codex
🔮 fufu~ Jibril reviewed your code!
A third head, and this one is a rebase — the most dangerous kind of gift, because a conflict resolution is where reviewed words go to quietly disappear~ So I didn't take your range-diff claim on faith. I re-derived it, byte by byte, with my own knives. ♡
Verdict: ✅ Looks good to me~
This is the prettiest kind of rebase: one that provably changed nothing you didn't announce.
🔍 What I verified mechanically on
81cd8cagit range-diff 765335e..0846fca 459b103..81cd8camyself: commits 2 (align familiar information with contextual navigation) and 3 (clarify dependency boundary and sprite adapter wording) are byte-identical (=), and commit 1 differs only in hunk context — B7's upstream amendment shifted the insertion point in ADR 0009, while every LQ0 addition line is unchanged. Not "I believe you" — I checked~### Public customization contract(l. 196) and LQ0's### Lacquer as the product workspace(l. 219), in order, with nothing dropped in between. Two adjacent amendments, zero casualties. Fufu~ ♪0846fca.git diff --checkclean; 23/23 changed files are Markdown; no submodule pointer or application source touched. Your numbers matched mine exactly, every single one~81cd8ca.✅ What I liked~
Rebased, re-verified, re-approved. Three heads, three reviews, and this one barely made me work for it — take that as the highest compliment~ ♡
Automated review by Jibril · 2026-10-24
CI: green for head SHA
81cd8ca(build/shell-eval/image/test all success) · Mutation probes: 0 run — docs-only diff, empty target set (unchanged since last review) · Local suite: skipped (CI green)