A. Room lifecycle (总管 review items A-1/A-2): the single cleanup owner now watches
ACTIVE session deadlines as well as debts - a session nobody touches expires into
the ledger and has its room really deleted with zero further requests (old code:
_cleanup_loop only scanned debts, expired rooms were never deleted until some
request ran _prune). Explicit end and the background sweep share ONE persisted
attempt/age cap and abandoned state - _sessions no longer bypasses it (the
manager measured 2 background attempts + 3 more interactive deletes = 5); an
abandoned debt raises CleanupAbandoned -> 503 with room/attempts/reason instead
of pretending done-404, success clears both _sessions and _debts, restart
continues the same persisted cap, close/cancel leaves no dangling coroutine, and
only this instance's own rooms are ever touched. New wake event makes a fresh
debt retry immediately instead of waiting out a deadline sleep.
B. Queued speech turns bind EXACTLY (总管 interleaving note_handle(H1)->H2->
note_stream(S1)->S2 gave H1=None, H2='turn-one'): CyberGrillLLM.chat now reads
the SDK's own agents_speech_handle ContextVar (livekit-agents 1.8.8 sets it in
_create_speech_task around every speech pipeline; source-verified on .12 and
asserted at runtime) so each ChatApiLLMStream carries speech_handle/speech_id of
the very handle whose generation created it. TurnLedger binds only on that exact
attribution - a newer speech NEVER retires an older pending handle, say/proactive
generations bind nobody, missing attribution fails closed (no ack, no subtitle
turn), interrupt/failure discards only its own binding.
C. Versioned subtitle contract v2 for the visual author's wiring card (HEL-898):
every cg.subtitle payload gains {v:2, seq} (session-monotonic), turn identity
(speech_id, turn_id, same-source message_id for proactive rows) and explicit
lifecycle events - turn_start/assistant_interim/assistant_final/turn_end/
turn_cancel(reason) plus proactive_start/proactive_end/proactive_cancel that
ride the say SpeechHandle instead of the receipt branch. Accumulators are
isolated per speech_id (a late delta after cancel is dropped producer-side);
seq + speech_id give the frontend two independent stale-rejection rules. Full
executable contract: docs/voice-subtitle-contract-v2.md with recorded samples in
docs/rework/hel879/subtitle-samples/ (scripts/gen_subtitle_samples_hel879.py).
web/src untouched.
Verification: the .12 real-SDK check now LOADS THE ACTUAL WORKER MODULE and drives
the callbacks IT registered on a real AgentSession (no hand-copied functions) -
both-queued delayed generations bind exactly, a say sandwiched between user turns
steals nothing, unattributed generations fail closed, interrupted turns cancel
without final/receipt (11/11 ALL PASS, source-level proof of the ContextVar wiring
included; evidence docs/rework/hel879/sdk-check-hel879-remaining.log). The old r2
lifespan test flake is root-caused (clock freeze and block were two separate portal
calls; a 10ms life tick could slip between them and claim the frozen day unblocked)
and fixed atomically without raising timeouts or dropping assertions - 12/12 solo,
full suite 350/350 three times in a row on .venv311; ruff back to the single
pre-existing error; voice chain 8/8; release offline R1-R7 with zero failures;
requirement checks 4/4 (origin/main and c121c42, each with/without handoff).
Zero real LLM/embed/ASR/TTS calls, zero downloads/installs, nothing deployed or
merged, real caps stay 0.
Co-authored-by: multica-agent <github@multica.ai>
CyberGril 需求入口
当前完整基准:v1.11。批准需求继承办法的主卡评论:01a1635b-b47c-79ed-ac66-9408c7905af9。
所有新任务、续办、设计、施工、审核先读完整当前需求,再读本次相关来源。只有白栖知明确修改的要求才可变更,阶段范围不能覆盖产品要求。
| 文件 | 用途 |
|---|---|
| current.md | 60条完整有效要求:原31条产品决定、13条视觉、3条补充、6条用户整改、6条执行约束、1条交付边界;另列未定项 |
| baseline.json | 同一份要求的结构化编号、边界和来源 |
| coverage.md / coverage.json | 实现与验收状态,未完成项目继续保留 |
| changes.md / changes.json | 后续修改的用户授权及影响 |
| task-handoff-template.md | 每张卡的必填交接项 |
| sources/ | 原始v1.6、完整初轮和视觉访谈、后续裁定、获批设计及分批方案 |
| source-manifest.json | 来源文件校验值,防止历史来源被静默改写 |
| implementation-verification.md | 本次实际写入、旧卡迁移与正反向验证记录 |
sources/comment-evidence.json 保存初版125条来源评论(保留原样)的身份、时间、原话和父评论;正文中的编号可追溯回主卡。原始初轮访谈与完整视觉访谈均用分页取齐,后续材料按关联决定保存。其余来源文档保留原样;其中旧进度、技术推荐、旧待定项不能自动成为当前事实。已经被后续明确决定覆盖的内容,见 current.md 和 changes.md。
原仓库 docs/SPEC.md 是初始构想,不能盖过后续逐项确认:例如旧VRM路线、允许经历改写出生核心等,均不能用于覆盖当前要求。
当前返工与交付边界
用户对当前前端的对话显示效果和整体沉浸界面不满意。本轮先完善需求约束,再制作独立样图及可操作动态小样;两主题、电脑/手机、长句与身份、历史展开、交流开合和空间过渡都须能查看。小样未认可前不替换正式界面。
完整产品仍需按获批批次接入生活、记忆、主动联系、语音、天气/素材/备份及有条件的视觉观察。R1已有档案与文字能力不能视为全部要求完成。嘴型研究独立,不阻塞完整界面设计。
检查与人工审核
先获取实际目标分支,再运行 python3 scripts/check_requirements.py --base-ref origin/main(实际目标不是 main 时替换引用);有交接JSON时加 --handoff <文件>。检查会对照目标阻止原地改需求却不升版、不追加新用户授权,且保留历史来源和变更记录。初次引入基准时由独立审核核对整包来源。此检查不证明画面好看、评论语义构成授权或用户已经满意。人工审核必须阅读原话,体验验收仍由白栖知决定。
当前没有配置仓库自动CI合并门禁;上述命令作为交付和独立审核的必做项,不声称平台已强制拦截所有修改。
v1.8新增来源见 sources/comment-evidence-visual-role-20261022.json,最新分工见GOV-005。
v1.9新增来源见 sources/comment-evidence-shared-space-20261023.json 与 sources/shared-space-static-acceptance-20261023.md;仅更新VIS-011交流静图认可边界,相处、过渡、雾光、手机与真实人物动态仍待验收。
v1.10新增来源见 sources/comment-evidence-dynamic-sample-20261023.json 与 sources/dynamic-sample-effect-acceptance-20261023.md;对坐动态小样 v2 整体效果(演示层面)记入 VIS-011 与 CG-004 边界,手机真机、候选容器启用、真实人物动态与完整功能仍待验收。
main 当前仍为 v1.8(提交 3a986726d8593e0ed2274f79db6cc5d807d2ebf0);v1.9 与 v1.10 在待合并分支线(agent/agent/54df81e610fd → agent/hel-811-handoff-f59f74dc4256 → 后续分支)上递进,合并入 main 由总管与白栖知决定。
后续以该入口所记录的最新版本为准,不把v1.7永久钉死为永不可新增的文档。每次新版本必须记录用户明确裁定,保留旧版来源与覆盖关系。
v1.11依据用户评论 01a1734c-1159-7a44-a81f-3105f1b8d98d 新增GOV-006成熟方案优先复用;来源见 sources/mature-reuse-principle-20261025.md,交流材料提炼见 discussion-takeaways-20261025.md。仅为治理记录,待审核入库,不声称main已更新或产品已采用任何底座。