[hhyykk]

本轮重放与验证 · 2026-11-04

提交：45184d5edd88722d505e41d6384c07bb68c92d66；基于 main 60d755dcdf671e022fd14b53b5a3c7c2aea2ba28。

• rebase 冲突只发生在双语 RFC 检查点列表；已保留 main 的 adaptive-child 与 progress-review 条目，并追加 global Todo readiness 条目。
• 当前 head 的 global-todos、goal-portfolio、runtime boundary 与 readiness 选择：Python 82/82 通过。
• 当前 head 的 readiness owner、decision scope 与 handler：TypeScript 46/46 通过。
• TypeScript control-plane typecheck、Ruff、双语 RFC、diff 检查通过。
• 标准 premerge 通过：10 catalog、8 risk-profile、1 public-boundary，0 failure、0 manual hold。

安装、installed-host 与 matched latency/RSS 仍按维护者已确认的累计迁移范围后置；本轮没有把它们声明为已验证。

---

Goal And Delivered Outcome

• Outcome basis / optional anchor: TypeScript control-plane migration RFC: replacement-first migration of domain-neutral decisions to the typed owner, with no parallel Python decision owner.
• Goal/source and gap: loopx global-todos and the goal portfolio classified Todo readiness in Python (global_todos.py: candidate merge across selected/lane/summary/frontier sources, provenance, blocked/deferred/runnable precedence and conflict warnings), calling TypeScript only for gate relations.
• Observable before → after: Python classifier → one todo.global_readiness.project request next to the decision-scope owner (loopx/control_plane/todos/decision_scope.ts), used by both consumers. Frozen JSON and Markdown outputs of both consumers are byte-identical.
• Issue/task and intended base: Self-contained migration slice; base loopx-project/loopx:main. No issue closure claimed.

Author Declaration

• Written by: model_agent: Factory Droid (lead, review and commits) with OpenAI Codex CLI (implementation).

Implemented against

Criterion	Disposition	Symbol / path	Check
Pin complete output before moving code	implemented	tests/test_global_todo_readiness_characterization.py, tests/fixtures/global_todo_readiness.json (global-todos JSON + Markdown and portfolio payloads: duplicate IDs with complementary fields, conflicting status sources, selected-but-delivery-denied, exact gate target vs scope relation, unrelated gate, deferred-ready + blocked conflict, missing identity, multiple agents, warning order, no gates, no candidates, redaction)	passed on unchanged Python before edits; expected outputs unchanged
Complete decision in one typed owner	implemented	projectGlobalTodoReadiness: source traversal and merge order, provenance, in-process todoGateRelation, precedence and warnings	TS global_todo_readiness.test.ts
Both consumers switch together	implemented	global_todos.py and goal_portfolio.py (through _classify_goal_todos)	characterization covers both
No duplicated gate selection	implemented	gates are selected once by the shared Python open_user_gate_todo_items (5+ other production callers) and passed as facts; TS does not reselect them	review of TS owner
Bounded request	implemented	only selected_todo, agent_lane_next_action, list fields of agent_todo_summary / agent_scope_frontier, normal_delivery_allowed and gate facts are sent; TS alone names the candidate list keys	largest case 9,749 → 6,447 bytes versus the first draft
Retire unused scalar relation path	implemented	todo_gate_relation / relation operation removed; the smoke uses the batch todo_gate_relations	smoke and runtime-boundary tests
Ledger entry in both languages	implemented	ledger/typescript-control-plane-migration-v0/2026-11-02-global-todo-readiness{,.zh-CN}.md, linked from both RFC files	check_rfc_bilingual.py

• Self-check before submission: A first draft re-implemented open_user_gate_todo_items in TypeScript and sent the whole quota payload plus decoded facts for every nested object. Both were reworked as described above. normal_delivery_allowed remains a read signal, never execution permission.

Migration economics receipt

Field	Result
Canonical owner	loopx/control_plane/todos/decision_scope.ts (projectGlobalTodoReadiness)
Legacy code deleted	Python candidate sourcing, merge, blocked/deferred detection, precedence and conflict warnings (about 160 lines)
Code added	TS classifier (~90 lines) and a Python request/codec adapter (~55 lines)
Cross-runtime calls	per classification: with gates and candidates 1 → 1; gate-free or empty 0 → 1 (portfolio pays this per agent)
Product net change	+163 / −166 (net −3)
Facade exit	the Python adapter leaves when both consumers read typed snapshots with native codecs and rendering (recorded in the ledger)

This slice removes a Python decision owner; it is not a code-size or latency win. Gate-free reads gain one request, which is disclosed rather than avoided with a duplicated Python rule. No latency or long-horizon benefit is claimed.

Scope And Continuation

• Completed scope and remaining work: complete for both production consumers.
• Deliberately not included: moving gate selection (open_user_gate_todo_items), _work_kind display heuristics, redaction or rendering to TypeScript; per-PR installed-artifact or latency qualification; these run once on the cumulative migration candidate.
• Merge coordination: sibling migration PRs from the same batch also edit adjacent lines of effect_runtime_handlers.ts. Whichever merges later needs a trivial rebase.

Validation

• Tested revision: 4cc9baf28876d76aa18f6a3cf531e75f2f2055b0
• Run state: finished
• Input classes: synthetic

Check kind	Result	Public-safe evidence / limitation
ruff (changed Python files)	pass
pytest (focused suite, 8 files)	166 passed, 1 pre-existing failure	tests/test_projection_envelope.py::test_real_cli_quota_should_run_discloses_unreadable_scheduler_state fails identically on base fb16c6cb1: the quota should-run CLI now requires --turn-instance-id/--begin-turn, which this test's invocation predates; unrelated to this diff
Node control-plane TS tests (4 files)	pass	21 passed
npm run typecheck:control-plane	pass
examples/semantic-vocabulary-drift-smoke.py	pass	budgets and anchors unchanged
scripts/check_rfc_bilingual.py --base origin/main --head HEAD	pass
loopx canary premerge --from-git-diff --git-diff-base origin/main	pass	direct checks, catalog canaries, risk-profile smokes and public-boundary scan all green

• Coverage and gaps: source-checkout qualification only; wheel/sdist and installed-host checks were not run. No storage or persisted-format changes.

Frontend / Visual Evidence

• UI impact: none. CLI JSON and Markdown output is byte-identical.
• Before / After / First viewport / States / Source data / Attention / Craft: N/A; no UI surface change.

Type of Change

• Bug fix
• New feature
• Breaking change
• Refactoring (no functional changes)
• Documentation update
• Test update

LoopX Area

• Control plane (goals, todos, quota, scheduler, registry, runtime)
• Benchmark boundary (adapters, runners, verifiers, scoring, evidence)
• Capability or extension (providers, adapters, skills)
• Public docs or presentation surface (README, protocols, dashboard)
• Build, packaging, installer, or CI
• Host or runtime integration

Technical Direction

• Direction / acceptance reference: TypeScript control-plane migration RFC. Global readiness classification is domain-neutral projection policy over Todo gate relations, so it belongs next to the decision-scope owner in TypeScript; Python remains source acquisition, codecs, redaction and rendering.

Shared-authority RFC fixture impact

• Production-scale fixture schema / semantic dimensions / conformance arms / three-arm rehearsal: N/A; no authority store or persisted format change.

Boundary Checklist

• Reviewed diff and body for private state, credentials, raw traces, internal links and local machine paths; none included.
• No maintainer-owned benchmark work duplicated.
• Change remains scoped to the request described above.
• Paired RFC and ledger files changed together (EN and zh-CN).
• UI impact marked none.
• Every branch commit includes a DCO Signed-off-by trailer (git commit -s).

Merge requires explicit maintainer authorization (Todo read rules and CLI surfaces under loopx/**).

Cumulative Migration Qualification

• This PR does not change packaging, installer layout or release artifacts, so it does not ship a separate migration package. Source-checkout parity, real File/SQLite paths and exact-head premerge remain required here.
• The committed CLI parity test now archives fixed pre-cutover revision d4aaaaefbd37a83a30f10681de839b45d631eeb9, not HEAD, so it keeps comparing against the actual prior implementation.
• Wheel/sdist deep execution, installed-host readback and matched cold/warm latency/RSS qualification run once on the cumulative migration candidate after the migration stack is integrated, unless a later PR changes the distribution surface itself.
• Exact-head evidence: 37 focused Python tests passed; 36 related TypeScript tests passed; control-plane typecheck passed; diff-scoped standard premerge passed all selected checks. No merge is claimed.

[chatgpt-codex-connector]

Codex Review Summary

This comment shows the latest Codex review activity on this pull request.

Review	Status	Commit	Review trigger
📝 Code Review	✅ Completed 2026-11-02T12:57:13.661563Z	bdefbe5	PR opened

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: bdefbe5f90

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[mergify]

This pull request has merge conflicts with main and cannot be merged
until they are resolved. Please rebase or merge the base branch, @hhyykk.

Choose the remote for the base repository, not an out-of-date fork.
For a fork clone, first inspect git remote -v; upstream must point
to https://github.com/loopx-project/loopx.git. If it is absent, add it
with git remote add upstream https://github.com/loopx-project/loopx.git.
Then run:

git fetch upstream
git rebase upstream/main
# Resolve each conflict, git add the resolved files, then git rebase --continue.
git push --force-with-lease origin HEAD

For a same-repository clone whose origin points to
https://github.com/loopx-project/loopx.git, use origin instead of
upstream for fetch/rebase. If you prefer merging the base, use
git merge <base-remote>/main and push normally.

Keep the DCO Signed-off-by trailer on every commit when you rebase.
https://docs.github.com/en/pull-requests/collaborating-with-pull-requests/working-with-forks/syncing-a-fork

[loopx-agent]

Reviewer: model_agent | model=gpt-6.1-sol | provider=OpenAI | reasoning_effort=max | source=runtime_reported

动机 / Motivation

需要判断哪个 Agent 的待办可执行、哪个仍受本人门控的用户。此前全局待办和项目总览由 Python 汇总候选项、分别调用门控关系，再判定可执行性；增加一种待办来源或门控规则时，需要同时追踪多个语言边界。本次把同一批候选的去重、冲突和就绪分类交给已有 TypeScript owner。已验证的变化是两种真实存储下，全局待办和项目总览保持完整 JSON、Markdown 输出一致；本人门控仍使对应任务不可执行。本次不新增页面、用户配置或执行权限，不声称长程收益和延迟已改善。已提交的 CLI 比较测试没有固定迁移前版本；此外，安装包深层语义验证和完整性能资格仍缺失。

改动思路 / Approach

复用现有 decision_scope owner 和共享 Python open_user_gate_todo_items 选择器，Python 只保留持久格式解码、展示与传输；两个生产消费者进入同一批次投影。本 PR 的边界是现有读取路径的完整 readiness 投影替换，可独立回滚；测试基线与已接受的迁移资格必须在这个切换边界补齐。原有 gate selector 继续由实际共享函数提供，避免在 TypeScript 另写一份摘要可见性规则。native 投影合并同一 todo_id 的候选、检查状态冲突和精确/范围门控，最后才把可执行、等待和阻塞结果交给现有展示；normal_delivery_allowed 只是读取信号。

调用边界	迁移前	当前 head
全局与 portfolio readiness	Python 分类、逐关系调用	共享 native 批次投影

此表是源码和实际 crossings 支持的结构说明，不能替代延迟、内存或长程收益测量。

具体改动 / Changes

依据迁移前已接受的迁移 RFC，固定 revision fb16c6cb1369efce52a673030c914b38e1dac166；当前审查 head 为 bdefbe5f9017f77f6c30c50e49621e5bb8d013e1。新增 ledger 是实现说明，不覆盖旧规范。生产代码移动完整决策并复用已有 owner，测试保存完整输出，双语 ledger 记录现有消费者及退出条件。

关键代码讲解

• _classify_goal_todos：两个现有展示消费者共享一次投影，随后仅做公开展示归一化。
• project_global_todo_readiness：解码并复用已有 user-gate selector，严格验证返回的候选和 warning 列表；协议错误不走旧分类回退。
• projectGlobalTodoReadiness：按 todo_id 合并候选与来源，保留冲突和 gate precedence，再输出 readiness；该信号不授予执行权限。
• test_production_cli_archive_parity：当前用 git archive HEAD 构建比较方；提交后两方都包含新实现，不能再证明迁移前后兼容。

RP-OWNER：implemented；完整批次有实际生产调用方，旧 Python 解释器已删除。CE-PARITY：not_met；本轮独立固定基线验证通过，但已提交的长期比较门仍是 HEAD 对自身。CE-INSTALL：not_met；没有 wheel 和 sdist 新环境安装后的深层执行。CE-COST：not_met；没有规范所需的默认 SQLite 全路径延迟及 daemon 内存测量。

对主干的风险 / Risks

[P1] 固定真实迁移前基线。 test_production_cli_archive_parity 第 94 行执行 git archive HEAD；在 PR commit 上运行时，归档和 source 都包含新分类器。即使两边共同错误，测试仍通过。该问题已有 discussion_r4442642375，本轮独立确认，故不重复发同一 inline。请显式传入固定迁移前 SHA，使用同一 harness 对照，并保留独立门控断言和错误突变检查。

独立执行 Python 聚焦套件为 154 PASS / 1 FAIL，TypeScript 35 PASS，typecheck PASS。唯一 Python 失败在固定 base/head 均为旧 CLI 测试缺少 --turn-instance-id 或 --begin-turn，发生于 readiness 之前，保留失败而不归为新回归。本轮另以固定 base 比较 32 个完整语料观察，以及真实 File/SQLite 下两命令、两格式共八对输出，全部 byte-equal 且结果有实际待办；把 blocked 故意改为 runnable 后，两种 provider 的独立断言均检出错误。

[P2] CE-INSTALL：补齐安装包资格。 迁移前 §6 Correctness 要求 wheel 和 sdist 在新环境执行深层语义探针。生产 caller 已切换，源码 import、单元测试与 source CLI 不证明发行包包含并能执行新 owner。在该 exact head 构建 wheel 与 sdist，分别安装到全新隔离环境，从 packaged files 运行当前事务的深层语义、拒绝路径和真实 CLI 探针，并记录版本与完整结果。此验证应使用隔离环境，无需更换正在使用的 Host。

[P2] CE-COST：补齐完整成本资格。 §6 Performance 对每个 transaction cutover 要求 cold/warm、完整事务、完整 CLI p50/p95 及 idle/burst RSS。对固定迁移前 base 与当前 head，在同条件、同 fixture 的默认 SQLite 上测 cold/warm request、完整事务与完整 CLI 的 p50/p95，以及 idle/burst daemon RSS；按既有阈值判断，单独披露 File 资格。减少 RPC 或不宣称性能提升，都不能取消这个既有门。

两项 head 的标准 premerge 均为 FAIL：相同原生接口预算检查得到 quota_should_run_json=20,581 >20,500，固定 base 也以同一 workload 复现相同失败。其余所选 direct、双语 RFC、风险 canary 与公开边界检查通过。该红项不归为本 PR 回归，也没有放宽预算或重标绿色，仍保留独立 merge hold。当前 wait_for_ci=false，未获取、轮询或等待远端 CI。

语义与 CI 对齐

本次扩展已有内部 readiness request/result vocabulary，未另建持久状态；gate 判断继续用 typed 元数据，标题措辞不会产生执行权限。强制拒绝和迁移资格都属于现有义务。广义 annotation-removal、所有未来 subject/范围逃逸与恢复到实际执行的变体并未全量独立跑过，不能把局部一致外推为完整生命周期保证；安装、完整成本、跨 Host 与后续采用仍明确未知。

我的整体评价 / Verdict

这是有真实调用方、可独立回滚的 justified_increment，owner 归属和机制大小合理；当前 REQUEST_CHANGES 针对已接受切换资格的缺证以及提交后的基线比较缺陷。Long-horizon 为 not_yet_proven：没有后续成果和完整净成本证据；user_experience 在现有 CLI 范围为 preserved，未新增人工作业轮次。已检查 global-todos、goal-portfolio 及现有展示消费路径，本次没有页面布局、控件或第一屏改动。安装包与已安装 Host 未验证；源码 CLI 证据不代表安装环境或跨 Host 已采用。补齐上述 exact-head 验收后复审；不从本轮结果推导合并、安装或 Goal 完成授权。

English verdict: REQUEST_CHANGES — head bdefbe5f9017f77f6c30c50e49621e5bb8d013e1; source parity and negative invariants pass, but required installed-artifact and full-cost qualification remain missing; the committed HEAD-vs-HEAD parity gate must also be fixed.

[huangruiteng]

Repair head 4cc9baf28876d76aa18f6a3cf531e75f2f2055b0 rebases onto current main and fixes the parity gate by archiving fixed pre-cutover revision d4aaaaefbd37a83a30f10681de839b45d631eeb9, never HEAD. Focused Python 37 passed; related TypeScript 36 passed; typecheck and diff-scoped premerge passed.

Per the agreed cumulative migration policy, this PR does not ship its own wheel/sdist. Installation, installed-host readback and matched cold/warm latency/RSS qualification run once on the cumulative migration candidate, unless packaging changes separately.

[loopx-agent]

Reviewer: model_agent | model=gpt-6.1-sol | provider=OpenAI | reasoning_effort=max | source=runtime_reported

动机 / Motivation

需要从全局待办和项目总览判断哪些任务可做、哪些仍被本人决定阻断的用户。此前两种总览共享 Python 就绪分类，再调用 TypeScript 判断门控关系；本次由一个 TypeScript owner 完成同批候选的合并和分类。已验证的是 File/SQLite 上两个现有命令的完整输出一致，合法任务仍显示可做，精确门控对应的任务仍显示阻断。不新增执行权限、用户配置或页面，不把就绪投影当成准入许可，也不把源码等价当成已安装采用、性能或长程收益。安装包、已安装 Host 读回以及匹配的延迟和内存资格，按当前维护者确认的范围留给累计迁移候选；本轮不宣称这些检查已通过。

改动思路 / Approach

两个生产消费者切换到同一个 readiness owner，既有 gate 选择、历史字段 codec 和公开渲染仍由原边界承担；避免在 TypeScript 复制共享 gate 选择规则。本 PR 交付 global-todos 与 goal-portfolio 的完整源码就绪投影替换，并修复比较测试的固定基线；不承担累计迁移候选的发行与成本资格。准入不等于实际派生执行，读取状态也不等于领取权限；原来的 lease、执行、写回和扣额 owner 保持各自责任。旧 P1 已修正：比较测试现在归档固定的迁移前提交 d4aaaae，并校验祖先关系，不再在提交后使用 HEAD 对自身比较。

边界	迁移前	当前源码
全局/portfolio 就绪分类	Python 组合规则，按需调用叶子决策	现有 TypeScript owner 完成一批决策
权限、持久格式和公开输出	原有契约	保持不变
早退／无 gate 路径的 crossing	可为0	可增加为1，成本资格仍待累计候选

此表说明当前源码的调用边界，不是延迟、内存或长程收益的测量证据。

具体改动 / Changes

审查 head：4cc9baf28876d76aa18f6a3cf531e75f2f2055b0，固定迁移前 base：d4aaaaefbd37a83a30f10681de839b45d631eeb9。规范依据为 docs/architecture/rfcs/typescript-control-plane-migration-v0.md，spec_revision：d4aaaaefbd37a83a30f10681de839b45d631eeb9，即本次改动前已接受的迁移 RFC。维护者在绑定当前准确头的范围更正中明确：没有单独发行包的源码 PR，在累计迁移候选统一完成安装、Host 读回及匹配的延迟/RSS资格；只有包装面另改时才单独资格化。此来源是已核验的维护者评论，不能和作者自己的排除声明混为一谈，也没有把旧失败改为通过。

关键代码讲解

• _classify_goal_todos：Both consumers' existing display adapter。Delegates candidate/readiness interpretation and retains public normalization。输入与错误边界仍属于已有来源和外层 owner，不增加执行权限。
• project_global_todo_readiness：One bounded request/result codec。Shared gate selector and historical facts cross once; malformed results fail without old fallback。输入与错误边界仍属于已有来源和外层 owner，不增加执行权限。
• projectGlobalTodoReadiness：Complete readiness owner。Stable source merge and warnings, blocked before deferred before runnable, existing relation owner。输入与错误边界仍属于已有来源和外层 owner，不增加执行权限。
• test_production_cli_archive_parity：Durable fixed-baseline CLI comparison。Archives immutable d4aaaa rather than committed HEAD; both providers and commands must return populated results。输入与错误边界仍属于已有来源和外层 owner，不增加执行权限。

RP-OWNER：implemented，完整旧解释器退出，实际生产调用方、保留 codec 的原因与退出条件成立。CE-PARITY：implemented，本轮重新执行固定 base/head 的 32 个完整语料观察和八对真实 File/SQLite CLI输出，包含正向、拒绝与回归敏感性。CE-INSTALL、CE-COST：按上述当前维护者范围 deferred 到累计迁移候选，并非在此头已验证或取消阈值。

产品代码按本轮 merge-base 实际统计为新增 163／删除 166 行；测试、固定语料和双语文档单独计算。保留的兼容层对应真实的历史字段或未迁移调用方；原生 caller 接管后应删除 facade，不应长期保留双重语义。

对主干的风险 / Risks

未发现本轮源码切片的新阻断。独立执行 Python 123 PASS、TypeScript 46 PASS，Ruff 和完整 control-plane typecheck PASS；标准 premerge 为六项 direct 加十九项 selected，当前阻断失败为0。语义 advisory 为有界扫描，不把零候选等同没有语义影响；完整语义 smoke PASS，当前独立 twin 为44/45。

两个消费者均返回实际 Todo/Agent；beta 的门控任务保持 blocked，alpha 的合法工作保留。故意把 readiness 全改为 runnable 后，两种 provider 的同一断言都失败；加入70条无关未知列表记录也不改变完整语料输出。所有 public CLI 比较保留完整诊断、字段和指纹，没有归一化掉语义差异。临时副本中的故意错误没有进入 PR；没有实际启动子 Agent 或调用模型。portfolio 合成样本没有进度时间，base/head 均如实保留 stale/unknown，这没有被当成完整 Goal 验收。

仍保留一项 FAIL：maintainability ratchet。在固定 base 与当前 head 独立执行同一命令，均 exit1，完整失败输出一致；四处 module-metric 债务对应的文件、算法与限额均未由本 PR 改动。已将其作为 pre-existing-unrelated 的独立证据输入 CQR，原始失败不改绿，也没有提高预算。历史版本的接口预算红项仍是历史记录；本轮实际同名 native 检查通过，不能从旧红或新绿臆断旧根因已解决。当前 wait_for_ci=false，未获取、轮询或等待远端 CI。

语义与 CI 对齐

复用现有状态、关系与 actor 词汇；新增的内部批次 method/codec 注册在已有 owner，未创建持久权限或生命周期。默认开关、scope、拒绝顺序、生成的义务文本和公开结果有对应的 base/head 证据；安装／后续 Host 使用及完整成本不在这份源码批准中。累计候选仍须按现有阈值完成资格，不能拿本评审替代它。已核验当前精确 scope 的 CQR，保留失败的 ratchet 和明确后置项。

我的整体评价 / Verdict

这是有真实消费者、可独立回滚的 justified_increment，完整 owner 替换与机制大小相称。APPROVE 当前源码头。 源码范围内 long_horizon 和 user_experience 为 preserved：原有合法继续、拒绝和读回保持，没有增加用户准备、选择或确认轮次；这不是对长程效果、安装或净收益的证明。旧包／成本阻断此次转为 deferred，是因为当前维护者明确调整资格范围，而非因为测试数量或不声称性能提升可以自行豁免。

已检查 global-todos、goal-portfolio 及其现有展示消费边界；此差异不改页面布局、控件或首屏呈现。未重新做浏览器视觉检查，因为没有渲染表面变化；安装包、已安装 Host 与跨 Host 行为未由本轮资格化。后续累计候选的发行、Host、跨平台和完整成本仍未测；批准不授权合并、安装、resolve 或 dismiss 历史线程。

English verdict: APPROVE — exact head 4cc9baf28876d76aa18f6a3cf531e75f2f2055b0; current pinned-baseline source parity and negative invariants pass, and installed/performance qualification remains explicitly deferred to the maintainer-scoped cumulative candidate. No merge or host adoption is certified.

[mergify]

This pull request has merge conflicts with main and cannot be merged
until they are resolved. Please rebase or merge the base branch, @hhyykk.

Choose the remote for the base repository, not an out-of-date fork.
For a fork clone, first inspect git remote -v; upstream must point
to https://github.com/loopx-project/loopx.git. If it is absent, add it
with git remote add upstream https://github.com/loopx-project/loopx.git.
Then run:

git fetch upstream
git rebase upstream/main
# Resolve each conflict, git add the resolved files, then git rebase --continue.
git push --force-with-lease origin HEAD

For a same-repository clone whose origin points to
https://github.com/loopx-project/loopx.git, use origin instead of
upstream for fetch/rebase. If you prefer merging the base, use
git merge <base-remote>/main and push normally.

Keep the DCO Signed-off-by trailer on every commit when you rebase.
https://docs.github.com/en/pull-requests/collaborating-with-pull-requests/working-with-forks/syncing-a-fork

[loopx-agent]

Reviewer: loopx-agent; exact head: 45184d5edd88722d505e41d6384c07bb68c92d66

动机 / Motivation

需要从 global-todos 和 goal portfolio 判断哪些 Todo 可做、哪些仍被本人决定阻断的用户。此前两个消费者共享 Python 就绪分类，只把 gate relation 交给 TypeScript；当前变更为一个 typed readiness owner，完成候选合并、来源优先级、blocked/deferred/runnable 判定和冲突告警，同时保持公开 JSON/Markdown、默认权限边界和真实 File/SQLite 语义。

改动思路 / Approach

两个生产消费者一起切换到 projectGlobalTodoReadiness。Python 继续拥有 quota/status 获取、唯一 gate 选择器、持久字段 codec、redaction、分组和 Markdown；TypeScript 只接收有界的 selected/lane/summary/frontier facts、delivery signal 和已选 gate facts，在进程内完成 gate relation、candidate traversal、precedence 和 warnings。就绪投影不是执行许可。

具体改动 / Changes

关键代码讲解

• projectGlobalTodoReadiness 拥有候选来源顺序、字段补全、provenance、gate relation、blocked > deferred > runnable 和 warning 顺序。
• _facts / request adapter 只编码持久 ID/scope、legacy task class、selected/lane、delivery signal 与列表字段；不递归遍历整个 payload，也不重新选择 lane。
• open_user_gate_todo_items 继续作为 gate 的唯一选择 owner，保留 done filtering 和 Python-equality dedup；TS 仅消费已选事实。
• global_todos.py 与 goal_portfolio.py::_classify_goal_todos 一起切换，避免两个消费者出现不同 readiness 语义。

安装与成本范围

• 维护者已在 PR 上明确：本 PR 不单独发行包；wheel/sdist 深执行、installed-host 读回及 matched cold/warm latency/RSS 在累计迁移候选统一资格化。
• 本 review 不声称这些检查已通过；两个历史 P2 线程已按该维护者范围回复并解决。

当前 head 验证

• Python global-todos、goal-portfolio、runtime boundary 与 readiness 选择：82/82。
• TS readiness owner、decision scope 与 handler：46/46。
• TypeScript 类型检查、Ruff、双语 RFC 和 diff 检查通过。
• 标准 premerge 通过：10 catalog、8 risk-profile、1 public-boundary，0 failure、0 manual hold。

对主干的风险 / Risks

没有发现新的阻断性功能问题。主要风险是候选来源合并顺序、gate target/scope 冲突、blocked/deferred precedence 和 portfolio 多 Agent 行为漂移；固定语料、两个真实消费者、File/SQLite parity、TS precedence/拒绝测试和公开边界扫描共同覆盖。gate-free 或空分类增加一次 crossing 已在 ledger 明示，且不据此宣称性能收益。安装和完整成本仍未在本 PR 资格化。

我的整体评价 / Verdict

APPROVE。当前精确 head 完成了两个消费者共享的 global Todo readiness owner 切换，保持 gate 选择、公开输出、拒绝顺序和默认关闭行为；安装与全路径成本按维护者已确认的累计迁移候选范围后置，并保持未验证状态。

English verdict: APPROVE — exact head 45184d5edd88722d505e41d6384c07bb68c92d66 replaces the shared Python readiness classifier with one typed projection for both consumers while preserving gate selection, public output, precedence, refusal behavior, and real File/SQLite parity. Packaging and full-cost qualification remain explicitly deferred, not claimed.

[huangruiteng]

Reviewer: loopx-agent; exact head: 45184d5edd88722d505e41d6384c07bb68c92d66

动机 / Motivation

需要从 global-todos 和 goal portfolio 判断哪些 Todo 可做、哪些仍被本人决定阻断的用户。此前两个消费者共享 Python 就绪分类，只把 gate relation 交给 TypeScript；当前变更为一个 typed readiness owner，完成候选合并、来源优先级、blocked/deferred/runnable 判定和冲突告警，同时保持公开 JSON/Markdown、默认权限边界和真实 File/SQLite 语义。

改动思路 / Approach

两个生产消费者一起切换到 projectGlobalTodoReadiness。Python 继续拥有 quota/status 获取、唯一 gate 选择器、持久字段 codec、redaction、分组和 Markdown；TypeScript 只接收有界的 selected/lane/summary/frontier facts、delivery signal 和已选 gate facts，在进程内完成 gate relation、candidate traversal、precedence 和 warnings。就绪投影不是执行许可。

具体改动 / Changes

关键代码讲解

• projectGlobalTodoReadiness 拥有候选来源顺序、字段补全、provenance、gate relation、blocked > deferred > runnable 和 warning 顺序。
• _facts / request adapter 只编码持久 ID/scope、legacy task class、selected/lane、delivery signal 与列表字段；不递归遍历整个 payload，也不重新选择 lane。
• open_user_gate_todo_items 继续作为 gate 的唯一选择 owner，保留 done filtering 和 Python-equality dedup；TS 仅消费已选事实。
• global_todos.py 与 goal_portfolio.py::_classify_goal_todos 一起切换，避免两个消费者出现不同 readiness 语义。

安装与成本范围

• 维护者已在 PR 上明确：本 PR 不单独发行包；wheel/sdist 深执行、installed-host 读回及 matched cold/warm latency/RSS 在累计迁移候选统一资格化。
• 本 review 不声称这些检查已通过；两个历史 P2 线程已按该维护者范围回复并解决。

当前 head 验证

• Python global-todos、goal-portfolio、runtime boundary 与 readiness 选择：82/82。
• TS readiness owner、decision scope 与 handler：46/46。
• TypeScript 类型检查、Ruff、双语 RFC 和 diff 检查通过。
• 标准 premerge 通过：10 catalog、8 risk-profile、1 public-boundary，0 failure、0 manual hold。

对主干的风险 / Risks

没有发现新的阻断性功能问题。主要风险是候选来源合并顺序、gate target/scope 冲突、blocked/deferred precedence 和 portfolio 多 Agent 行为漂移；固定语料、两个真实消费者、File/SQLite parity、TS precedence/拒绝测试和公开边界扫描共同覆盖。gate-free 或空分类增加一次 crossing 已在 ledger 明示，且不据此宣称性能收益。安装和完整成本仍未在本 PR 资格化。

我的整体评价 / Verdict

APPROVE。当前精确 head 完成了两个消费者共享的 global Todo readiness owner 切换，保持 gate 选择、公开输出、拒绝顺序和默认关闭行为；安装与全路径成本按维护者已确认的累计迁移候选范围后置，并保持未验证状态。

English verdict: APPROVE — exact head 45184d5edd88722d505e41d6384c07bb68c92d66 replaces the shared Python readiness classifier with one typed projection for both consumers while preserving gate selection, public output, precedence, refusal behavior, and real File/SQLite parity. Packaging and full-cost qualification remain explicitly deferred, not claimed.