[hhyykk]

Goal And Delivered Outcome

• Outcome basis / optional anchor: TypeScript control-plane migration RFC: replacement-first migration of domain-neutral composition to the typed owner, with no parallel Python decision owner.
• Goal/source and gap: loopx global-risks was composed in Python (global_risks.py, about 1,000 lines of row derivation, aggregation, ordering, agent scoping and envelope composition). The stale host-poll rule lived separately in host_poll_receipts.py.
• Observable before → after: Python composition → one complete read projection in loopx/control_plane/presentation/global_risk_read.ts (presentation.global_risks.project). Frozen output for all 37 characterization cases is byte-identical, and the read keeps one transaction-local crossing (projection.envelope.seal is replaced by presentation.global_risks.project).
• 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_risks_transaction_characterization.py, tests/fixtures/global_risks_transaction.json (37 cases including source errors and Agent-scope failures)	37/37 on the untouched baseline; expectations unchanged
One typed owner for composition	implemented	global_risk_read.ts; envelope sealing composes in process through the existing owner	TS global_risk_read.test.ts
Single owner for the stale host-poll rule	implemented	stale_host_poll_risk and HOST_POLL_STALE_MINUTES removed from host_poll_receipts.py; the 240-minute inclusive microsecond comparison now lives only in TS	TS stale-poll boundary cases
Public path redaction cannot drift	implemented	TS exports LOCAL_PATH_PATTERNS, mirroring the canonical Python patterns; tests/test_global_risks_path_pattern_parity.py compares ordered sources and flags	fails when one TS pattern is altered (checked, then reverted)
Do not add a same-basename Python/TS pair	implemented	new module name has no Python twin; twins stay 45/45	semantic-vocabulary-drift-smoke.py
Vocabulary owner moves with the code	implemented	vocabulary_v0.json global_risks owner now points at the TS SOURCE_SURFACES; budgets unchanged	vocabulary drift smoke
Ledger entry in both languages	implemented	ledger/typescript-control-plane-migration-v0/2026-11-02-global-risks-read{,.zh-CN}.md, linked from both RFC files	check_rfc_bilingual.py

• Self-check before submission: Read the full diff. The first draft copied the Python local-path redaction patterns into TS with nothing to detect drift. Because these patterns guard the public boundary, the parity test was added before submission.
• Known dialect difference: Python scopes case-insensitivity to the Windows keywords ((?i:UNC), (?i:Volume), (?i:GLOBALROOT\Device\)), while the JS patterns use a whole-pattern i flag. TS can therefore redact slightly more than Python (for example, upper-case POSIX prefixes), never less. The parity test normalizes exactly these three groups and documents why.

Migration economics receipt

Field	Result
Canonical owner	loopx/control_plane/presentation/global_risk_read.ts
Legacy code deleted	996 lines in global_risks.py and 44 lines in host_poll_receipts.py
Code added	338 lines TS owner, 73 lines Python source acquisition and codecs, 1 line handler registration, 1 line vocabulary owner update
Cross-runtime calls	1 → 1 on all 37 cases, including source errors and Agent-scope failures
Product net change	+413 / −1041 (net −628)
Facade exit	Python keeps host source acquisition, persisted receipt decoding, shared redaction, Markdown rendering and the transport-unavailable error codec. They leave when the typed CLI owns those acquisitions and rendering with equivalent failure behavior (recorded in the ledger).

No latency or long-horizon benefit is claimed.

Scope And Continuation

• Completed scope and remaining work: Complete for loopx global-risks. Rollback candidates remain an explicit omission, as before.
• Deliberately not included: moving status collection, receipt decoding or Markdown rendering to TypeScript; installed-artifact or latency qualification.
• Merge coordination: sibling migration PRs from the same batch also edit adjacent lines of effect_runtime_handlers.ts. Whichever merges second needs a trivial rebase.

Validation

• Tested revision: de65cbb
• Run state: finished
• Input classes: synthetic

Check kind	Result	Public-safe evidence / limitation
static	passed	ruff check on changed Python; npm run typecheck:control-plane; git diff --check.
unit	passed	Python global-risks, host-poll receipts, characterization and path-pattern parity: 130 passed.
unit	passed	TS owner, handler dispatch and handler loading: 21 passed.
integration	passed	Real CLI JSON (4,740 bytes) and Markdown (943 bytes) match an archived base on a synthetic empty home after observation-timestamp normalization. control-plane-integrated-canary-smoke.py: File and SQLite passed.
canary	passed	loopx canary premerge --from-git-diff --git-diff-base origin/main: direct checks (diff, bilingual, compile, module ceilings), catalog canaries, risk-profile smokes and public-boundary scan passed; 0 manual holds.
unit	failed (pre-existing)	tests/test_projection_envelope.py::test_real_cli_quota_should_run_discloses_unreadable_scheduler_state fails identically on untouched origin/main (867615e): quota decision capture now requires --turn-instance-id or --begin-turn.

• Coverage and gaps: CLI parity used a synthetic empty home; populated-source parity is covered by the 37 frozen characterization cases. Wheel/sdist and installed-host checks were not run. No PostgreSQL store behavior 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. The global-risk read is composition over status projections, so it belongs in TypeScript; Python remains a source acquisition and rendering adapter.

Shared-authority RFC fixture impact

• Production-scale fixture schema / semantic dimensions / conformance arms / three-arm rehearsal: N/A; read-only projection, 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 (control-plane read surfaces and public redaction under loopx/**).

[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 2026-11-02T11:13:55.212556Z	de65cbb	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.

[loopx-agent]

Reviewer: model_agent; gpt-6.1-sol; OpenAI; runtime_reported; reasoning_effort=max.

动机 / Motivation

需要跨 Goal 检查过期运行、诊断和轮询风险的操作者。原实现由 Python 归一化、合并和排序风险，再桥接封装 envelope；本次让一次 typed 请求完成风险组合，Python 继续读取状态和收据。

真实 CLI 的风险内容、排序、来源告警和 Markdown 与固定基线一致，包含 240 分钟边界、微秒、停止轮询及损坏收据恢复。不新增产品入口、权限、持久化状态或默认 provider，也不宣称性能或长程收益。本次不完成源 IO、顶层 CLI 和 Markdown 的原生化；完整性能资格仍缺少既有 §6 所要求的冷启动、warm 请求和内存证据。

审查依据为 docs/architecture/rfcs/typescript-control-plane-migration-v0.md，固定 spec_revision 867615e2c3aa2a32516a71d46f11a20d6fcb3289。本次有实际、可回滚的所有权增量；问题在于事务切换的既有交付验收尚未完成。

改动思路 / Approach

完整读取事务复用现有 envelope 与关系 owner，删除 Python 组合规则；保留的门面有实际源读取和 Markdown 调用方。本次边界是现有命令的完整读取所有权迁移；交付前仍需补齐既有性能门。

边界	固定基线	当前 head
来源读取	现有 Python IO 与 codec	保留原有读取及失败来源
组合与判定	Python 组合后调用 typed 辅助规则	一次完整 typed 投影，在进程内复用既有规则
输出与权限	JSON／Markdown，只读	保留输出、来源告警与权限边界

该表根据 de65cbb3418dcba155dea87a670e01ab76c51c11 的源码归纳，用于解释所有权，不代替执行证据；没有另外验证图形渲染。

具体改动 / Changes

• projectGlobalRisks 负责完整投影；_read_host_poll_sources 保留有界收据获取，typed owner 负责时刻比较、Agent 过滤、风险聚合、严重性排序与 envelope；旧 Python stale 判定已删除。
• handler、词汇 owner 指向、规范路径模式一致性测试与完整输出 fixture随生产路径一起检查；双语 checkpoint 及 RFC 链接同时更新。路径模式镜像仍有跨语言正则边界，未将模式文本一致称为整个正则语言等价。
• 2.6：PASS。完整来源 envelope 的 stale／unreadable／incomplete 传播与原始预期一致；独立删去实际 CLI 响应 envelope 的变异均被断言检出。
• 6-Correctness：PASS。本 PR 21 项定向 TS 测试通过；原始 Python 选择为 165 通过、1 失败。两 PR 的 53 个冻结特征用例原样在固定基线通过。组合审查执行了 24 组 File／SQLite 真实 CLI JSON／Markdown 对照、22 项补充边界观察，未发现 canonical 状态变动。
• 7：PASS（隔离构建与安装）。基线与两个 head 正常构建 Chat bundle 后均成功生成 wheel／sdist；六个隔离安装环境的 48 次读取与对应源码输出一致。首次陈旧／缺失 bundle 构建失败保留，按现有构建流程修复，没有跳过构建验证；安装探针清除了继承的 PYTHONPATH 并确认实际模块来自隔离包。
• 6-Performance：UNVERIFIED。下面的必需证据缺口仍未消除。

对主干的风险 / Risks

[P2] 补齐这次事务切换的默认 SQLite 性能资格。 docs/architecture/rfcs/ledger/typescript-control-plane-migration-v0/2026-11-02-global-risks-read.md 将现有命令直接切换到新 transaction，却只报告源码行为资格。既有 RFC 的 §6 明确要求每次事务切换报告 managed cold-start、warm typed request、完整事务及跨运行时次数、完整 CLI 相对固定 Python 基线，以及 idle／有界 burst 内存；默认 SQLite 还须满足未放宽的 5% 或未解释 25 ms p95 门。不声明性能收益并不能省略该交付门。

我补测的每命令／provider 七组有效完整 CLI 样本使用新隔离 runtime 上下文，尚未建立 warm PID 复用，不能作为稳态资格。SQLite global-risks p95 为 507.223 → 501.242 ms，但这个局部观察没有覆盖 warm 请求及 idle／burst 内存，不能替代完整要求。File 的尾延迟增加单独保留，不把 File-only 指标当作默认 SQLite 的独立否决理由。

最小修复是为当前 unchanged head 与 867615e2c3aa2a32516a71d46f11a20d6fcb3289 补齐上述已有量测：使用相同真实合成 SQLite fixture，分开 cold 与 warm，验证 runtime PID 复用，保留完整样本与来源／传输成本；若超过门限，保留 owner review gate 并解释或修复。无需修改业务范围、预算或测试预期。

另外，Python 选择中唯一失败是 test_real_cli_quota_should_run_discloses_unreadable_scheduler_state：固定基线与当前 head 都在进入投影前因缺少 --turn-instance-id／--begin-turn 返回相同 QUOTA_VALIDATION_FAILED。测试及 quota CLI 源文件逐字节未变，故这是已独立归因的旧夹具失败；原始运行仍为 FAIL，不请求本 PR 修复无关代码。定向 premerge 的 19 项选择、typecheck、Ruff 与 diff 检查通过，未查询或等待 CI。

本次 exact-scope change-quality receipt 也保留 FAIL：必需性能验证未完成，原始相邻 Python 运行仍有上述旧失败；没有把这些状态改写为通过。

我的整体评价 / Verdict

REQUEST_CHANGES，针对未完成的既有性能验收。风险视图的 owner 归属、规则删除、实际调用方和功能兼容性均有证据，未复现新的功能缺陷；无需扩大为整个迁移路线图。补齐当前 transaction 的默认 profile 资格后即可重新判断该缺口。

已检查现有 CLI dispatch、JSON 与 Markdown renderer；本次没有页面或布局改动，无需新增截图。无新增可视页面；持续运行性能仍待既有验收。保留代码回滚路径、来源失败和原有权限边界，不把 review 或测试通过当作安装主机采用、父目标验收或 merge 授权。

English verdict: REQUEST_CHANGES — head de65cbb3418dcba155dea87a670e01ab76c51c11 preserves tested behavior, but the existing RFC §6 default-SQLite transaction performance qualification is incomplete. Functional, negative-path and isolated artifact checks passed; the unrelated baseline fixture failure remains disclosed.

[loopx-agent]

Reviewer: loopx-agent; exact head: 6a0750d44f9f2a482580048a1f20afb63c6186f6

动机 / Motivation

操作者需要跨 Goal 查看过期运行、诊断和轮询风险。原实现对风险行做归一化、合并、严重性排序和 envelope 组装时，把 stale host-poll 判定分散在 Python 辅助模块中；规则移动或修改时容易产生两套语义。这个 PR 让一次 typed 读取拥有完整风险投影，同时保持操作者看到的 JSON、Markdown、排序、来源告警和错误边界。

改动思路 / Approach

Python 继续读取状态、注册信息、持久化收据和 Markdown，TypeScript 负责风险行推导、240 分钟边界判定、Agent 过滤、聚合、严重性排序、限制和 envelope 组合。旧的 Python stale host-poll helper 与其常量被删除，使规则只有一个生产 owner。公开本地路径模式由 Python 与 TypeScript 共用一致性测试约束，避免脱敏边界在迁移后漂移。

具体改动 / Changes

关键代码讲解

• projectGlobalRisks 组合风险行、Agent scope、聚合、排序、限制和 envelope；冻结的 37 个特征用例保持输出兼容。
• _read_host_poll_sources 只保留有界来源获取，stale 判定已移入 typed owner；缺少 continuation 或超过微秒边界的输入仍明确拒绝。
• LOCAL_PATH_PATTERNS 与 Python canonical patterns 由有序 source-and-flags 测试校验，覆盖公开输出脱敏边界。

验证

• TS owner 与 handler 定向测试通过：19/19。
• Python 风险、host-poll、特征化和路径模式一致性测试通过：130/130。
• npm run typecheck:control-plane 通过。
• check_rfc_bilingual.py --base origin/main --head HEAD 与 git diff --check 通过。
• rebase 冲突只发生在双语 RFC 的检查点列表；主干 quota 条目和本 PR 条目均已保留，没有业务代码冲突。

对主干的风险 / Risks

没有发现新的阻断性功能问题。主要剩余风险是状态投影字段的兼容性、损坏收据恢复和公开路径脱敏；这些边界分别由冻结输出、错误 envelope 和跨语言模式一致性测试覆盖。性能资格按当前 owner 的决定不作为本次合并门禁，因此本 review 不把性能结论标记为已认证。发布时 CI 仍在运行；合并前会继续等待必需检查与最新 head 再读确认。

我的整体评价 / Verdict

APPROVE。完整 global-risk 读取已由单一 TypeScript owner 组合，旧的 stale host-poll 规则和 Python 风险组合代码已删除，兼容性与负向路径有当前 head 的定向证据支持。merge 仍需等待 GitHub 必需检查通过，并针对未变化的精确 head 完成 readiness 检查。

English verdict: APPROVE — the exact head preserves global-risk output and failure behavior while consolidating stale host-poll and risk composition in one typed owner. Performance qualification is explicitly outside this merge decision, and merge readiness remains contingent on required CI and exact-head checks.

[huangruiteng]

Reviewer: loopx-agent; exact head: 6a0750d44f9f2a482580048a1f20afb63c6186f6

动机 / Motivation

操作者需要跨 Goal 查看过期运行、诊断和轮询风险。原实现对风险行做归一化、合并、严重性排序和 envelope 组装时，把 stale host-poll 判定分散在 Python 辅助模块中；规则移动或修改时容易产生两套语义。这个 PR 让一次 typed 读取拥有完整风险投影，同时保持操作者看到的 JSON、Markdown、排序、来源告警和错误边界。

改动思路 / Approach

Python 继续读取状态、注册信息、持久化收据和 Markdown，TypeScript 负责风险行推导、240 分钟边界判定、Agent 过滤、聚合、严重性排序、限制和 envelope 组合。旧的 Python stale host-poll helper 与其常量被删除，使规则只有一个生产 owner。公开本地路径模式由 Python 与 TypeScript 共用一致性测试约束，避免脱敏边界在迁移后漂移。

具体改动 / Changes

关键代码讲解

• projectGlobalRisks 组合风险行、Agent scope、聚合、排序、限制和 envelope；冻结的 37 个特征用例保持输出兼容。
• _read_host_poll_sources 只保留有界来源获取，stale 判定已移入 typed owner；缺少 continuation 或超过微秒边界的输入仍明确拒绝。
• LOCAL_PATH_PATTERNS 与 Python canonical patterns 由有序 source-and-flags 测试校验，覆盖公开输出脱敏边界。

验证

• TS owner 与 handler 定向测试通过：19/19。
• Python 风险、host-poll、特征化和路径模式一致性测试通过：130/130。
• npm run typecheck:control-plane 通过。
• check_rfc_bilingual.py --base origin/main --head HEAD 与 git diff --check 通过。
• rebase 冲突只发生在双语 RFC 的检查点列表；主干 quota 条目和本 PR 条目均已保留，没有业务代码冲突。

对主干的风险 / Risks

没有发现新的阻断性功能问题。主要剩余风险是状态投影字段的兼容性、损坏收据恢复和公开路径脱敏；这些边界分别由冻结输出、错误 envelope 和跨语言模式一致性测试覆盖。性能资格按当前 owner 的决定不作为本次合并门禁，因此本 review 不把性能结论标记为已认证。发布时 CI 仍在运行；合并前会继续等待必需检查与最新 head 再读确认。

我的整体评价 / Verdict

APPROVE。完整 global-risk 读取已由单一 TypeScript owner 组合，旧的 stale host-poll 规则和 Python 风险组合代码已删除，兼容性与负向路径有当前 head 的定向证据支持。merge 仍需等待 GitHub 必需检查通过，并针对未变化的精确 head 完成 readiness 检查。

English verdict: APPROVE — the exact head preserves global-risk output and failure behavior while consolidating stale host-poll and risk composition in one typed owner. Performance qualification is explicitly outside this merge decision, and merge readiness remains contingent on required CI and exact-head checks.

[loopx-agent]

Reviewer: loopx-agent; exact head: cd2e0f0a97e439ab6c8f76a6e4b6db1063c1a0f2

动机 / Motivation

操作者需要跨 Goal 查看过期运行、诊断和轮询风险。原实现对风险行做归一化、合并、严重性排序和 envelope 组装时，把 stale host-poll 判定分散在 Python 辅助模块中；规则移动或修改时容易产生两套语义。这个 PR 让一次 typed 读取拥有完整风险投影，同时保持操作者看到的 JSON、Markdown、排序、来源告警和错误边界。

改动思路 / Approach

Python 继续读取状态、注册信息、持久化收据和 Markdown，TypeScript 负责风险行推导、240 分钟边界判定、Agent 过滤、聚合、严重性排序、限制和 envelope 组合。旧的 Python stale host-poll helper 与其常量被删除，使规则只有一个生产 owner。公开本地路径模式由 Python 与 TypeScript 共用一致性测试约束，避免脱敏边界在迁移后漂移。

具体改动 / Changes

关键代码讲解

• projectGlobalRisks 组合风险行、Agent scope、聚合、排序、限制和 envelope；冻结的 37 个特征用例保持输出兼容。
• _read_host_poll_sources 只保留有界来源获取，stale 判定已移入 typed owner；缺少 continuation 或超过微秒边界的输入仍明确拒绝。
• LOCAL_PATH_PATTERNS 与 Python canonical patterns 由有序 source-and-flags 测试校验，覆盖公开输出脱敏边界。

验证

• 合并 7892 后的 TS risk、summary 与 handler 定向测试通过：25/25。
• 合并后的 Python 风险、host-poll、特征化、路径模式一致性、summary 和摘要特征化测试通过：161/161。
• npm run typecheck:control-plane 通过。
• check_rfc_bilingual.py --base origin/main --head HEAD 与 git diff --check 通过。
• rebase 后双语 RFC 的 summary/gates、quota 和 risk 条目均已保留；相邻 handler 自动合并且两类投影均已注册，复验通过。

对主干的风险 / Risks

没有发现新的阻断性功能问题。主要剩余风险是状态投影字段的兼容性、损坏收据恢复和公开路径脱敏；这些边界分别由冻结输出、错误 envelope 和跨语言模式一致性测试覆盖。性能资格按当前 owner 的决定不作为本次合并门禁，因此本 review 不把性能结论标记为已认证。发布时 CI 仍在运行；合并前会继续等待必需检查与最新 head 再读确认。

我的整体评价 / Verdict

APPROVE。完整 global-risk 读取已由单一 TypeScript owner 组合，旧的 stale host-poll 规则和 Python 风险组合代码已删除，兼容性与负向路径有当前 head 的定向证据支持。merge 仍需等待 GitHub 必需检查通过，并针对未变化的精确 head 完成 readiness 检查。

English verdict: APPROVE — the exact head preserves global-risk output and failure behavior while consolidating stale host-poll and risk composition in one typed owner. Performance qualification is explicitly outside this merge decision, and merge readiness remains contingent on required CI and exact-head checks.

[huangruiteng]

Reviewer: loopx-agent; exact head: cd2e0f0a97e439ab6c8f76a6e4b6db1063c1a0f2

动机 / Motivation

操作者需要跨 Goal 查看过期运行、诊断和轮询风险。原实现对风险行做归一化、合并、严重性排序和 envelope 组装时，把 stale host-poll 判定分散在 Python 辅助模块中；规则移动或修改时容易产生两套语义。这个 PR 让一次 typed 读取拥有完整风险投影，同时保持操作者看到的 JSON、Markdown、排序、来源告警和错误边界。

改动思路 / Approach

Python 继续读取状态、注册信息、持久化收据和 Markdown，TypeScript 负责风险行推导、240 分钟边界判定、Agent 过滤、聚合、严重性排序、限制和 envelope 组合。旧的 Python stale host-poll helper 与其常量被删除，使规则只有一个生产 owner。公开本地路径模式由 Python 与 TypeScript 共用一致性测试约束，避免脱敏边界在迁移后漂移。

具体改动 / Changes

关键代码讲解

• projectGlobalRisks 组合风险行、Agent scope、聚合、排序、限制和 envelope；冻结的 37 个特征用例保持输出兼容。
• _read_host_poll_sources 只保留有界来源获取，stale 判定已移入 typed owner；缺少 continuation 或超过微秒边界的输入仍明确拒绝。
• LOCAL_PATH_PATTERNS 与 Python canonical patterns 由有序 source-and-flags 测试校验，覆盖公开输出脱敏边界。

验证

• 合并 7892 后的 TS risk、summary 与 handler 定向测试通过：25/25。
• 合并后的 Python 风险、host-poll、特征化、路径模式一致性、summary 和摘要特征化测试通过：161/161。
• npm run typecheck:control-plane 通过。
• check_rfc_bilingual.py --base origin/main --head HEAD 与 git diff --check 通过。
• rebase 后双语 RFC 的 summary/gates、quota 和 risk 条目均已保留；相邻 handler 自动合并且两类投影均已注册，复验通过。

对主干的风险 / Risks

没有发现新的阻断性功能问题。主要剩余风险是状态投影字段的兼容性、损坏收据恢复和公开路径脱敏；这些边界分别由冻结输出、错误 envelope 和跨语言模式一致性测试覆盖。性能资格按当前 owner 的决定不作为本次合并门禁，因此本 review 不把性能结论标记为已认证。发布时 CI 仍在运行；合并前会继续等待必需检查与最新 head 再读确认。

我的整体评价 / Verdict

APPROVE。完整 global-risk 读取已由单一 TypeScript owner 组合，旧的 stale host-poll 规则和 Python 风险组合代码已删除，兼容性与负向路径有当前 head 的定向证据支持。merge 仍需等待 GitHub 必需检查通过，并针对未变化的精确 head 完成 readiness 检查。

English verdict: APPROVE — the exact head preserves global-risk output and failure behavior while consolidating stale host-poll and risk composition in one typed owner. Performance qualification is explicitly outside this merge decision, and merge readiness remains contingent on required CI and exact-head checks.