[huangruiteng]

Goal And Delivered Outcome

The controller RFC named broad C3/C4 lifecycle work without identifying which existing responsibilities can converge, what should be deleted, or which similar-looking components already share an owner. That made a contributor handoff prone to another adapter or a fixed cleanup agenda.

This proposal adds a pinned source audit and nine linked investigation/delivery packages, each with current callers, target ownership, specific retirement, decisive acceptance, dependencies and rollback. The RFC and overall roadmap explicitly require selection from current global outcomes and demonstrated bottlenecks: package IDs and row order are references, never global priority or permanent Agent assignments.

• Basis: controller/executor RFC C3/C4/C7/C8/C9 and the TypeScript migration change-locality contract at 27991e27ba998439d7faf2fc113efcfe82882144.
• Intended base: loopx-project/loopx:main.

Author Declaration

• Written by: model_agent; OpenAI Codex.

Implemented against

Specification: docs/architecture/rfcs/controller-executor-unified-contract-v0.md and docs/development/README.md#engineering-scope-and-refactoring at the source-audit revision above, refined by the maintainer's global-priority and code-simplification direction.

Criterion	Disposition	Location / evidence
Reuse existing owners and remove demonstrated duplication	implemented	Paired source audit distinguishes overlapping operation flows, misplaced shared handlers, existing reducers and non-equivalent lifecycles. No implementation or deletion is claimed.
Actionable contributor handoff across multiple slices	implemented	C3.1/C3.2, C4.1–C4.3, C7.1, C8.1, C9.1/C9.2 identify callers, retirement, acceptance and rollback. Conditional candidates retain comparison requirements.
Preserve global work selection and real dependencies	implemented	RFC and roadmap reject fixed package priority and permanent Agent lanes, require current outcome/bottleneck/PR/Todo reconciliation and permit unrelated higher-value work.
Preserve C1/C2, authority and original settlement boundaries	implemented	Existing composition/capture work, journal recovery, next-work decisions, wake identity, native capability differences and original effect readback remain distinct.
Bilingual and public-safe delivery	implemented	English/Chinese RFC, audit and roadmap mirrors; exact-file boundary scan and bilingual gates.

Self-check: read the affected source boundaries and current public PR scopes, reviewed both language mirrors and rejected false-positive consolidation candidates. No model run or runtime equivalence claim.

Scope And Continuation

Complete within this documentation/handoff scope. This is a bounded source audit, not an exhaustive repository audit, nine promised PRs, a new task store or a controller prerequisite queue. Implementation remains in existing C1–C9 and shared-authority/TypeScript acceptance. Claimants must refresh main, active PRs and canonical work; this document assigns no peer work.

The three responsibilities are existing Core semantic owners, shared managed execution and native Host adapters. A BaseHost hierarchy, new store, scheduler, per-field RPC proliferation or generic effect engine is not proposed. Measure cross-language production changes, actual deleted rules, complete-path crossings and the next representative rule's modification sites.

Validation

• Tested revision: e01486e97594c73093b549b0fe7134158adb5f6a.
• Run state: finished.
• Input classes: public_fixture, synthetic; no private execution artifacts.

Check kind	Result	Evidence / limitation
static	passed	Staged and exact-head bilingual RFC checks; generated RFC status index is current; diff hygiene clean.
static	passed	New/changed local link destinations and fragments passed. This is not whole-repository link certification.
static	passed	Exact outgoing-document boundary scan: 8 files, no blockers or unread inputs.
static	passed	examples/docs-governance-smoke.py on the committed proposal. Earlier checks caught the missing execution-ledger appendix and derived status index; both are now included.
integration	passed	Standard risk-based loopx canary premerge --from-git-diff --git-diff-base origin/main: four direct checks and all 19 selected checks passed, zero failures. An earlier run rejected a concurrent tracked-document refinement and is not used as qualification.
real_entrypoint / real_backend	not_run	No production code changes. Host/provider suites in the audit are implementation acceptance, not runtime equivalence evidence from this PR.

A broad loopx check also returned unrelated registered-Goal source-routing errors; its outgoing-document scan was clean. The explicit exact-file scan above qualifies only these publication candidates, not overall installation health. Required CI and independent review remain separate.

Frontend / Visual Evidence

UI impact: none. This edits architecture text and a source-audit ledger, without documentation chrome, runtime behavior, schema, settings or product UI changes. Screenshots and packaged frontend tests are not applicable here; caller-facing implementation packages retain their existing frontend failure/recovery and visual-evidence obligations.

Type of Change / LoopX Area

• Documentation update
• Public docs
• Control-plane architecture planning

Shared-authority fixture impact: none. No authority transaction, provider routing, state format, scoring, installation or default is changed. The real backend/Host suites listed in the handoff are future implementation acceptance, not results from this PR.

Boundary Checklist

• Eight public documentation files only; no private Goal state, raw logs, credentials, internal links or host paths.
• Existing RFC language pairs and new audit language pair changed together.
• Commit includes DCO sign-off.
• No merge, model execution or benchmark launch authority is inferred.