diff --git a/docs/architecture/rfcs/README.md b/docs/architecture/rfcs/README.md
index a773baa4ec..08168db90f 100644
--- a/docs/architecture/rfcs/README.md
+++ b/docs/architecture/rfcs/README.md
@@ -273,6 +273,14 @@ Paired diffs do not certify translation accuracy.
 
 ## Planning, Research, And Adaptive Intelligence
 
+- [LoopX × LayerFS — Engineering Exploration and Evaluation v0](layerfs-engineering-exploration-evaluation-v0.md)
+  ([中文版](layerfs-engineering-exploration-evaluation-v0.zh-CN.md))
+  - **Delivery on `main`:** Proposal only; no provider integration or measured benefit.
+  - **Current boundary:** Optional execution-workspace provider under existing
+    Explore/Host owners; real-environment branch/handoff conformance, then warm
+    worktree comparisons and native/passive/governed outcome/cost attribution.
+    Filesystem checkpoints do not qualify complete research-state continuation.
+
 - [Goal-scoped Capability Portfolio and Connector Lifecycle v0](goal-scoped-capability-portfolio-v0.md)
   ([中文版](goal-scoped-capability-portfolio-v0.zh-CN.md))
   - **Delivery on `main`:** Proposal; catalog, Agent-context hooks and connector
diff --git a/docs/architecture/rfcs/STATUS.md b/docs/architecture/rfcs/STATUS.md
index 36b77b1e59..6d2656d887 100644
--- a/docs/architecture/rfcs/STATUS.md
+++ b/docs/architecture/rfcs/STATUS.md
@@ -18,7 +18,7 @@ appendix may keep dated history, but no dated log heading may precede it.
 
 [中文版](STATUS.zh-CN.md) is the semantic mirror of this file.
 
-## Accepted (45)
+## Accepted (46)
 
 | RFC | Header status | Supersedes / closes | Ledger |
 | --- | --- | --- | --- |
@@ -49,6 +49,7 @@ appendix may keep dated history, but no dated log heading may precede it.
 | [RFC: Human Attention Wishlist v0](human-attention-wishlist-v0.md) | Accepted | none | — |
 | [RFC: Human-confirmed domain operations (v0)](human-confirmed-domain-operations-v0.md) | Accepted | none | — |
 | [RFC: Intelligent Review and Dynamic Presentation Surfaces v0](intelligent-review-presentation-surfaces-v0.md) | Accepted | none | — |
+| [RFC: LoopX × LayerFS — Engineering Exploration and Evaluation (v0)](layerfs-engineering-exploration-evaluation-v0.md) | Accepted | none | — |
 | [RFC: Live Team Workspace v0](live-team-workspace-v0.md) | Accepted | none | — |
 | [RFC: Long-Horizon Harness Benchmark and Research Program v0](long-horizon-harness-benchmark-research-program-v0.md) | Accepted | none | — |
 | [RFC: Long-Running Agent Reliability Diagnostics and Governed Delivery v0](long-running-agent-reliability-diagnostics-governed-delivery-v0.md) | Accepted | none | — |
diff --git a/docs/architecture/rfcs/STATUS.zh-CN.md b/docs/architecture/rfcs/STATUS.zh-CN.md
index 2891613ba8..f8791b87d4 100644
--- a/docs/architecture/rfcs/STATUS.zh-CN.md
+++ b/docs/architecture/rfcs/STATUS.zh-CN.md
@@ -15,7 +15,7 @@
 
 [English](STATUS.md) 与本文互为语义镜像。
 
-## 已接受 (45)
+## 已接受 (46)
 
 | RFC | 头部状态 | 替代 / 关闭 | Ledger |
 | --- | --- | --- | --- |
@@ -46,6 +46,7 @@
 | [RFC：Human Attention Wishlist v0](human-attention-wishlist-v0.zh-CN.md) | 已接受 | 无 | — |
 | [RFC：用户确认的垂域操作（v0）](human-confirmed-domain-operations-v0.zh-CN.md) | 已接受 | 无 | — |
 | [RFC：智能化审阅与动态展示面 v0](intelligent-review-presentation-surfaces-v0.zh-CN.md) | 已接受 | 无 | — |
+| [RFC：LoopX × LayerFS——复杂工程探索与效果评测（v0）](layerfs-engineering-exploration-evaluation-v0.zh-CN.md) | 已接受 | 无 | — |
 | [RFC：团队实时工作区 v0](live-team-workspace-v0.zh-CN.md) | 已接受 | 无 | — |
 | [RFC：长程 Harness Benchmark 与研究计划 v0](long-horizon-harness-benchmark-research-program-v0.zh-CN.md) | 已接受 | 无 | — |
 | [RFC：长程 Agent 可靠性诊断与治理交付 v0](long-running-agent-reliability-diagnostics-governed-delivery-v0.zh-CN.md) | 已接受 | 无 | — |
diff --git a/docs/architecture/rfcs/layerfs-engineering-exploration-evaluation-v0.md b/docs/architecture/rfcs/layerfs-engineering-exploration-evaluation-v0.md
new file mode 100644
index 0000000000..368abb17a3
--- /dev/null
+++ b/docs/architecture/rfcs/layerfs-engineering-exploration-evaluation-v0.md
@@ -0,0 +1,416 @@
+# RFC: LoopX × LayerFS — Engineering Exploration and Evaluation (v0)
+
+- **RFC status:** Accepted (effective on merge; this change is a proposal)
+- **Supersedes / closes:** none
+- **Delivery maturity:** Proposal; no integration, experiment result or product promotion is claimed.
+- **Authors / owners:** LoopX exploration and benchmark maintainers; LayerFS maintainers are invited to review the provider boundary, not assigned work.
+- **Created:** 2026-10-26
+- **Last normative revision:** 2026-10-26
+- **Implementation baseline:** LoopX `15803cd7dcc2fc517049809bbe58998f04f5acac`; LayerFS source `ad78914e36b258b360daf13aec164b5655814689`.
+- **Tracking:** [LoopX #7406](https://github.com/loopx-project/loopx/issues/7406) · [LayerFS #461](https://github.com/Ephemeral-AI-Lab/layerfs/issues/461).
+- **Language mirror:** [中文版](layerfs-engineering-exploration-evaluation-v0.zh-CN.md)
+- **Related contracts:** [Explore](research-exploration-control-plane-v0.md), [research-state branching](hierarchical-agent-stride-control-v0.md#77-research-state-branching-and-matched-continuation), [measurement](long-horizon-harness-benchmark-research-program-v0.md#6-measurement-model), [overall roadmap](loopx-overall-roadmap-v0.md).
+
+## Document map and maintenance contract
+
+Sections 1–3 and 5–11 define the proposed experiment and integration boundaries;
+section 4 records source observations, not live qualification. Section 12 lists
+decisions needed before execution. Appendix A records provenance. Keep the
+Chinese mirror synchronized. Merge accepts a design basis, not a runtime
+default, benchmark result, spending commitment or upstream implementation promise.
+Future measured checkpoints belong in the existing [RFC ledger](ledger/README.md).
+
+## 1. Decision summary
+
+Evaluate LayerFS as an **optional execution-workspace provider for multi-agent
+engineering exploration**. LoopX keeps questions, hypotheses, evidence, work
+ownership, budget and continuation; LayerFS keeps branchable filesystem state;
+the host keeps model sessions and tool execution. Git remains the reviewed
+source-delivery boundary. No new generic capability or scheduler is proposed.
+
+The first experiment asks whether several agents can investigate a hard repair,
+reuse a prepared environment, reject an approach, hand off a promising checkpoint
+and deliver an independently verified patch with less total time or cost.
+It does not equate more branches, cheaper mounts or more checkpoints with progress.
+
+Start with one host and three workers, using the same model and aggregate budget
+as a warm Git-worktree control. Qualify workspace correctness before model runs.
+Keep native/passive/governed controls for attributing joint benefit, and report
+an absence of benefit as a useful result. Integration remains default-off.
+
+## 2. Problem and hypotheses
+
+In a multi-package repair, alternative explanations often require different
+dependency versions, generated code, build outputs and test state. Git preserves
+tracked source well, while environment reconstruction and transferring useful
+untracked state can dominate short branches. Copying complete environments can
+consume disk; sharing writable caches can contaminate comparisons. Conversely,
+long-lived warm worktrees already avoid much repeated setup.
+
+| Hypothesis | Observable evidence | Falsifier or limit |
+| --- | --- | --- |
+| H1: branching a prepared full environment reduces exploration overhead | Lower init-amortized fork→test-ready, capture and handoff cost at equal restored-state fidelity | Mount is fast but commit, I/O, cleanup or cold import cancels the saving |
+| H2: structured evidence plus cheap environment branches improves engineering outcomes | More independently accepted repairs at fixed total budget, or lower complete cost at comparable accepted outcomes | More attempts or attractive history with unchanged/worse outcomes |
+| H3: a receiver can reuse a useful candidate without repeating setup or investigation | Receiver verifies the exact candidate, consumes a relevant finding and performs the next accepted work | A read receipt, filesystem fork or copied summary without adopted work |
+
+The likely fit is frequent branching with expensive, reusable environment state.
+Read-only code search, a few persistent checkouts, large sequential I/O, or
+workloads dominated entirely by model thinking may gain little or regress.
+Filesystem deduplication does not deduplicate model reasoning or context tokens.
+
+## 3. Scope and non-goals
+
+The slice covers public engineering fixtures, isolated disposable workspaces,
+candidate/result lineage, one end-to-end demo and a reproducible comparative
+study. It maps to roadmap **S4/S8/S11**, with existing S3/S7 authority and budget
+constraints; it is not a replacement for R2/R3 team qualification.
+
+Excluded: a replacement Goal store, global memory, a new agent framework,
+production durability, untrusted multi-tenant isolation, cross-machine shared
+Store writes, automatic Git publication, full process/VM snapshots, autonomous
+MCTS policy, and promotion of LayerFS 2.0 roadmap items as current support.
+
+## 4. Audited current-system contract
+
+### LoopX
+
+- [`explore`](../../../loopx/capabilities/explore/README.md) already owns
+  node-linked findings, candidate/input identities, bounded measurement feedback
+  and execution observations. A measurement or observation is producer evidence,
+  not proof of benefit. Reuse its TypeScript decision owners.
+- [Stride §7.7](hierarchical-agent-stride-control-v0.md#77-research-state-branching-and-matched-continuation)
+  defines a complete research point across conversation, evidence, artifacts and
+  evaluation opportunities. The existing [capture adapter](../../../benchmark/runtime/checkpoint.py)
+  and [artifact owner](../../../loopx/control_plane/runtime/checkpoint_artifacts.ts)
+  do not already accept arbitrary LayerFS references. The current file-capture
+  path rejects symlinks; a pnpm tree cannot simply be passed through that path.
+- The [DSH plugin](../../../packages/dsh-loopx-plugin/README.md) supplies native
+  binding, continuation and Goal readback. It is a possible composition host,
+  not an existing LayerFS integration.
+
+### LayerFS
+
+The following facts refer to the pinned source above; release 1.1.0 and moving
+web documentation are not interchangeable with that commit. Source evidence is
+listed in Appendix A.
+
+| Mechanism | Integration consequence |
+| --- | --- |
+| SQLite content-addressed Store; fork shares committed objects; Workspace mounts a writable view | Candidate branches can retain source, dependencies and generated outputs without full payload copies; initialization and later writes still cost work |
+| `@layerfs/agent`, `@layerfs/backend`, `@layerfs/runtime` split | Use public APIs at one exact source version; do not build against `@layerfs/agent/internal` |
+| File mutations commit; `workspace.exec`, `workspace.commit`, `workspace.unmount` are separate calls | The integration must implement the per-tool lifecycle; a Bash command is not automatically a durable checkpoint |
+| Direct Store open requires one Linux kernel and local ext4/xfs/btrfs; macOS/Windows have a worker route | First qualify Linux or one Linux VM; macOS worker setup must be tested explicitly, including upstream [#456](https://github.com/Ephemeral-AI-Lab/layerfs/issues/456) |
+| Disposable profile, WAL with synchronous OFF; no crash/power-loss durability claim | Keep authoritative LoopX state and an independent source/result copy outside the Store; do not advertise backup or compare fsync speed as equal durability |
+| Shared publication lane; conflicts, `.lww`, bounded diagnostics and explicit unknown outcomes | Measure contention and refusals; protect source, lockfiles and evaluator inputs from last-writer-wins; never turn a lost reply into an automatic retry |
+| Append-only history; no branch retirement/unreachable-history GC ([#411](https://github.com/Ephemeral-AI-Lab/layerfs/issues/411)) | Discarding a Workspace does not reclaim committed branch history; account for all failed candidates and bound each experiment's Store lifetime |
+| Commands can write outside the Workspace; process memory, network effects and external services are not captured | Use a declared execution envelope and disposable external fixtures; a filesystem rewind is not an experiment rewind |
+
+The published v1.0.0 [dedup study](https://layerfs.ai/benchmarks/dsh-dedup/)
+reports 157 historical trees retained in about 100 MB versus about 1,995 MB for
+Btrfs+zstd under its rewrite workload. Its [fio study](https://layerfs.ai/benchmarks/fio/)
+also reports sequential-write/read and random-write regressions. These are
+upstream measurements at `e5c7aaae05`, not reproduced here, not 1.1.0 results,
+and not evidence of engineering-task success or lower total agent cost.
+
+## 5. Proposed architecture and demo
+
+### Ownership and placement
+
+| Owner | Responsibility |
+| --- | --- |
+| LoopX existing Explore / Todo / quota owners | Hypothesis and finding lineage, claims/leases, admitted work, result adoption, remaining budget and replan |
+| Host/runtime | Native sessions, model/tool routing, cancellation, settled capture boundary and execution isolation |
+| Optional LayerFS adapter | Map a candidate to exact Store/project/Commit/root and image identities; fork, mount, inspect, commit and return evidence |
+| Existing benchmark toolkit / task evaluator | Frozen workload, independent oracle, complete cost collection and comparative analysis |
+| Existing Git/review workflow | Export the selected source patch, validate in a clean checkout and propose it for review |
+
+Placement rationale: `capability_id: explore`, current built-in provider
+`loopx-core`; LayerFS is a candidate optional provider implementation, initially
+experiment-owned under the benchmark workspace. A production distribution would
+follow the [extension placement rules](../../reference/extensions.md#placement-decision-for-agents)
+only after a real caller qualifies. Do not register a provider id, new shared
+enum or package solely for this design. New provider-neutral decisions belong
+in the existing TypeScript owner; specialized capture remains an adapter.
+
+Keep two graphs: the Explore graph explains questions and findings; the LayerFS
+graph identifies filesystem ancestry. They are linked, not one-to-one. Several
+tool commits can implement one experiment, and one filesystem checkpoint can
+support several findings. A LayerFS Commit is not a Git commit or an accepted
+LoopX outcome.
+
+The experiment manifest binds existing task/experiment/result ids to provider
+source revision, Store/project identity, parent and result Commit/root, runtime
+image digest, environment manifest, evaluator revision and artifact digests.
+Mutable branch names are display handles, never sufficient evidence identities.
+Keep raw manifests and traces outside model packets; attach bounded summaries
+and opaque evidence refs using existing Explore fields. This is a local study
+record, not a silently introduced shared runtime schema.
+
+### A complete small demo
+
+Use the public failure in [LoopX #6869](https://github.com/loopx-project/loopx/issues/6869)
+as the first **candidate fixture**, starting at source
+`907ede6e1faff1fab0f2f765dde4358c69c7b7ff`: an unreadable spend artifact can
+silently reduce accounted spend. Source inspection confirms the missing-event
+path; the first milestone must reproduce the real CLI failure before admitting
+this fixture. [PR #6875](https://github.com/loopx-project/loopx/pull/6875) is existing
+repair work, not an accepted oracle or work to duplicate in this proposal.
+The demo is an isolated historical repair exercise, not a claim on that issue.
+
+1. Prepare a standalone clone at the pinned base, with locked dependencies,
+   built TypeScript runtime and synthetic File/SQLite task data. The solver's
+   source repository is distinct from the installed LoopX controller and its
+   runtime state. Remove credentials and host-local data before import. A Git
+   worktree's `.git` pointer is not a portable complete repository.
+2. Give three workers the same task and declared tool/feedback envelope. They
+   choose and record hypotheses, may share findings and may reject or continue
+   approaches. Do not script their conclusions or force one branch to win.
+3. Start branches from an exact prepared checkpoint. A deterministic conformance
+   run first exercises competing edits, a negative result and a receiver handoff;
+   a separate model run determines whether those mechanisms are actually used.
+4. Record each probe's code/environment identity, test result and cost. The
+   independent evaluator checks that missing, malformed and truncated spend
+   evidence never silently increases authorization; valid spend/void behavior,
+   history and admission stay consistent, including at time-window boundaries.
+   Do not use blanket permanent throttling or one patch's expected output as
+   the correctness oracle.
+5. A receiver forks a selected immutable candidate, verifies its manifest,
+   reads relevant findings and performs the next test or repair. Then export
+   only the selected source patch into a fresh checkout and rerun the evaluator
+   with clean dependencies. An apparently passing cache or branch is insufficient.
+6. Present the resulting patch, failed alternatives, receiver's actual work,
+   held-out result and full cost breakdown. A recorded UI replay is labeled as
+   such; scripted conformance is not autonomous model adoption.
+
+This public historical task may be contaminated by prior model exposure. It
+qualifies the demo path only. The study must add previously unseen, independently
+prepared tasks in at least a TypeScript monorepo and a Rust workspace, including
+dependency/generated-code changes and a low-branching negative-control stratum.
+Fixtures retain real package graphs, lockfiles, symlinks, ignored dependencies,
+build outputs and representative error logs. Artificial mutations are labeled;
+repeated strings or empty files cannot substitute for those state shapes.
+
+### Execution and failure transitions
+
+For the first host, investigate co-installing the existing DSH plugins in a
+disposable profile; pin DSH, both plugins and LayerFS together. Existing
+co-installability, tool routing and session inheritance are **unverified**.
+DSH availability does not authorize silently changing the study's host.
+
+Before each filesystem command, the existing host admits the work and the
+adapter binds its candidate. Direct file calls use the SDK; shell calls use
+mount → exec → explicit commit/readback → clean unmount. Inspect nonzero exits
+as experiment evidence, not successful fixes. Do not accidentally discard
+useful failed-test state; an explicit discard is recorded when chosen. Same
+Workspace operations remain sequential; workers own separate Workspaces.
+
+After a conflict, preserve the original inputs and inspect every affected path,
+including paginated/truncated diagnostics. Resolve on new working forks and
+reevaluate. For merges, freeze input heads first; a passing source branch's test
+does not cover a newly merged root. On a lost response, reconcile exact history
+and Workspace status before any new operation. If success cannot be established,
+keep the attempt unknown and stop that branch; no blind command replay or forced
+commit. Cleanup never claims an unresolved operation succeeded.
+
+For filesystem-only forks, start fresh sessions with the same declared context
+in both arms. A later same-conversation continuation experiment must satisfy
+Stride §7.7's host cursor, evidence, artifacts and opportunity barrier. LayerFS
+can replace only the artifact participant after its provider contract qualifies;
+it cannot make the current file-only capture accept symlinks or restore leases,
+budgets, evaluator attempts or external side effects by copying their files.
+
+## 6. Alternatives and comparisons
+
+| Option | Why retain it |
+| --- | --- |
+| Warm Git worktrees + locked dependency cache | Primary practical control; allow realistic caching and preserve needed untracked outputs in its checkpoint path |
+| Btrfs/reflink or overlay snapshot on a suitable Linux host | Strong alternative for cheap full environments; compare when available, disclose unavailable coverage |
+| LayerFS workspace provider | Test shared complete-tree history, incremental capture and branch handoff, including FUSE and publication costs |
+| Ordinary persistent single-worker checkout | Negative control when branching is rare; establishes whether parallel coordination itself helps |
+
+Do not compare a prewarmed LayerFS tree only against cold package downloads.
+Measure both cold setup and repeated warm reuse, each with equal reachable
+state and retention requirements. Keep dependency caches immutable or isolated
+per branch. Native snapshots can remain the better choice.
+
+Start each attempt's wall clock before its environment preparation and include
+terminal validation/export. Report one-time provider build/image preparation
+separately and in a declared amortized total; do not hide it or charge it only
+to the control. The steady-state view supplements the cold total, not replaces it.
+
+## 7. Safety, privacy and compatibility
+
+Run on disposable public fixtures in one Linux kernel, with fixed CPU/RAM/disk
+limits and separately controlled network access. Do not import a developer's
+working directory: LayerFS intentionally captures ignored files too. Keep
+credentials, model histories, live LoopX state, evaluator internals, daemon
+endpoints and Docker access outside agent-controlled trees. Use a fixed empty
+`.lww` policy for the first study; no automatic newer-wins for lockfiles, source
+or tests. A policy change makes a different experimental treatment.
+
+Check executable modes, symlinks, hardlinks, absolute paths, native dependency
+ABI and packages writing outside the root. Imported environments must be built
+for the same guest OS/architecture/image; an arm64 macOS dependency tree is not
+a Linux environment. LayerFS isolation is not evidence of tenant security.
+
+Feature-off must keep existing execution and state unchanged. Adapter failure
+is visible; a fallback starts a separately labeled attempt rather than silently
+changing an arm. No model-visible instruction set can substitute for verifying
+where tools actually read and write. LoopX control calls stay on their owning
+host, outside the candidate tree being repaired.
+
+## 8. Migration and rollback
+
+No existing Store or Goal migration is proposed. Initialize a new study Store
+from an independently preserved fixture. Keep selected source patches and
+verification evidence outside it; test clean-checkout replay before declaring
+success. Stop new admissions, reconcile in-flight operations, export qualified
+results and close owned Workspaces/Sandboxes before removing the study's Store.
+Never describe Workspace discard as historical GC. Disable the optional adapter
+to return to the original workspace path. A source-format upgrade starts a new
+study generation and is not an in-place repair of old evidence.
+
+## 9. Evaluation and acceptance
+
+### Separate substrate effects from agent effects
+
+First replay the same predetermined filesystem/build/test operations on both
+workspace providers, with 1/3/8 workers and branch counts 1/4/16 where resource
+limits allow. Record excluded cells before running. Keep topology, retained
+states and command semantics equal. This estimates substrate overhead only.
+
+Then use this controlled matrix, with the same host/model/reasoning settings,
+task split, test visibility, aggregate resource budgets and three-worker ceiling:
+
+| Control policy | Warm worktrees (W) | LayerFS (F) |
+| --- | --- | --- |
+| Native host (N) | N-W | N-F |
+| Passive LoopX (P) | P-W | P-F |
+| Governed LoopX (G) | G-W | G-F |
+
+The first bounded model pilot may run **G-W vs G-F**; it estimates workspace
+effect under that LoopX policy, not LoopX's benefit or joint synergy. The full
+matrix separates observation overhead, governance and workspace effects. Claim
+an interaction only from a predeclared difference-of-differences on a common
+outcome scale, with uncertainty; superiority of G-F alone is insufficient.
+
+Freeze task ids/base commits, mutation generator and evaluator hashes, model
+revision, SDK/image/plugin pins, branch/depth limits, feedback schedule, dollar/
+token/compute/wall ceilings and missing-data rules before model evaluation.
+Charge coordinator and all workers to **one aggregate budget**. Model-selected
+branching may differ in the agent study but stays within that ceiling. Never
+compensate one provider's overhead with extra time, worker slots or tokens.
+
+Use paired tasks/seeds, interleaved arm order and at least five repetitions per
+promoted comparison cell, following the existing research contract; five is not
+a power justification. Select tasks and any sequential stop rule before seeing
+comparative results. Reserve held-out tasks/evaluation from all solver-accessible
+files and network routes. Treat development feedback, scheduled evaluation and
+terminal grading as distinct opportunities. Retain setup failures, cancellation,
+timeouts, missing treatment and invalid evaluation in the denominator with their
+own classifications. Freeze a power/precision plan before making general claims.
+
+### Metrics
+
+Primary: independently accepted repair rate at fixed aggregate budget and
+best valid retained result over time. For equivalent accepted outcomes, report
+time and complete cost, with all attempts included. Estimate any break-even
+branch count from measured cold setup and marginal reuse costs, not fork latency
+alone. Secondary metrics explain the result:
+
+- setup, dependency acquisition, mount/fork, tool execution, capture/commit,
+  conflict repair, handoff, validation, export and cleanup wall/CPU time;
+- uncached/cached input and output tokens, actual billed model cost, control and
+  evaluator work, human interventions and measured attention time;
+- physical allocated Store plus WAL, overlays, container scratch, caches,
+  independent exports and peak disk/RSS; logical history bytes separately;
+- contention, refused/unknown operations, recovery work, duplicated probes and
+  evidence→decision→executed receiver work, without rewarding branch counts.
+
+Use LayerFS traces/backend counters as diagnostics, plus an independent runner
+clock, OS resource measurements and provider usage. Trace writes can be lost;
+missing trace data is not zero cost. Follow the existing measurement contract:
+packet bytes are not model tokens, dedup ratios are not total disk saved, and
+unknown billing/attention prevents a complete-cost claim. Do not sum overlapping
+wall intervals as elapsed duration; also report total worker compute time.
+
+### Acceptance rows
+
+| Claim | Required evidence | Pass boundary |
+| --- | --- | --- |
+| A1: real environment branch fidelity | Deterministic tree/content/mode/link checks and same tests after fresh remount, including ignored outputs and a symlink-rich package tree | Every selected checkpoint has matching declared state; metadata differences are explicitly normalized, not ignored wholesale |
+| A2: isolation and conflict integrity | Parallel independent edits, same-file conflict, merge-root retest, external-path sentinel and no-op/read-only calls | No candidate contamination or silent overwrite; no fabricated Commit for unchanged state |
+| A3: truthful failure/recovery | Nonzero tool exit, lost reply, interrupted Workspace, absent worker, disk cap and missing evidence | Unknown remains unknown; no duplicate execution, false success or automatic authority restore |
+| A4: useful handoff and output | Receiver verifies candidate, uses a scoped finding, performs new work; exported patch passes independent clean-checkout evaluation | Demonstrates the complete scripted path; a separate model run must prove actual adoption |
+| A5: causal efficiency evidence | Frozen matched cells, complete attempt accounting, paired uncertainty and native outcomes | Claim only the scope supported by the contrast; no promotion from a demo or microbenchmark |
+| A6: usable optional journey | Packaged host discovers availability, performs authorized run, shows conflict/unavailable states, reads results and disables provider | Required before product promotion; CLI-only experiment is labeled partial |
+
+Before the confirmatory study, preregister the primary contrast, minimum useful
+effect and acceptable outcome-loss margin using pilot variance and operator
+costs. These values cannot be chosen after inspecting the confirmatory result.
+If the confidence interval spans material harm and benefit, keep the conclusion
+inconclusive. Correctness/isolation failures block qualification regardless of
+speed. Benefit restricted to storage is reported as storage benefit.
+
+## 10. Operational and product contract
+
+Use the existing Goal/Explore research history and DSH conversation for task
+and outcome readback. LayerFS Work/History may show the filesystem lineage;
+it cannot replace the explanation of why a candidate was kept. No new top-level
+screen is required for the experiment. A shipped provider still needs existing
+settings/capability readiness, authorized launch, unavailable/error recovery and
+disable/readback coverage in App/CLI and any advertised Lark entrypoint.
+
+Cap branch count, total Store growth, concurrent workers and command duration
+at the experiment runner. Forecast retention from failed as well as successful
+branches; cleanup only this study's resources. Show stale/unknown checkpoints
+and unavailable observations. Do not feed raw logs or growing history into
+every model turn; detail is fetched by immutable reference and measured as
+actual consumed context. This RFC changes no runtime packet or compute quota.
+
+## 11. Normative delivery plan
+
+| Milestone | Entry | Exit evidence | Rollback / owner |
+| --- | --- | --- | --- |
+| M0: freeze the demo | This proposal and public fixture candidate | Reproduced base failure; reviewed independent oracle; pinned host/provider manifest and explicit bounded run budget | No live state change; LoopX experiment owner |
+| M1: one reliable integration path | M0; isolated runtime; compatible SDK/native/image | A1–A4 deterministic conformance through real LayerFS, LoopX readback and clean exported patch; disabled path parity | Remove disposable profile/provider; LoopX adapter owner, upstream consulted on API gaps |
+| M2: comparative demo and pilot | M1 | G-W/G-F paired pilot, failed alternatives, full cost coverage and an honest effect/limit report; preregistration for any larger study | Retain evidence; no default change; benchmark owner |
+| M3: attribution and optional adoption decision | Adequate pilot evidence and predeclared study | Native/passive/governed matrix, held-out task strata, A5 and independently repeated results; A6 before any product claim | Keep provider experimental or reject it; existing maintainers decide |
+
+One LoopX tracking issue owns this sequence; one LayerFS issue requests API and
+evaluation feedback. Reuse upstream #411/#456 and roadmap
+[#446](https://github.com/Ephemeral-AI-Lab/layerfs/issues/446) rather than filing
+duplicate GC, worker or generic worktree-benchmark tasks. Portable substrate
+measurements may contribute to `layerfs-benchmark`; LoopX owns governance and
+task-outcome experiments. Upstream assent is useful but does not block local
+fixture/oracle preparation with currently available interfaces.
+
+## 12. Open decisions
+
+| Decision / owner | Recommendation | Evidence required / due |
+| --- | --- | --- |
+| Host profile / LoopX host maintainer | DSH composition first; retain runtime-neutral design | Verify plugin versions, control-call routing, worker inheritance, cancellation and no double commit before M1 |
+| Fixture and budget / experiment owner | Public #6869 demo; held-out TS/Rust task strata for effectiveness | Reproduce base, license/privacy review, oracle and cost caps before M0 exits |
+| Provider API / LayerFS maintainer feedback | Public 1.1-era packages at one exact commit; no internal API or durability changes | Confirm worker settings, exact-root comparison, ambiguous-operation readback and cleanup scope before claiming provider support |
+| Promotion / existing LoopX maintainers | Remain optional until replicated value and A6 | Predeclared practical-effect thresholds, power/precision and actual journey evidence before M3 closes |
+
+## Appendix A: Evidence registry
+
+Inspected 2026-10-26. Source inspection is not a reproduced benchmark or a live
+integration. No LayerFS build, model experiment or runtime installation was run
+for this RFC; a local Docker availability check only informs feasibility.
+
+| Evidence | Finding / limitation |
+| --- | --- |
+| [LayerFS pinned README](https://github.com/Ephemeral-AI-Lab/layerfs/blob/ad78914e36b258b360daf13aec164b5655814689/README.md) | Released-source/API split, Disposable restriction and explicitly scoped v1.0 measurements |
+| [Agent SDK guide](https://github.com/Ephemeral-AI-Lab/layerfs/blob/ad78914e36b258b360daf13aec164b5655814689/api/how-to-use/agent.md) and [exec implementation](https://github.com/Ephemeral-AI-Lab/layerfs/blob/ad78914e36b258b360daf13aec164b5655814689/api/agent/sdk/src/workspace/exec.js) | Public calls, explicit commit lifecycle, guest settings, streamed output and bounded read/merge diagnostics |
+| [Runtime guide](https://github.com/Ephemeral-AI-Lab/layerfs/blob/ad78914e36b258b360daf13aec164b5655814689/api/how-to-use/runtime.md) | Host-owned initialization, macOS/Windows worker and resource cleanup; no proof of plugin coexistence |
+| [Backend/trace guide](https://github.com/Ephemeral-AI-Lab/layerfs/blob/ad78914e36b258b360daf13aec164b5655814689/api/how-to-use/backend.md) | Root comparison limitations, storage accounting and lossy diagnostic trace |
+| [Store architecture](https://github.com/Ephemeral-AI-Lab/layerfs/blob/ad78914e36b258b360daf13aec164b5655814689/core/docs/architecture/84-global-sqlite-store.md) | Local-kernel boundary, publication semantics and no unreachable-history GC |
+| [DSH integration](https://github.com/Ephemeral-AI-Lab/dsh-layerfs/blob/baac1b92b97f0bd492002c285f6a13401b4f4c0b/README.md) | Existing Bash/SDK route and Work/History UI; its LayerFS pin must be reconciled with the proposed source pin before execution |
+| [LoopX historical readback](https://github.com/loopx-project/loopx/blob/907ede6e1faff1fab0f2f765dde4358c69c7b7ff/loopx/control_plane/quota/ledger_readback.py) | Missing/unreadable events yield no contribution; real-entry reproduction and correct repair oracle remain M0 work |
+
+The website's broad shared-storage wording does not supersede the supported
+single-kernel Store contract. Older generic concept pages and pre-1.1 SDK names
+are not implementation guidance for this study. Open issue reports are useful
+risks, not independently reproduced defects or promised fixes.
diff --git a/docs/architecture/rfcs/layerfs-engineering-exploration-evaluation-v0.zh-CN.md b/docs/architecture/rfcs/layerfs-engineering-exploration-evaluation-v0.zh-CN.md
new file mode 100644
index 0000000000..f0b9bf0a14
--- /dev/null
+++ b/docs/architecture/rfcs/layerfs-engineering-exploration-evaluation-v0.zh-CN.md
@@ -0,0 +1,352 @@
+# RFC：LoopX × LayerFS——复杂工程探索与效果评测（v0）
+
+- **RFC status:** Accepted (合并后生效；本变更仍为提案)
+- **替代 / 关闭：** 无
+- **Delivery maturity:** Proposal；尚无集成实现、实验结果或产品推广结论。
+- **Authors / owners:** LoopX 探索与 benchmark 维护者；邀请 LayerFS 维护者评审 provider 边界，不替对方分配工作。
+- **Created:** 2026-10-26
+- **Last normative revision:** 2026-10-26
+- **Implementation baseline:** LoopX `15803cd7dcc2fc517049809bbe58998f04f5acac`；LayerFS 源码 `ad78914e36b258b360daf13aec164b5655814689`。
+- **Tracking:** [LoopX #7406](https://github.com/loopx-project/loopx/issues/7406) · [LayerFS #461](https://github.com/Ephemeral-AI-Lab/layerfs/issues/461)。
+- **Language mirror:** [English](layerfs-engineering-exploration-evaluation-v0.md)
+- **Related contracts:** [Explore](research-exploration-control-plane-v0.zh-CN.md)、[研究状态分叉](hierarchical-agent-stride-control-v0.zh-CN.md)、[度量契约](long-horizon-harness-benchmark-research-program-v0.zh-CN.md)、[整体路线](loopx-overall-roadmap-v0.zh-CN.md)。
+
+## 文档地图与维护契约
+
+第 1–3、5–11 节规定拟议实验与集成边界；第 4 节记录源码观察，不代表真实运行资格。
+第 12 节列出执行前待定事项。附录 A 保存来源。中英文保持语义同步。
+合并接受的是设计基础，不是运行时默认值、benchmark 结果、支出承诺或上游实施承诺。
+后续实测 checkpoint 使用现有 [RFC ledger](ledger/README.zh-CN.md)。
+
+## 1. 决策摘要
+
+将 LayerFS 作为**多 Agent 工程探索的可选执行工作区 provider**进行评估。
+LoopX 保持问题、假设、证据、工作归属、预算与接续的权威；LayerFS 保存可分叉的
+文件系统状态；Host 保持模型会话与工具执行；Git 保持源码评审交付边界。
+本提案不新增通用 capability 或 scheduler。
+
+第一个实验检验：多个 Agent 能否在复杂修复中复用已准备好的环境，排除一条路线，
+把有潜力的 checkpoint 交给接手者，并以更低总时间或总成本交付独立验证的补丁。
+更多分支、更快 mount、更多 checkpoint 本身不算进展。
+
+从单 Host、三个 worker 开始，与预热 Git worktree 使用相同模型和总预算。
+先验证工作区正确性，再运行模型。联合收益归因保留 native/passive/governed 对照；
+没有收益也是有效结果。集成默认关闭。
+
+## 2. 问题与假设
+
+多包工程修复中的不同解释常常需要不同依赖版本、生成代码、构建产物和测试状态。
+Git 能很好地保存跟踪源码；但短分支的环境重建、有效未跟踪状态的交接可能占据大量时间。
+完整复制环境消耗磁盘，共享可写缓存又可能污染对比。另一方面，长期保留的预热
+worktree 已经避免了很多重复准备，不能忽略这条强基线。
+
+| 假设 | 可观察证据 | 反证或适用边界 |
+| --- | --- | --- |
+| H1：分叉完整预热环境可降低探索开销 | 恢复状态等价时，计入初始化摊销后的 fork→可测试、capture、handoff 成本更低 | mount 虽快，commit、I/O、清理或首次导入抵消收益 |
+| H2：结构化证据结合低成本环境分叉，改善工程成果 | 固定总预算下独立验收通过的修复更多，或同等成果的完整成本更低 | 尝试次数和历史展示更丰富，成果却不变或变差 |
+| H3：接手者能复用有效候选，减少重复准备和调查 | 接手者验证精确候选，采用相关 finding，执行下一项有效工作 | 只有读取回执、文件分叉或复制摘要，没有实际采用 |
+
+最可能适合的是频繁分叉、环境准备昂贵且可复用的任务。
+只读代码搜索、少量常驻 checkout、大文件顺序 I/O，或主要时间都花在模型思考上的任务，
+可能收益很小甚至退化。文件系统去重不等于模型推理或上下文 token 去重。
+
+## 3. 范围与非目标
+
+本切片包含公开工程 fixture、隔离的一次性工作区、候选与结果谱系、一条端到端 demo
+和可复现的比较研究。对应路线 **S4/S8/S11**，继承 S3/S7 的权限和预算边界；
+不替代 R2/R3 团队资格验证。
+
+不包含：替换 Goal store、全局记忆、新 Agent 框架、生产级持久性、不可信多租户隔离、
+跨机器共享 Store 写入、自动 Git 发布、进程/VM 完整快照、自主 MCTS 策略，
+或把 LayerFS 2.0 路线项当作当前支持。
+
+## 4. 已审阅的当前系统契约
+
+### LoopX
+
+- [`explore`](../../../loopx/capabilities/explore/README.md) 已拥有节点关联 finding、
+  candidate/input 身份、有界度量反馈和 execution observation。
+  度量和观察只是生产方证据，不直接证明收益；复用其 TypeScript 决策 owner。
+- [Stride §7.7](hierarchical-agent-stride-control-v0.zh-CN.md) 的完整研究点涵盖对话、
+  证据、产物与评测机会。现有 [capture adapter](../../../benchmark/runtime/checkpoint.py)
+  和 [artifact owner](../../../loopx/control_plane/runtime/checkpoint_artifacts.ts)
+  尚不接受任意 LayerFS 引用。当前文件 capture 路径拒绝 symlink，不能直接把 pnpm
+  环境树塞入现有路径。
+- [DSH plugin](../../../packages/dsh-loopx-plugin/README.md) 提供原生绑定、接续和
+  Goal 回读，可作为组合 Host，但当前不等于已有 LayerFS 集成。
+
+### LayerFS
+
+以下事实针对上述精确源码；1.1.0 release 与持续变化的网站文档不能替代该 commit。
+源码来源见附录 A。
+
+| 机制 | 对集成的影响 |
+| --- | --- |
+| SQLite 内容寻址 Store；fork 共享已提交对象；Workspace 挂载可写视图 | 候选可以保留源码、依赖和生成产物，避免完整 payload 复制；初始化和后续写入仍有成本 |
+| `@layerfs/agent`、`@layerfs/backend`、`@layerfs/runtime` 分层 | 使用同一精确版本的公开 API，不依赖 `@layerfs/agent/internal` |
+| 文件变更调用会提交；`workspace.exec`、`workspace.commit`、`workspace.unmount` 是独立调用 | 集成方负责每工具调用的生命周期；一次 Bash 不自动成为持久 checkpoint |
+| 直接打开 Store 要求同一 Linux kernel、本地 ext4/xfs/btrfs；macOS/Windows 有 worker 路由 | 先验证 Linux 或单 Linux VM；单独验证 macOS worker 配置，包括上游 [#456](https://github.com/Ephemeral-AI-Lab/layerfs/issues/456) |
+| Disposable profile、WAL、synchronous OFF；不保证崩溃/掉电持久性 | LoopX 权威状态与源码/结果的独立副本留在 Store 外；不宣传为备份，不把不同持久性的 fsync 速度直接比较 |
+| 共享发布通道；冲突、`.lww`、有界诊断、显式未知结果 | 记录竞争与拒绝；源码、lockfile、评测输入不能自动 last-writer-wins；回复丢失不能触发自动重试 |
+| 追加历史；尚无 branch retirement / 不可达历史 GC（[#411](https://github.com/Ephemeral-AI-Lab/layerfs/issues/411)） | 丢弃 Workspace 不回收已提交分支历史；计入所有失败候选，限定每次实验 Store 生命周期 |
+| 命令可写 Workspace 外部；进程内存、网络效果、外部服务不在快照内 | 声明执行环境和一次性外部 fixture；文件系统回退不等于实验回退 |
+
+公开 v1.0.0 [去重研究](https://layerfs.ai/benchmarks/dsh-dedup/) 在特定整文件重写负载下，
+报告 157 个历史树约占 100 MB，Btrfs+zstd 约 1,995 MB；
+[fio 研究](https://layerfs.ai/benchmarks/fio/) 同时报告顺序读写、随机写的退化。
+这些是上游 `e5c7aaae05` 的测量，本次未复现，不能当作 1.1.0 结果，也不证明工程任务
+成功率或 Agent 完整成本改善。
+
+## 5. 拟议架构与 demo
+
+### Owner 与放置
+
+| Owner | 职责 |
+| --- | --- |
+| LoopX 现有 Explore / Todo / quota owner | 假设与 finding 谱系、claim/lease、工作准入、结果采用、剩余预算和 replan |
+| Host/runtime | 原生会话、模型/工具路由、取消、已结算 capture 边界与执行隔离 |
+| 可选 LayerFS adapter | 候选关联精确 Store/project/Commit/root 与 image 身份；fork、mount、inspect、commit、返回证据 |
+| 现有 benchmark toolkit / 任务 evaluator | 冻结工作负载、独立 oracle、完整成本收集和比较分析 |
+| 现有 Git/review 流程 | 导出选定源码补丁，在干净 checkout 验证并提交评审 |
+
+放置理由：`capability_id: explore`，当前内置 provider 为 `loopx-core`；LayerFS 是
+候选的可选 provider 实现，初期放在 benchmark 实验工作区。只有真实调用方完成验证后，
+才按 [extension 放置规则](../../reference/extensions.md#placement-decision-for-agents)
+讨论产品 distribution。不要仅为此设计注册 provider id、新共享 enum 或 package。
+新的 provider-neutral 决策继续归现有 TypeScript owner，专用 capture 留在 adapter。
+
+保留两张图：Explore 图解释问题和发现；LayerFS 图描述文件系统祖先关系。
+二者关联而非一一对应。多个工具 Commit 可实现一次实验，同一个文件 checkpoint
+也可支持多个 finding。LayerFS Commit 不等于 Git commit，也不等于已验收的 LoopX 成果。
+
+实验 manifest 将现有 task/experiment/result id 关联 provider 源码版本、Store/project
+身份、父/结果 Commit/root、runtime image digest、环境 manifest、evaluator 版本与
+产物 digest。可变 branch 名仅作为展示句柄，不能独自充当证据身份。
+原始 manifest/trace 留在模型 packet 外；使用现有 Explore 字段挂有界摘要和 opaque
+evidence ref。这是研究局部记录，不是悄悄新增的共享运行时 schema。
+
+### 一条完整的小 demo
+
+第一项**候选 fixture**使用公开 [LoopX #6869](https://github.com/loopx-project/loopx/issues/6869)，
+从源码 `907ede6e1faff1fab0f2f765dde4358c69c7b7ff` 开始：不可读 spend artifact
+可能静默降低已计入的消耗。源码确认了 missing-event 路径；第一个里程碑仍须通过真实
+CLI 复现才可纳入 fixture。[PR #6875](https://github.com/loopx-project/loopx/pull/6875)
+是已有修复工作，不是已验收 oracle，也不是本提案要重复交付的工作。
+这是隔离的历史修复练习，不代表认领该 issue。
+
+1. 准备固定 base 的独立 clone，含锁定依赖、已构建 TypeScript runtime、合成的
+   File/SQLite 任务数据。solver 修改的源码仓库与已安装 LoopX controller 及其 runtime
+   状态分离。导入前移除凭据和机器本地资料；Git worktree 的 `.git` 指针不是可移植完整仓库。
+2. 三个 worker 获得相同任务和声明的工具/反馈权限，自主提出、记录假设，分享发现，
+   拒绝或继续路线。不脚本化结论，也不强迫某个分支获胜。
+3. 从精确预热 checkpoint 建立分支。先用确定性 conformance 执行竞争编辑、负结果和
+   receiver handoff；另行运行模型实验，判断这些机制是否被实际使用。
+4. 每个 probe 保存代码/环境身份、测试结果和成本。独立 evaluator 验证 missing、
+   malformed、truncated spend 证据不会静默增加权限；有效 spend/void、history、
+   admission 保持一致，并覆盖时间窗口边界。不能用永久全量 throttling 或某份补丁
+   预期输出代替正确性 oracle。
+5. 接手者从选定不可变候选 fork，验证 manifest、读取相关 finding，执行下一项测试
+   或修复。最终仅导出选定源码补丁到新 checkout，以干净依赖重跑 evaluator。
+   只有分支或缓存中的通过结果不够。
+6. 展示补丁、失败路线、接手者的实际工作、held-out 结果与完整成本分解。
+   录制 UI 回放须标注；脚本 conformance 不算自主模型采用。
+
+这个公开历史任务可能被模型预先见过，只能验证 demo 路径。效果研究还须加入此前未见、
+独立准备的任务，至少覆盖 TypeScript monorepo 与 Rust workspace，包括依赖/生成代码
+变更，以及低分叉负对照。Fixture 保留真实 package graph、lockfile、symlink、
+ignored dependencies、构建产物和代表性错误日志。人为注入故障须标明，不能用重复
+字符串或空文件代替这些状态形态。
+
+### 执行与失败转换
+
+首个 Host 优先调查两个现有 DSH plugin 在一次性 profile 中共存；固定 DSH、双方
+plugin 和 LayerFS 的组合版本。共存、工具路由、session 继承目前**未经验证**。
+不能因 DSH 可用就静默改变研究所选 Host。
+
+每个文件操作前，由现有 Host 准入工作、adapter 绑定候选。直接文件调用经 SDK，
+shell 经 mount → exec → 显式 commit/readback → clean unmount。
+非零退出作为实验证据，不能当修复成功；不要意外丢弃有价值的失败测试状态。
+决定 discard 时显式记录。同一 Workspace 的操作保持串行，worker 各有独立 Workspace。
+
+冲突后保留原输入，检查全部相关路径，包括分页或截断的诊断。在新的工作 fork 中
+解决并重新评测。Merge 前冻结输入 head；原分支的测试通过不覆盖新合并 root。
+回复丢失时，先对照精确 history 与 Workspace status 再执行新操作；无法确认成功就
+保持 unknown、停止该分支，不能盲目重放命令或 force commit。
+清理也不能把未决操作改记为成功。
+
+仅文件系统分叉时，两组都以相同声明上下文开启新会话。后续如研究同对话接续，
+必须满足 Stride §7.7 的 host cursor、证据、产物与机会 barrier。
+LayerFS 的 provider 契约通过验证后，只能替换 artifact participant；不能使当前
+file-only capture 自动接受 symlink，也不能通过复制文件恢复 lease、预算、
+evaluator 次数或外部效果。
+
+## 6. 替代方案与比较
+
+| 方案 | 保留它的理由 |
+| --- | --- |
+| 预热 Git worktree + 锁定依赖缓存 | 主要实用对照；允许合理缓存，其 checkpoint 路径也须保存必要未跟踪产物 |
+| 合适 Linux Host 上的 Btrfs/reflink 或 overlay snapshot | 完整环境低成本分叉的强替代；可用时比较，不可用时披露覆盖缺口 |
+| LayerFS workspace provider | 测完整树共享历史、增量 capture、分支交接，也计入 FUSE 与发布成本 |
+| 普通常驻单 worker checkout | 分叉很少时的负对照，同时判断并行协调本身是否有益 |
+
+不能只用已预热 LayerFS 对比每次冷下载依赖的对照。分别测冷启动和多次热复用，要求
+可恢复状态与保留历史等价。依赖缓存只读共享或按分支隔离。原生 snapshot 仍可能更优。
+
+每次 attempt 在环境准备前启动 wall 时钟，计入终局验证/导出。
+Provider 的一次性 build/image 准备单列，并计入已声明摊销口径的总成本；
+不能隐去，也不能只计给对照。稳态视图补充冷启动总量，而不替代它。
+
+## 7. 安全、隐私与兼容性
+
+使用同一 Linux kernel 内的一次性公开 fixture，固定 CPU/RAM/磁盘上限，单独控制网络。
+不要导入开发者正在使用的目录：LayerFS 会有意保存 ignored files。
+凭据、模型历史、真实 LoopX state、evaluator 内部资料、daemon endpoint 与 Docker
+访问权限放在 Agent 可写树之外。首个研究使用固定空 `.lww`，源码、lockfile、测试不能
+自动 newer-wins；改变策略即改变 treatment。
+
+检查可执行 mode、symlink、hardlink、绝对路径、native dependency ABI、写 root 外的
+package。导入环境必须匹配 guest OS/架构/image；arm64 macOS 依赖树不是 Linux 环境。
+LayerFS 隔离本身不证明租户安全。
+
+Feature-off 保持已有执行与状态。Adapter 失败要可见；fallback 是单独标记的新 attempt，
+不能静默换 arm。模型提示词不能替代对真实工具读写位置的验证。
+LoopX 控制调用留在其 owning Host，不放进正在修复的候选树。
+
+## 8. 迁移与回退
+
+不迁移已有 Store 或 Goal。从独立保存的 fixture 初始化新研究 Store。
+选定源码补丁和验证证据存放在外部；宣告成功前验证 clean-checkout replay。
+停止新准入、核对在途操作、导出合格结果、关闭本实验的 Workspace/Sandbox，再移除
+研究 Store。不要把 Workspace discard 描述为历史 GC。
+关闭可选 adapter 即恢复原工作区路径。源码格式升级开启新的研究 generation，
+不原地改造历史证据。
+
+## 9. 评测与验收
+
+### 区分底层机制收益与 Agent 收益
+
+首先在两种 workspace provider 上重放同一预定文件/构建/测试序列，在资源允许时
+覆盖 1/3/8 个 worker 与 1/4/16 个分支，运行前记录排除的单元。
+保持拓扑、保留状态和命令语义一致；它只估计底层开销。
+
+随后使用以下矩阵，固定 Host/model/reasoning、任务划分、测试可见范围、总资源预算
+和最多三个 worker：
+
+| 控制策略 | 预热 worktree（W） | LayerFS（F） |
+| --- | --- | --- |
+| Native Host（N） | N-W | N-F |
+| Passive LoopX（P） | P-W | P-F |
+| Governed LoopX（G） | G-W | G-F |
+
+第一轮小规模模型 pilot 可只运行 **G-W vs G-F**，只能估计该 LoopX 策略下的 workspace
+效果，不能估计 LoopX 收益或联合协同。完整矩阵区分观察开销、治理与 workspace 作用。
+协同声明需要在共同 outcome 标尺上预先声明 difference-of-differences，并给出不确定性；
+仅 G-F 获胜不足以证明协同。
+
+模型评测前冻结 task id/base commit、mutation generator/evaluator hash、模型版本、
+SDK/image/plugin pin、branch/depth 上限、反馈时序、美元/token/compute/wall 上限与
+缺失数据规则。协调者与所有 worker 共用**一个总预算**。Agent 研究允许自主改变
+分叉策略，但不能突破该预算；不能给某 provider 额外时间、worker 槽位或 token
+补偿其开销。
+
+使用配对任务/seed、交错 arm 顺序，按现有研究契约，每个拟推广比较单元至少五次重复；
+五次不是统计功效依据。任务选择和 sequential stop 规则须先于比较结果确定。
+所有 solver 可访问文件和网络均不得泄漏 held-out task/evaluation。
+开发反馈、计划中的评测和终局评分作为独立机会处理。
+准备失败、取消、timeout、treatment 未交付、无效评测均保留在分母并单独分类。
+广泛结论之前冻结功效/精度计划。
+
+### 度量
+
+主要度量是固定总预算的独立验收通过率，以及随时间变化的最佳有效保留成果。
+同等成果再比较时间和完整成本，计入所有 attempt。
+根据实测冷准备与边际复用成本估算盈亏平衡分支数，不能只看 fork latency。
+解释性指标包括：
+
+- 初始化、依赖获取、mount/fork、工具执行、capture/commit、冲突解决、handoff、
+  验证、导出、清理的 wall/CPU 时间；
+- 未缓存/缓存输入与输出 token、真实模型账单、控制与 evaluator 工作、人工介入次数
+  及已测注意力时间；
+- 物理分配 Store 与 WAL、overlay、container scratch、cache、独立导出和峰值
+  disk/RSS，另列 logical history bytes；
+- 竞争、拒绝/未知操作、恢复工作、重复 probe，以及证据→决策→receiver 实际工作，
+  不奖励分支计数。
+
+LayerFS trace/backend counter 只作诊断，同时使用独立 runner 时钟、OS 资源测量与
+provider usage。Trace 写入可能丢失，缺记录不是零成本。
+遵循现有度量契约：packet byte 不是模型 token，去重比不是总磁盘节省；账单/注意力
+未知就不能声明完整成本。重叠 wall 区间不能相加当 elapsed duration，同时报告总
+worker compute time。
+
+### 验收行
+
+| 声明 | 所需证据 | 通过边界 |
+| --- | --- | --- |
+| A1：真实环境分叉保真 | fresh remount 后独立树/内容/mode/link 检查、相同测试，覆盖 ignored output 和 symlink 密集 package tree | 每个选定 checkpoint 的声明状态一致；显式归一化元数据差异，不整体忽略 |
+| A2：隔离与冲突完整性 | 并行独立编辑、同文件冲突、merge root 重测、外部路径 sentinel、no-op/read-only 调用 | 无候选污染、静默覆盖；未变状态不伪造 Commit |
+| A3：如实处理失败/恢复 | 非零工具退出、回复丢失、Workspace 中断、worker 缺失、磁盘上限、证据缺失 | Unknown 保持 unknown；无重复执行、假成功或自动权限恢复 |
+| A4：有效交接与产物 | Receiver 验证候选、采用 scoped finding、执行新工作；导出补丁通过独立 clean-checkout 评测 | 证明完整脚本路径；模型实际采用须单独运行验证 |
+| A5：因果效率证据 | 冻结 matched cell、完整 attempt 统计、配对不确定性和原生结果 | 结论限于对比支持的范围；demo/microbenchmark 不能直接 promotion |
+| A6：可用的可选旅程 | 打包 Host 可发现可用性、执行授权运行、展示 conflict/unavailable、读取结果和关闭 provider | 产品推广前必须满足；CLI-only 实验标为局部交付 |
+
+确认性研究前，依据 pilot 方差和使用方成本，预注册主要对比、最小实用效果和
+可接受成果损失界限。不能看完确认性结果再选这些值。
+置信区间同时覆盖实质伤害与收益时，结论保持不确定。
+正确性/隔离失败无论多快都不合格。收益仅在存储时，就只报告存储收益。
+
+## 10. 运行与产品契约
+
+复用现有 Goal/Explore research history 与 DSH conversation 回读任务和结果。
+LayerFS Work/History 可显示文件谱系，但不能替代保留候选的理由。
+实验不需要新增顶层页面。产品 provider 仍须在现有设置/capability 中提供 readiness、
+授权启动、unavailable/error 恢复，以及 App/CLI 和任何宣称支持的 Lark 入口的关闭/回读覆盖。
+
+实验 runner 限制分支数、Store 总增长、并发 worker 与命令时长。
+按成功和失败分支共同预测保留成本，只清理本研究资源。
+展示 stale/unknown checkpoint 与不可用观察。不要每轮向模型灌入原始日志和不断增长的
+历史；按不可变引用获取详情，度量实际消费上下文。本 RFC 不改变 runtime packet
+或 compute quota。
+
+## 11. 规范性交付计划
+
+| 里程碑 | 进入条件 | 退出证据 | 回退 / owner |
+| --- | --- | --- | --- |
+| M0：冻结 demo | 本提案与公开候选 fixture | 真实复现 base 失败；评审独立 oracle；固定 Host/provider manifest 与明确有界预算 | 不改活跃状态；LoopX 实验 owner |
+| M1：一条可靠集成路径 | M0；隔离 runtime；兼容 SDK/native/image | 真实 LayerFS、LoopX 回读、干净导出补丁通过 A1–A4 确定性 conformance；关闭路径 parity | 移除一次性 profile/provider；LoopX adapter owner，上游就 API 缺口提供意见 |
+| M2：比较 demo 与 pilot | M1 | G-W/G-F 配对 pilot、失败路线、完整成本覆盖和如实效果/限制报告；较大研究先预注册 | 保留证据、不改默认；benchmark owner |
+| M3：归因与可选采用决策 | 充分 pilot 证据与预声明研究 | Native/passive/governed 矩阵、held-out 任务分层、A5 与独立复现；产品声明前满足 A6 | 保持实验性质或拒绝 provider；现有维护者决策 |
+
+一条 LoopX issue 承接上述序列；一条 LayerFS issue 请求 API/评测反馈。
+复用上游 #411/#456 和 [#446 路线](https://github.com/Ephemeral-AI-Lab/layerfs/issues/446)，
+不重复创建 GC、worker 或通用 worktree benchmark 任务。
+可移植底层测量可贡献至 `layerfs-benchmark`；治理与任务结果实验由 LoopX 负责。
+上游意见有价值，但不能阻塞通过现有接口可独立推进的 fixture/oracle 准备。
+
+## 12. 待定事项
+
+| 决策 / owner | 建议 | 所需证据 / 截止边界 |
+| --- | --- | --- |
+| Host profile / LoopX Host 维护者 | 优先 DSH 组合，设计保持 runtime-neutral | M1 前核验 plugin 版本、control-call 路由、worker 继承、取消与无双重 commit |
+| Fixture 与预算 / 实验 owner | #6869 公开 demo，held-out TS/Rust 分层研究效果 | M0 退出前完成 base 复现、license/privacy 审阅、oracle、cost cap |
+| Provider API / LayerFS 维护者反馈 | 同一精确 commit 的公开 1.1 时代 packages；不改 internal API 或持久性 | 宣称 provider 支持前确认 worker 配置、exact-root 比较、未知操作回读与清理范围 |
+| Promotion / LoopX 现有维护者 | 重复收益和 A6 满足前保持可选 | M3 关闭前具备预声明实用效果阈值、功效/精度和真实旅程证据 |
+
+## 附录 A：证据登记
+
+检查日期为 2026-10-26。源码审阅不等于 benchmark 复现或集成实测。
+本 RFC 未运行 LayerFS build、模型实验或 runtime 安装；本地 Docker 可用性检查仅用于
+判断可行性。
+
+| 证据 | 发现 / 限制 |
+| --- | --- |
+| [LayerFS 固定 README](https://github.com/Ephemeral-AI-Lab/layerfs/blob/ad78914e36b258b360daf13aec164b5655814689/README.md) | 发布源码/API 分层、Disposable 限制、有明确范围的 v1.0 测量 |
+| [Agent SDK guide](https://github.com/Ephemeral-AI-Lab/layerfs/blob/ad78914e36b258b360daf13aec164b5655814689/api/how-to-use/agent.md) 与 [exec 实现](https://github.com/Ephemeral-AI-Lab/layerfs/blob/ad78914e36b258b360daf13aec164b5655814689/api/agent/sdk/src/workspace/exec.js) | 公开调用、显式 commit 生命周期、guest 设置、流式输出、有界读取/merge 诊断 |
+| [Runtime guide](https://github.com/Ephemeral-AI-Lab/layerfs/blob/ad78914e36b258b360daf13aec164b5655814689/api/how-to-use/runtime.md) | Host 所有的初始化、macOS/Windows worker、资源清理；不证明 plugin 共存 |
+| [Backend/trace guide](https://github.com/Ephemeral-AI-Lab/layerfs/blob/ad78914e36b258b360daf13aec164b5655814689/api/how-to-use/backend.md) | Root 比较限制、存储口径和可能丢失的诊断 trace |
+| [Store 架构](https://github.com/Ephemeral-AI-Lab/layerfs/blob/ad78914e36b258b360daf13aec164b5655814689/core/docs/architecture/84-global-sqlite-store.md) | 单 kernel 边界、发布语义、无不可达历史 GC |
+| [DSH 集成](https://github.com/Ephemeral-AI-Lab/dsh-layerfs/blob/baac1b92b97f0bd492002c285f6a13401b4f4c0b/README.md) | 现有 Bash/SDK 路由、Work/History UI；执行前需核对其 LayerFS pin 与提议源码 pin |
+| [LoopX 历史 readback](https://github.com/loopx-project/loopx/blob/907ede6e1faff1fab0f2f765dde4358c69c7b7ff/loopx/control_plane/quota/ledger_readback.py) | Missing/unreadable event 不贡献消耗；真实入口复现和正确修复 oracle 仍属于 M0 |
+
+网站广义共享存储描述不能替代受支持的单 kernel Store 契约。
+旧概念页和 1.1 之前的 SDK 名称不作为本研究实现指导。
+Open issue 报告用于识别风险，不代表本次独立复现的缺陷或承诺修复。
diff --git a/docs/architecture/rfcs/loopx-overall-roadmap-v0.md b/docs/architecture/rfcs/loopx-overall-roadmap-v0.md
index 75b59e1276..7cc8feb9d4 100644
--- a/docs/architecture/rfcs/loopx-overall-roadmap-v0.md
+++ b/docs/architecture/rfcs/loopx-overall-roadmap-v0.md
@@ -490,6 +490,7 @@ subsystem was not performed. Section 8 records the focused audit.
 | [Human-confirmed domain operations (v0)](human-confirmed-domain-operations-v0.md) | S8/S9 + R2/R3 | Accepted; canonical operation seam exists, managed native transport under owner review | Qualify source-context vs admitted-executor separation, exact human approval→one-shot consumption→domain evidence→original-route return; owned Turn/delegation need not await Desktop authentication. Live approval/effect/wakeup remain unqualified; finance provider stays separate |
 | [Provider-side authorization at effect acceptance (v0)](provider-effect-acceptance-v0.md) | S8/S9, supporting S2/S4 | Accepted design only; no runtime integration or qualified provider | M1: controlled provider and deterministic revoke/crash/replay conformance; strict production wiring remains gated by exact Goal lifetime, receipt retention and independent provider qualification |
 | [Research Exploration Control Plane v0](research-exploration-control-plane-v0.md) | S11/S3 | Accepted; partial M2 composition/successor; §11.5 result/adoption refinement proposed | P1: one real provider result -> scoped interpretation -> adopted step/successor, exact lineage/shared write gate and packaged readback; qualify shadow before opt-in, retain composition gaps; defer inference and M4 portfolio ranking |
+| [LoopX × LayerFS Engineering Exploration and Evaluation v0](layerfs-engineering-exploration-evaluation-v0.md) | S4/S8/S11 | Proposal; no integration or benefit claim | Bounded experiment: real environment fork/handoff/clean export, then matched warm-worktree and control-policy comparisons; preserve existing research-state/authority owners, count complete cost and negative results; no new priority or default |
 | [Hierarchical Agent Stride Control v0](hierarchical-agent-stride-control-v0.md) | S11/S7 | Accepted; M1 observation and source-level replan decision followthrough; adaptive policy/continuation unqualified | P2: §7.6 mechanism/evidence scope review -> actual probe/adoption/full cost; §7.7 same-state continuation and recovery before adaptive promotion. Preserve per-Turn settlement and authority; no forced-hour pivot or default cadence change |
 | [Goal-scoped Capability Portfolio v0](goal-scoped-capability-portfolio-v0.md) | S1/S3/S6/S8/S11 | Accepted; read-only settings/context inspection slice | P1: correction→fresh-session decision across existing owners, then demand-driven composition and measured method evolution; no universal memory database or second opt-in |
 | [Post-Outcome Memory Utility Attribution v0](post-outcome-memory-utility-attribution-v0.md) | S6/S11 | Accepted; Stage 1 verified-outcome binding | P1 read-only reducer→P2 pilot: distinguish recalled/applied/utility, no automatic ranking change |
diff --git a/docs/architecture/rfcs/loopx-overall-roadmap-v0.zh-CN.md b/docs/architecture/rfcs/loopx-overall-roadmap-v0.zh-CN.md
index 5108402ea7..e4a9694f3e 100644
--- a/docs/architecture/rfcs/loopx-overall-roadmap-v0.zh-CN.md
+++ b/docs/architecture/rfcs/loopx-overall-roadmap-v0.zh-CN.md
@@ -376,6 +376,7 @@ Muse 设计页在浏览器超时，其文章通过网页检索读取。本次调
 | [Human-confirmed domain operations (v0)](human-confirmed-domain-operations-v0.zh-CN.md) | S8/S9 + R2/R3 | 已接受；规范操作接缝存在，受管原生传输待 owner review | 验收来源上下文与已准入执行者分离、精确用户批准→单次消费→垂域证据→原路返回；自有 Turn/delegation 不等待 Desktop 认证。真实批准/效果/唤醒仍未资格化；金融 provider 保持独立 |
 | [Provider 在效果接受点执行授权（v0）](provider-effect-acceptance-v0.zh-CN.md) | S8/S9，S2/S4 支撑 | 已接受设计；尚未接入 runtime，也未准入 provider | M1：controlled provider 与 deterministic revoke/crash/replay conformance；strict production 接线仍需精确 Goal 生命周期、receipt retention 与独立 provider 资格 |
 | [Research Exploration Control Plane v0](research-exploration-control-plane-v0.zh-CN.md) | S11/S3 | 已接受；M2 composition/successor 局部实现；§11.5 result/adoption 修订为提案 | P1：一条真实 provider result -> 有范围 interpretation -> adopted step/successor，精确 lineage/shared write gate 与 packaged readback；shadow 资格先于 opt-in，保留 composition gap；推断与 M4 portfolio 排序继续 defer |
+| [LoopX × LayerFS 复杂工程探索与效果评测 v0](layerfs-engineering-exploration-evaluation-v0.zh-CN.md) | S4/S8/S11 | 提案；无集成或收益声明 | 有界实验：真实环境 fork/handoff/干净导出，然后匹配预热 worktree 与控制策略进行比较；保留现有研究状态/权限 owner，计入完整成本和负结果；不新增优先级或默认值 |
 | [Hierarchical Agent Stride Control v0](hierarchical-agent-stride-control-v0.zh-CN.md) | S11/S7 | 已接受；M1 观测及源码层面的 replan 决策接续；自适应策略/恢复接续未资格化 | P2：§7.6 机制/证据范围复盘 -> 真实探针/采用/完整成本；§7.7 同起点接续及恢复验证先于自适应 promotion。保留每 Turn 结算与权限，不按固定小时强制转向、不改变默认 cadence |
 | [Goal-scoped Capability Portfolio v0](goal-scoped-capability-portfolio-v0.zh-CN.md) | S1/S3/S6/S8/S11 | 已接受；只读配置/上下文检查切片 | P1：复用既有 owner 验收纠正→新会话决策，再做按需组合与可测方法演化；不建万能记忆库或第二个启用开关 |
 | [Post-Outcome Memory Utility Attribution v0](post-outcome-memory-utility-attribution-v0.zh-CN.md) | S6/S11 | 已接受；Stage 1 verified-outcome 绑定 | P1 只读 reducer→P2 pilot：区分 recalled/applied/utility，归因不自动改 ranking |