[Dong1017]

Governing Issue and delivery slice

Refs #372

• Accepted scope / delegation: Bounded archive-only slice delegated by the repository owner (Dong1017) on 2026-11-10 (UTC+8). Public record: docs: archive stale planning notes under docs/plans #383 (comment). This slice only archives the two root planning notes and marks the plans under docs/plans/ as historical. It does not treat the 3+1 split as a newly accepted plan, and it does not close refactor: track architecture migration, integration PRs, and repository layout #372 or discussion: Knowledge / Harness / Benchmark 三模块目标架构——现状、原语、交接时序与遗留问题 #362.
• What this PR delivers: An in-place archive. PHASE1_REPORT.md and SUBMISSION_NOTES.md move into docs/plans/ under dated names, the phase-1 report's internal links are retargeted, the one in-repo path mention is updated, and each plan file under docs/plans/ has the same short historical-archive notice. File bodies are otherwise preserved. This is an archive (move + historical notice), not a deletion.
• What it does not deliver: No file is removed from history. AGENTS.md, evidence/, the skill library, skill-eval runbooks, both official matrices, docs/skill-coverage.md, docs/kv-reclamation-qualification.md, and docs/examples/harness-improvement/summary.json are unchanged. No docs/migration/ directory is created. Acceptance boxes on refactor: track architecture migration, integration PRs, and repository layout #372 and discussion: Knowledge / Harness / Benchmark 三模块目标架构——现状、原语、交接时序与遗留问题 #362 are untouched.

Knowledge authorship declaration

Not applicable. This PR does not change protected knowledge paths or knowledge content. No policy exception is requested.

{
  "schema_version": 1,
  "head_sha": "08246fcaad2865c49ba5c15389a262f7935099e8",
  "human_only": false,
  "material_ai_authors": [
    {
      "tool": "Cursor",
      "model_id": "grok-4.7",
      "effort": "unknown",
      "role": "author"
    }
  ]
}

Changed artifacts and evidence

Moved with git mv. A search of the tree before the move found one in-repo path hit for PHASE1_REPORT.md and none for SUBMISSION_NOTES.md. No evidence manifest or other hash-pinned path named either file, so both moves proceeded.

Old path	New path	Content change
PHASE1_REPORT.md	docs/plans/2026-10-14-phase1-report.md	Three relative links now point at ../architecture.md, ../reference-matrix.md, and ../searchspace-matrix.md. Rename similarity 98%.
SUBMISSION_NOTES.md	docs/plans/2026-10-28-submission-notes.md	Same historical-archive notice as the other plans, inserted under the title. The date is the 2026-10-28 commit that added the file.

Historical-archive notice, inserted under the title and otherwise leaving the body as it was:

• docs/plans/2026-10-18-inferrsi-rename.md
• docs/plans/2026-10-18-negative-diagnostics.md
• docs/plans/2026-10-19-campaign-plan-rfc.md
• docs/plans/2026-10-28-submission-notes.md (follow-up on the same branch)

The notice line is identical in all four files:

Historical archive. Kept for provenance. Not current implementation or validation status.

In-repo reference update:

• docs/reviews/source-audit-ac70272.md now names docs/plans/2026-10-14-phase1-report.md. The rest of docs/reviews/ is unchanged.

git grep on the committed tree for PHASE1_REPORT and SUBMISSION_NOTES returned no hits. Already-landed Issue and PR text was left as written. #372 still names the old root paths in its body; that text was not edited. Closed issue #43 also names PHASE1_REPORT.md and was left alone.

Requirement	Actual change	Check / evidence and result	Remaining gap
Move both root notes into docs/plans/	Two git mv renames above	Root listing no longer contains either file; both live under docs/plans/	None in this diff
Keep phase-1 links resolving	Three relative links in the moved report	tests/test_template_layout.py::test_maintained_documentation_links_resolve passed with the draft smoke and claim tests (11 passed) on the move, and again alone after the submission-notes notice (1 passed)	None
Update in-repo path mentions	Source-audit backtick path	git grep for the old filenames: no hits	Issue and PR comments outside the tree can still name the old paths
Mark every plan under docs/plans/ historical	One notice line on each of the four files	Diff is those two added lines per file; the four notice lines match	None

Checks and dependencies

• Checks actually run (result and tested scope): On the move commit: python3 -m pytest tests/test_template_layout.py tests/test_smoke.py tests/test_claim.py — 11 passed in 0.23s, and python3 -m ruff check src tests — all checks passed (ruff 0.17.4). After the submission-notes notice: python3 -m pytest tests/test_template_layout.py::test_maintained_documentation_links_resolve — 1 passed in 0.20s. The moved submission notes contain no relative links to retarget.
• Failed, skipped, timed-out or not-run checks (if applicable): Full pytest was not run. No workflow, Python, or evidence file changed.
• Dependencies / shared interfaces / integration assumptions (if applicable): None. The archive does not change a runtime path or a frozen evidence hash.

Remaining work and Issue closure

• Remaining Issue requirements / follow-up: refactor: track architecture migration, integration PRs, and repository layout #372 still tracks the integration sequence and its acceptance boxes. This PR is one archive slice. The coordinator can request review after the PR exists.
• Goal-acceptance owner or accepted closure delegation: Unchanged. This PR does not record acceptance.
• Goal ready for acceptance? Not claimed

  

[cursor]

LGTM

PR review verdict — PASS
Reviewed base 3f56064 and head f1e5a64. Full diff on opened; no prior review.

Verified:

• Rename PHASE1_REPORT.md → docs/plans/2026-10-14-phase1-report.md. The historical-snapshot banner links now use ../architecture.md#three-objective-delivery-framing, ../reference-matrix.md, and ../searchspace-matrix.md. From docs/plans/ those resolve to the same docs/ pages the previous root-relative docs/... links targeted.
• Pure rename SUBMISSION_NOTES.md → docs/plans/2026-10-28-submission-notes.md (similarity 100%).
• Historical-archive banner added after the title in docs/plans/2026-10-18-inferrsi-rename.md, docs/plans/2026-10-18-negative-diagnostics.md, and docs/plans/2026-10-19-campaign-plan-rfc.md, with a blank line before each file's existing blockquote.
• docs/reviews/source-audit-ac70272.md now names docs/plans/2026-10-14-phase1-report.md in the licensing observation.

Optional follow-up: add the same historical-archive banner to docs/plans/2026-10-28-submission-notes.md. The phase1 report already carries a 2026-10-14 snapshot disclaimer; the three in-place plans gained the banner in this diff.

Not verified here: docs-only change, tests not run, CI not checked. Seeded evidence is the full patch, so unchanged lines inside the two renamed files were not re-read for other relative links, and the #three-objective-delivery-framing anchor was not re-opened in docs/architecture.md.

[Dong1017]

@zengchuang-hw @amy-why-3459 Please review this PR.

This is an archive-only follow-up on the #372 layout track: PHASE1_REPORT.md and SUBMISSION_NOTES.md move into docs/plans/ under dated names, the three existing plan files get a short historical-archive notice, and the one in-repo path mention is updated. Nothing is deleted. AGENTS.md is unchanged, and this PR does not close #372 or #362.

Current head: f1e5a642be3945587ee35a7c01a1ffde79a2864d.

[amy-why-3459]

LGTM

Reviewed the full diff at f1e5a642be3945587ee35a7c01a1ffde79a2864d (base 3f560642a44a1496aa4984ab936517693be2c788).

• The two renames are archive-only. SUBMISSION_NOTES.md moves with its bytes unchanged. In the phase-1 report, only the three relative links in the header notice are retargeted, to ../architecture.md, ../reference-matrix.md and ../searchspace-matrix.md, and those were the only relative links in the file.
• Each of the three existing plans gets one historical notice line, with no other body changes. docs/reviews/source-audit-ac70272.md §15 now names the new path.
• No protected knowledge path is touched (checked against .github/knowledge-model-policy.json). declarations passed.
• CI on this head is green: full-check (3.10/3.11), installed-package (3.10/3.11), lint-and-smoke and pr-references.

[Dong1017]

@zengchuang-hw @amy-why-3459 Please review the latest head.

A follow-up commit is now on this PR: 08246fcaad2865c49ba5c15389a262f7935099e8. It addresses the optional cursor[bot] suggestion by adding the same historical-archive notice to docs/plans/2026-10-28-submission-notes.md. No other files changed in that commit.

The earlier approvals from amy-why-3459 and cursor[bot] were on f1e5a642be3945587ee35a7c01a1ffde79a2864d and have been dismissed after this push. Please look at the current head.

Checks on the new head so far: declarations, pr-references, lint-and-smoke, and installed-package (3.10/3.11) have passed. full-check (3.10/3.11) is still running. Check records: 08246fc

[zengchuang-hw]

Interim non-voting review of head 08246fc, against base 3f56064.

Incremental scope: f1e5a64 → 08246fc. I read the complete one-commit diff and the current submission-notes file. The only delta adds the historical-archive notice and a blank line. This resolves the optional suggestion in #383 (review). No additional technical defect or optional change found in this increment.

The prior full six-file assessment still applies: the phase-1 report's three relative links resolve, the original submission-note body is preserved, and no runtime or protected knowledge path changed. Previous review reference: #383 (review) (on the old head, now dismissed).

Before approval, please link the accepted scope/delegation for this bounded two-file archive slice. #372 still opens with “Status: Proposed” and explicitly says it does not authorize moves; its sole discussion comment does not supply that acceptance. The PR description says this is a follow-up but does not link a separate accepted decision. This is an evidence gap, not a claim that the archive is technically faulty, and does not require accepting the entire 3+1 proposal or closing #372/#362.

Checks: independently inspected the GitHub check records for this exact head. All eight check runs are now completed successfully, including full-check and installed-package on Python 3.10/3.11, lint-and-smoke, pr-references and both declarations runs.. I did not run local pytest or Ruff. The author's local test results remain author-reported. This comment is interim feedback, not an Approve or Request changes vote.

[amy-why-3459]

LGTM

Incremental review f1e5a642be3945587ee35a7c01a1ffde79a2864d..08246fcaad2865c49ba5c15389a262f7935099e8 (one commit). My earlier approval on f1e5a642 was dismissed by this push.

• The only change adds the same historical-archive notice line to docs/plans/2026-10-28-submission-notes.md, directly under the title, matching the other archived plans. The note body is otherwise unchanged.
• My full-diff assessment from the previous head still applies; no protected knowledge path is touched.
• CI on this head is green: declarations, pr-references, lint-and-smoke, full-check (3.10/3.11) and installed-package (3.10/3.11).

[Dong1017]

@zengchuang-hw Thank you for the interim review on head 08246fcaad2865c49ba5c15389a262f7935099e8.

This PR is a bounded archive-only slice, delegated by the repository owner (Dong1017) on 2026-11-10 (UTC+8). That confirmation was given outside this issue thread, so this comment is the public record of the delegation. It is not an acceptance of the #372 3+1 layout proposal, and it does not authorize closing #372 or #362.

Accepted scope for this slice:

• Move PHASE1_REPORT.md to docs/plans/2026-10-14-phase1-report.md and retarget only its header relative links.
• Move SUBMISSION_NOTES.md to docs/plans/2026-10-28-submission-notes.md, preserving the original body and adding the same historical-archive notice used on the other archived plans.
• Add that historical-archive notice to the three existing files under docs/plans/.
• Update the one in-repo path mention in docs/reviews/source-audit-ac70272.md.
• Do not delete files, do not change AGENTS.md, and do not close refactor: track architecture migration, integration PRs, and repository layout #372 or discussion: Knowledge / Harness / Benchmark 三模块目标架构——现状、原语、交接时序与遗留问题 #362.

Current head remains 08246fcaad2865c49ba5c15389a262f7935099e8.

[zengchuang-hw]

LGTM

PR review verdict — PASS
Reviewed base 3f56064 and head 08246fc.

This completes my interim review: #383 (review). I independently assessed the original six-file diff and the incremental commit f1e5a64..08246fc. The increment only adds the historical-archive notice to the submission notes, resolving the optional suggestion.

The bounded archive scope and delegation are now publicly recorded at #383 (comment). This resolves the missing scope-record item from my interim feedback; it does not accept the broader #372 proposal or authorize closing #372/#362.

Verified: the two notes retain their original bodies apart from the declared header changes, the phase-1 report's three relative links and architecture anchor resolve, the in-repo reference is updated, and the plan notices are consistent. No runtime or protected knowledge content changes. No outstanding blocking findings or additional optional suggestions.

All nine current check runs on this head are successful, including declarations, pr-references, lint-and-smoke, full-check (3.10/3.11), and installed-package (3.10/3.11). I inspected remote checks; I did not run local pytest or Ruff, and the author's local results remain author-reported.

Independent reviewer: dot acting for zengchuang-hw under the user's direct instruction to review and approve passing changes. This is a revision-scoped PR approval, not Issue-level or campaign acceptance.