Skip to content

refactor: keep split facades' base public names bound and typed - #15675

Merged
bolichen97 merged 1 commit into
mainfrom
refactor/qf2-facade-name-surface
Oct 1, 2026
Merged

bolichen97 merged 1 commit into
mainfrom
refactor/qf2-facade-name-surface

Conversation

@bolichen97

Copy link
Copy Markdown
Contributor

Problem / Motivation

Goal: Six split facades resolve every public name they offered before their split again, two forwarding facades stop hiding public re-exports from mypy, and the AutoNudge loader always reads the file its store writes.

The campaign audit of the merged facade splits (quality report §3.2 #14, #15 and #20; rework plan S2, S3 and S8) found three latent gaps that break nothing on main today:

  • X-08: six facades stopped resolving public kiro_crew names they bound before their split. These are acp/client.py (B09, 6 names), cron.py (B11, 5), autonudge.py (B12, 28), the Ops Mission Control routes (I02, 21), the Issue Radar routes (I06, 3) and mcp_gateway/gatewayd.py (I10, 23). A caller or a star import that still uses one gets AttributeError. gatewayd.py's docstring also claimed that "every name an owner holds is re-exported".
  • X-09: pod/runtime.py (B06) and the source-providers handler (B01) hide their forwarding __getattr__ from mypy, but their TYPE_CHECKING imports listed only part of the public _EXPORTS. As a result, 37 and 19 names resolve at run time but fail type checking with attr-defined.
  • B12-3: AutoNudgeService.__init__ copied self._store.path into self._path. _load read the copy while writes went to self._store.path, so reassigning the store's path left the loader reading a different file.

Why it matters

The campaign's facade contract says a split must not shrink a facade's public surface or hide it from the type checker. Each gap is a trap for the next caller or test: an import of a name the facade has always offered fails at run time, a mistyped or removed rt.<name> read type-checks as Any, and two copies of one path can drift apart.

Not a goal

  • No behaviour change, no new feature and no change to the owners' code.
  • Re-bound names stay plain re-exports, not patch seams: owners keep reading their own bindings, and the facade docstrings now say so. Adding corpus-derived patch-reach guards to the cron, Ops Mission Control and Issue Radar facades is the S9/M5 follow-up work.
  • Standard-library, typing and third-party imports and private names the splits dropped are not restored. Examples are cron's cron_descriptor Options / get_description and gatewayd's five private imports.
  • onboarding_import.py and artifacts.py (QF1) and I04 (QF4) are not touched.

What changed (motivation → approach → change)

Splitting these modules into owners shrank their public surface. Callers lost names they could import, mypy lost sight of names that still worked at run time, and one path was held twice. This PR puts the surface back without changing what any owner does. Each facade gets its fix through the mechanism it already uses:

  • acp/client.py (forwarding facade): the six names (launcher_refusal, sandbox_init_remediation, LAUNCHER_EXIT_PREFIXES, the two SANDBOX_LAYER_*, ACP_BACKENDS_HOST_AUTH_CALLBACK) join _EXPORTS_BY_OWNER["kiro_crew.acp.transport_errors"], the one module that reads them, and its TYPE_CHECKING block. This follows the existing corroborate_launcher_refusal precedent, so a patch through the client still reaches the reader.
  • Plain facades (cron.py, autonudge.py, src/kiro_crew/apps/builtins/ops_mission_control/backend/routes.py, src/kiro_crew/apps/builtins/issue_radar/backend/routes.py, gatewayd.py): # noqa: F401 re-export lines. Each docstring now says these names resolve but are not patch seams. In gatewayd.py every line names one import and is marked not a seam, so isort cannot merge them into the blocks of names the module uses. Two function-local targets assignments in autonudge.py take # noqa: F811, now that the module re-exports probes.targets.
  • pod/runtime.py, src/kiro_crew/dashboard/handlers/source_providers.py: every public _EXPORTS name is imported under TYPE_CHECKING from the owner the table names.
  • AutoNudgeService._path is a read-only property returning self._store.path.
  • Specs: learn-cron-dashboard.md (the cron and AutoNudge surface sentences, and the _path state row) and mcp-gateway-daemon-lifecycle.md (one sentence that matches the gatewayd.py docstring word for word).

Backwards compatibility

Compatible: names are only added to the six facades. No caller-visible behaviour changes. _path is still readable, and no code in src or test assigns it.

Tests

  • New test/test_campaign_facades_base_public_names.py:
    • freezes each S2 facade's base public project names, with the audit ID per facade, and asserts every one resolves;
    • asserts in a fresh interpreter that each re-bound name is its home module's own object.
  • test_pod_runtime_refactor_facade.py: every public _EXPORTS name is declared under TYPE_CHECKING.
  • test_source_providers_refactor_facade.py: every public _EXPORTS name is declared under TYPE_CHECKING, imported from the owner the table names.
  • test_autonudge_refactor_contract.py: _load reads a reassigned store path, and _path cannot be assigned.
  • Local run: scripts/local-gate.py --dry-run selected 2,001 backend and 1,427 frontend related files, more than this shared host runs whole. A targeted set of 46 files ran instead: the touched facades' contract, create-guard and owner tests, test_mirrored_owner_storage.py, test_lazy_reexport_storage.py, test_security_posture.py, test_spawn_audit.py, the boot smoke and the AutoNudge _path readers. Result: 4,308 passed, 13 skipped, 5 xfailed. mypy over src/kiro_crew is clean (1,996 files), and the CI semgrep rule set in baseline mode reports 0 new findings. CI runs the broad set.
  • Mutation proof: with the src changes reverted and the tests kept, all 10 new checks fail (6 facade resolutions, the identity probe, both TYPE_CHECKING pins and the _path test). A mypy probe that reads every public forwarded name shows 62 attr-defined errors at base (37 + 19 + 6) and 0 after the change.

Manual verification

N/A: unit coverage is sufficient, because the change adds bindings and type-only imports and nothing runs differently.

Related Issues

Closes campaign audit items X-08 (B09-1, B11-1, B12-2, I02-F3, I06-F2, I10-F1, and the D6 gatewayd.py docstring), X-09 (B06, B01) and B12-3. No claimed item was skipped.

Follow-ups this PR leaves alone, all of them false or partial before this PR too:

  • mcp_gateway/daemon/__init__.py and autonudge_service/__init__.py say "every name an owner holds is re-exported"; both are outside the claim.
  • The cron, Ops Mission Control and Issue Radar facades have no corpus-derived patch-reach guard (S9/M5).
  • The source-providers bare-global pin skips the TYPE_CHECKING block.

Checklist

  • At most two commits (one is the norm), with a Conventional Commits title
  • Existing tests pass and new tests added for new functionality
  • Self-review completed; code follows project style guidelines
  • Documentation updated (if applicable)
  • No secrets, credentials, or internal references in the diff

Contribution License Agreement

@bolichen97
bolichen97 requested a review from a team as a code owner October 1, 2026 02:07
@github-actions github-actions Bot added the readiness: checking Automated validation is still running label Oct 1, 2026
@github-actions

github-actions Bot commented Oct 1, 2026 •

Copy link
Copy Markdown
Contributor

Design Review (Fable 5) — ✅ PASS

Design-level review of 55a88bdc0611bf9c50d1c95bd9854a16a3736dc7 — updated in place on each push. A BLOCK verdict blocks PR readiness; PASS/CONCERNS are advisory.

Design-Verdict: PASS

Verify the frozen _BASE_PUBLIC_NAMES inventories were dumped from the pre-split modules, not hand-reconstructed — the pin ratifies any name still missing.

Suggestions

  • The seam/not-a-seam split now exists only as comment convention until S9/M5 lands; land that guard soon, since a patch on a not a seam name silently never reaches its owner.

[DESIGN-REVIEWED] 55a88bd

Description read

[DESCRIPTION-READ] a487f41ce49f81af541cd5830aa7c68d07c810aea121ad0d40f4dd9dd9411403

That sha256 names everything this verdict read: the PR description, plus the 1 evidence file(s) the workflow collected from it and handed this review. It covers that evidence because the media strip replaces every attachment URL with the same placeholder, so an edit swapping one attachment for another would leave the description's captured text identical while what this review judged changed. Recompute it with .github/scripts/pr-description-capture.sh over the same inputs; if it differs, the description or that evidence changed after this verdict was formed and any finding drawn from either is unproven.

@github-actions

github-actions Bot commented Oct 1, 2026 •

Copy link
Copy Markdown
Contributor

GPT 5.6 Review — ✅ no blocking findings

GPT 5.6 completed its review of 55a88bdc0611bf9c50d1c95bd9854a16a3736dc7 and found no blocking issues.

This comment is updated in place on each push.

Review details

No findings.
[GPT-REVIEWED] 55a88bd

False positive or not applicable? A repository writer can comment:
/ai-review override gpt 55a88bdc0611bf9c50d1c95bd9854a16a3736dc7: <one-sentence reason>

@github-actions

github-actions Bot commented Oct 1, 2026 •

Copy link
Copy Markdown
Contributor

First Principles Review (Fable 5) — 🟡 CONCERNS

Premise-level review of 55a88bdc0611bf9c50d1c95bd9854a16a3736dc7 — why this exists and whether the shipped surface is the smallest honest version. Updated in place on each push. A BLOCK verdict blocks PR readiness; PASS/CONCERNS are advisory.

First-Principles-Verdict: CONCERNS

86 zero-consumer re-exports are restored on the word of an audit this repo does not carry, then frozen as a permanent 500-name inventory pin.

Not justified as shipped

  • Items 1–6, 9, 10 — inherited: the only provenance for restoring the imported names is "quality report §3.2 … rework plan S2, S3 and S8" — grepping the repo for X-08|X-09|B12-3|rework plan finds nothing, and the base docstrings promise less than the PR claims (cron/gatewayd pin only defined/owner names; autonudge only imported names "callers and tests read off it"). Grepping cron., autonudge., gatewayd., routes., client.-dotted reads of the restored names counts 0 non-test consumers, matching the description's own "break nothing on main today". The new frozen test then makes every zero-consumer name a surface the facade "deliberately" offers, which every future facade change must maintain.
    Subtraction: shrink the # noqa: F401 -- re-exported blocks and _BASE_PUBLIC_NAMES/_REBOUND in test_campaign_facades_base_public_names.py to names with a counted consumer; defer the rest.
    Clears when: a counted in-repo consumer or a linked issue showing a broken import of a restored name through its facade, or the zero-consumer re-exports and their inventory rows are deleted.

What this change ships

Inventory (10 items) — 2 justified

Intent: FIX — restore public names six split facades stopped resolving, make mypy see two facades' forwards, and stop AutoNudge holding its store path twice.

  1. acp/client resolves six sandbox/types names again — inherited, zero consumers (see above)
  2. cron.py resolves five imported names again — inherited, zero consumers (see above)
  3. autonudge.py resolves 28 imported names again (plus two F811 locals) — inherited, zero consumers (see above)
  4. Ops Mission Control routes resolve 21 names again — inherited, zero consumers (see above)
  5. Issue Radar routes resolve three names again — inherited, zero consumers (see above)
  6. gatewayd.py resolves 23 names again, docstring reworded — inherited, zero consumers (see above)
  7. mypy now sees all 56 public forwarded names on pod/runtime and source_providers — justified
  8. _path is a live read-only view of the store's path (no longer assignable) — justified
  9. Docstrings/specs now promise "every public imported name", marked not a seam — rides along, widens the base pins' promise (see above)
  10. New 350-line frozen inventory test pinning ~500 names per facade — rides along, pins the inherited surface (see above)

[FIRST-PRINCIPLES-REVIEWED] 55a88bd

Description read

[DESCRIPTION-READ] baed2513038b1daab34128872a335b4679bae582b5f5aeca841fcd26731a0cf4

That sha256 names the PR description this verdict read. Recompute it with .github/scripts/pr-description-capture.sh; if it differs, the description changed after this verdict was formed and any finding drawn from the description is unproven.

@bolichen97

Copy link
Copy Markdown
Contributor Author
  • span=1daafe508f76 Suggestion: guard, not a docstring, should carry the not-a-seam distinction — rebutted (out of goal here)

Disposition: rebutted for this PR. "Not a goal" excludes patch-reach guards for the cron, Ops Mission Control and Issue Radar facades. Adding one is new hardening this PR's goal does not need.
gatewayd already has one: test_no_owner_reads_a_patched_name_as_its_own_global derives every patched name from the suite and fails on a patch of any of its 23 re-bound names. acp/client forwards its six, so a patch through it reaches the reader.
The remaining guard work (S9/M5) is queued with the refactor campaign's lead as a separate change. Until then, each plain facade's docstring states that its re-exports are not seams.

@bolichen97

Copy link
Copy Markdown
Contributor Author
  • span=bfd3b66d6cd8 Item 2: zero-reader re-bound imports are not justified — rebutted

A facade is the import path for callers this repo cannot grep: builtin and external apps, operator scripts, out-of-tree tests. A zero in-repo count does not show the name is unused.
Main already treats a facade dropping a public name it bound as a defect. #14602 and #14606 re-bound names their splits dropped, and test_source_providers_refactor_facade.py::_HISTORICAL_SURFACE pins imported names, stdlib included.
The rule is now stated in-repo: each touched facade's docstring and spec says its public imports keep resolving. The refactor's owner chose this rule, and this PR applies it.
The two are pragmas on function locals. They change no behaviour, and flake8 stays clean.

@bolichen97

Copy link
Copy Markdown
Contributor Author
  • span=48c74025e761 Item 5: the new inventory duplicates the cron/autonudge contract pins — rebutted

The overlap is partial and the pins differ in kind. cron's and autonudge's _BASE_DEFINED pin DEFINED names, and _BASE_REACHED_IMPORTS pins only imports a caller reads through the facade. Neither pins the imported public names this PR restores, which is the gap the drop went through.
The new file is the one cross-facade pin of each facade's whole pre-split public surface. It covers four facades (acp/client, both app routes, gatewayd) that have no such inventory.
Shrinking it to _REBOUND would leave their other base public imports unpinned. A second edit is needed only when a public name is deliberately removed, and that is when both pins should be reviewed.

@bolichen97
bolichen97 force-pushed the refactor/qf2-facade-name-surface branch from 5a0f249 to 79d02d3 Compare October 1, 2026 02:33
@github-actions github-actions Bot added readiness: action required A blocking check or review needs attention and removed readiness: checking Automated validation is still running labels Oct 1, 2026
@bolichen97

Copy link
Copy Markdown
Contributor Author

CI note for head 79d02d34f. None of the reds below comes from this diff.

Infra, rerun once with --failed. These lanes died on API rate limit exceeded for installation before producing a verdict, between 02:35Z and 02:38Z:

  • UX Review (run 36806439613)
  • GPT 5.6 Review (36806439625)
  • Design Review (36806439638)
  • First Principles Review (36806439803)
  • Opus 5 Review (36806439653)
  • Build / Resolve desktop matrix (36806439661, request A538:2CE812:1A924BA:57BEADC:6ABDC6D4)
  • PR Readiness publish

The lanes logged "could not read this PR's comments/description after 3 attempts".

Unrelated test reds:

  • Backend Tests (Windows) (3), job 110192271848, rerun once: test/test_agent_panel_crew_log.py::test_eviction_ranks_on_fold_order_not_on_a_stamp hit a KeyError after dropped -1272 torn trailing byte(s) from the crew-log store. The test passes on Linux at this head, and the diff touches neither crew_log nor the agent panel.

  • macOS Tests (4), run 36806439928, not rerun because it is deterministic on main:

    • test/test_sandbox_mount_pinned_target.py::test_the_directory_loop_records_every_window_it_binds raises AttributeError: module 'os' has no attribute 'O_PATH'. fix(sandbox): skip a leaf absent under a mask the launcher already placed #15564 added that test without the module's _LINUX_LINK_PIN skip.
    • test/test_terminal_commands.py::TestRunProbe::test_the_environment_is_an_allowlist_not_a_filtered_inherit fails the same way on a checkout of main 7460284 with no change applied.

    The diff touches neither sandbox nor terminal code. This is reported to the campaign lead as a main-red.

@github-actions github-actions Bot added readiness: checking Automated validation is still running and removed readiness: action required A blocking check or review needs attention labels Oct 1, 2026
@bolichen97

Copy link
Copy Markdown
Contributor Author
  • span=b11c0429cf12 Items 1–6: restored re-exports have no counted consumer — rebutted

The rule rests on in-repo precedent, not only the campaign report. #14602 and #14606 on main re-bound names their splits dropped, and test_source_providers_refactor_facade.py::_HISTORICAL_SURFACE pins every imported name that facade bound, stdlib included.
A facade's consumers are mostly outside this repo's grep: builtin and external apps, operator scripts, out-of-tree tests. A split that drops a name breaks them at import time with no in-repo signal, which is why counting in-repo readers is not the test.
The widened docstring and spec sentences state the rule each facade now follows. The B09 sentence quoted is the facade's own choice; the audit flagged that choice, and this change applies the replacement rule the refactor's owner decided.
If a maintainer rules otherwise, the subtraction is self-contained: the six re-export hunks plus the new test. X-09 and the _path fix do not depend on them.

@github-actions github-actions Bot added readiness: action required A blocking check or review needs attention and removed readiness: checking Automated validation is still running labels Oct 1, 2026
@github-actions

github-actions Bot commented Oct 1, 2026 •

Copy link
Copy Markdown
Contributor

Opus 5 Review — ✅ no blocking findings

Reviewed 55a88bdc0611bf9c50d1c95bd9854a16a3736dc7 — this comment is updated in place on each push.

Review details

No findings.

[OPUS-REVIEWED] 55a88bd

Verdict parsed from the review's SHA-scoped output markers for commit 55a88bdc0611bf9c50d1c95bd9854a16a3736dc7.

False positive or not applicable? A repository writer can comment:
/ai-review override fable 55a88bdc0611bf9c50d1c95bd9854a16a3736dc7: <one-sentence reason>

@github-actions github-actions Bot added readiness: passed Eligible automated validation passed for the current revision and removed readiness: action required A blocking check or review needs attention labels Oct 1, 2026
@bolichen97
bolichen97 enabled auto-merge (squash) October 1, 2026 05:11
@bolichen97
bolichen97 force-pushed the refactor/qf2-facade-name-surface branch from 79d02d3 to db9749f Compare October 1, 2026 07:13
@github-actions github-actions Bot added readiness: checking Automated validation is still running readiness: action required A blocking check or review needs attention and removed readiness: passed Eligible automated validation passed for the current revision readiness: checking Automated validation is still running labels Oct 1, 2026
Six split facades stopped resolving public kiro_crew names they bound
before their owners moved out, two forwarding facades hid most public
re-exports from mypy, and AutoNudgeService held its store path twice.

- acp/client.py, cron.py, autonudge.py, the Ops Mission Control and
  Issue Radar routes and mcp_gateway/gatewayd.py bind again every such
  name (86 in all); the client forwards its six to transport_errors,
  the module that reads them. The plain facades' docstrings say the
  re-exports are not patch seams, and gatewayd's states exactly which
  names it re-exports instead of every name an owner holds.
- pod/runtime.py and the source-providers handler import every public
  forwarded name under TYPE_CHECKING (37 and 19).
- AutoNudgeService._path is a read-only view of the store's path, so
  the loader reads the file the store writes.

A repo-wide test freezes each facade's base public names and checks
every re-bound name is its home module's object in a fresh
interpreter; the two facades' contract tests pin the TYPE_CHECKING
coverage, and the AutoNudge contract test pins the path view.
@bolichen97
bolichen97 force-pushed the refactor/qf2-facade-name-surface branch from db9749f to 55a88bd Compare October 1, 2026 08:34
@github-actions github-actions Bot removed the readiness: action required A blocking check or review needs attention label Oct 1, 2026
@github-actions github-actions Bot added the readiness: checking Automated validation is still running label Oct 1, 2026
@bolichen97

Copy link
Copy Markdown
Contributor Author
  • span=ea7b0343d1a1 Suggestion: land the S9/M5 guard soon so not-a-seam is enforced, not commented — rebutted (out of goal here)

Same class as span 1daafe508f76 on 5a0f249, and the same ruling applies. This PR's frozen "Not a goal" excludes the cron, Ops Mission Control and Issue Radar patch-reach guards.
Where a guard exists it already enforces the rule: gatewayd's suite-derived test_no_owner_reads_a_patched_name_as_its_own_global fails on a patch of any of its 23 not-a-seam names. acp/client forwards its six, so those patches reach the reader.
The remaining guard is queued with the refactor campaign's lead as its own change.

@bolichen97

Copy link
Copy Markdown
Contributor Author
  • span=afc45494f7d1 Items 1–6, 9, 10: zero-consumer re-exports and their frozen inventory — rebutted

Same class as spans bfd3b66d6cd8, 48c74025e761 and b11c0429cf12, and the same ruling applies. Re-binding what a facade bound before its split follows main's own precedent: #14602 and #14606 re-bound names their splits dropped, and test_source_providers_refactor_facade.py::_HISTORICAL_SURFACE pins imported names, stdlib included.
A facade's importers include apps, operator scripts and out-of-tree tests that no in-repo grep counts, so a zero count is not evidence the name is unused. The inventory exists to make a future removal deliberate rather than silent.
The rule and its scope are the refactor owner's decision, and this PR applies it. If a maintainer rules otherwise, the subtraction is the six re-export hunks plus the new test; X-09 and the _path fix do not depend on them.

@github-actions github-actions Bot added readiness: passed Eligible automated validation passed for the current revision and removed readiness: checking Automated validation is still running labels Oct 1, 2026

@buluoray buluoray left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Approve at 55a88bdc0611bf9c50d1c95bd9854a16a3736dc7.

This PR restores the public names that six split facades dropped, makes the TYPE_CHECKING blocks of two forwarding facades cover their whole public _EXPORTS, and turns AutoNudgeService._path into a live read-only view of _store.path. Without it, an import of a name these facades used to offer raises AttributeError, 56 forwarded rt./sp. reads fail mypy with attr-defined, and _load keeps reading the old file after the store's path is reassigned.

What I verified at this head

  • The inventory is complete and correct. I derived each facade's top-level public kiro_crew imports independently, from the parent of its split commit (#15153, #14615, #14994, #14537, #14545, #14976). Names that did not resolve at base: acp/client 6, cron 5, autonudge 28, OMC routes 21, Issue Radar routes 3, gatewayd 23. That is exactly the 86 the PR re-binds, and all of them resolve at head.
  • Patches through the acp/client facade still reach the owner. monkeypatch.setattr(client, <name>, sentinel) reaches transport_errors.<name> for all six forwarded names, and undo restores it. The forwarding goes through _ReExportModule.__setattr__.
  • No current test is broken by the new plain-facade re-exports. I grepped test/ and src/ for patches of every re-bound plain-facade name (string targets, setattr, patch.object) and found none, so no existing test patch silently misses its owner. The gatewayd guard test_no_owner_reads_a_patched_name_as_its_own_global also catches such a patch: a temporary test patching gatewayd.hash_command turned it red (daemon/launch.py:352: hash_command).
  • mypy. mypy on the 8 touched src files reports no issues. A probe that reads every public _EXPORTS name of pod/runtime and source_providers gets 56 attr-defined errors at base (37 + 19, which matches the added TYPE_CHECKING names) and 0 at head. flake8 is clean on the touched files.
  • _path is never assigned on AutoNudge. The only svc._path = ... writes in test/ are on CronService (test_portability.py) and the new negative assertion. The three AutoNudgeService.__new__ sites in test_autonudge_wake_judge.py do not touch _path, and that file passes (293).
  • Targeted tests pass. The new and changed facade tests pass (504), as do the gatewayd composition and cron contract tests (123) and the acp facade, Issue Radar seams/surface and OMC ledger tests (426).

Mutation checks (each one reddened, then restored)

  • I put back self._path = self._store.path and removed the property. test_autonudge_refactor_contract.py::...::test_the_loader_reads_the_path_the_store_holds_now failed.
  • I dropped gatewayd's BackendUnavailable re-export. test_every_base_public_name_resolves_on_its_facade[kiro_crew.mcp_gateway.gatewayd] and test_a_rebound_name_is_the_object_its_home_module_holds failed.
  • I dropped loopback_owner_pids from pod/runtime's TYPE_CHECKING block. test_every_public_reexport_is_named_for_type_checkers failed.
  • I dropped source_providers' sanitize TYPE_CHECKING import. test_type_checkers_resolve_every_public_forwarded_name_from_its_owner failed.

Red "Publish readiness signal" check

This is not a code issue. Run 36837121314 (09:34Z) died on API rate limit exceeded for installation (core limit 0/15000) at gh api repos/.../pulls/15675, before it evaluated anything. A later evaluation published the PR Readiness commit status as success at 10:00:08Z ("Eligible automated validation passed for this revision") and applied readiness: passed at 10:00:11Z. The red check-run is the stale rate-limited attempt. All other check-runs are success (85) or skipped (11).

Bot dispositions

  • Design (PASS), "verify the inventories were dumped from the pre-split modules": checked. My independent derivation from the split parents matches the re-bound set exactly. Does not gate.
  • Design suggestion, "land the S9/M5 guard soon": gatewayd already has a working guard (confirmed by mutation above). cron, autonudge, OMC and Issue Radar have none: a future patch of, say, kiro_crew.cron.platform_compat would silently not reach the owner. No test does that today. This is a follow-up, and the PR declares it out of scope. Does not gate.
  • First Principles (CONCERNS), "zero-consumer re-exports and a frozen inventory": this is a scope objection, not a defect. The restored names are exactly what each facade bound before its split, which follows the #14602/#14606 precedent. Does not gate.
  • GPT 5.6 and Opus 5: no findings.

Non-blocking follow-ups

  • Add the S9/M5 patch-reach guard for the cron, autonudge, Ops Mission Control and Issue Radar facades, so the not a seam rule is enforced by a test rather than by docstrings.
  • Ops Mission Control routes.py now imports kiro_crew.cron at import time (CronStoreUnreadable), as it did before the split. This is fine; I note it only as an import-graph edge.

The squash commit body is accurate: 86 names, and 37 + 19 TYPE_CHECKING names.

@bolichen97
bolichen97 merged commit 35171e2 into main Oct 1, 2026
103 of 111 checks passed
@bolichen97
bolichen97 deleted the refactor/qf2-facade-name-surface branch October 1, 2026 15:21
@github-actions github-actions Bot removed the readiness: passed Eligible automated validation passed for the current revision label Oct 1, 2026
bolichen97 added a commit that referenced this pull request Oct 2, 2026
…n main) (#16032)

#12781 added a function-local 'from kiro_crew import probes' in
_is_work_ledger_monitor while #15675 bound probes at module level as a
re-export, so main's Backend Lint fails with F811 at autonudge.py:1797.
Read probes off the module binding and drop the now-stale lazy-import
rationale from the docstring. No behaviour change.

Co-authored-by: w <w@k.local>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants