From 5041813e2ca97ed013c4441b67a55752ae130d0f Mon Sep 17 00:00:00 2001
From: Seongho Bae <me@seonghobae.me>
Date: Wed, 2 Sep 2026 19:19:54 +0900
Subject: [PATCH 01/10] fix(noema): remove unsupported local job-deadline
 policy

---
 .../source_fix_pr1715_no_model_job_timeout.py | 102 ++++++++++--------
 1 file changed, 58 insertions(+), 44 deletions(-)

diff --git a/scripts/ci/source_fix_pr1715_no_model_job_timeout.py b/scripts/ci/source_fix_pr1715_no_model_job_timeout.py
index 497d109678..b37e76353b 100644
--- a/scripts/ci/source_fix_pr1715_no_model_job_timeout.py
+++ b/scripts/ci/source_fix_pr1715_no_model_job_timeout.py
@@ -1,4 +1,4 @@
-"""One-shot exact-head repair for PR #1715's Noema model timeout contract."""
+"""One-shot exact-head repair for the merged PR #1715 timeout-policy regression."""
 
 from __future__ import annotations
 
@@ -15,14 +15,28 @@ def replace_once(text: str, old: str, new: str, label: str) -> str:
     """Replace one literal block and fail closed when branch contents moved."""
     count = text.count(old)
     if count != 1:
-        raise SystemExit(f"PR1715 {label}: expected one literal block, found {count}")
+        raise SystemExit(f"PR1715 successor {label}: expected one literal block, found {count}")
     return text.replace(old, new, 1)
 
 
 def patch_workflow() -> None:
-    """Keep bounded cleanup but remove elapsed-time authority from model work."""
+    """Remove repository-authored elapsed-time termination from both Noema jobs."""
     text = WORKFLOW.read_text(encoding="utf-8")
-    old = '''    # Bound this job well short of GitHub's 360-minute platform default. Its
+    cleanup_old = '''    # Bound this job well short of GitHub's 360-minute platform default. Its
+    # only step is a single-repository, status-filtered gh api --paginate
+    # list-and-cancel sweep (up to 3 passes x 5 statuses), no branch update
+    # or merge -- lighter than pr-review-merge-scheduler.yml's scan-pr-queue
+    # job (PR #1702), which got timeout-minutes: 30 for a comparable
+    # single-repo scan that also dispatches a review and updates a branch.
+    timeout-minutes: 20
+'''
+    cleanup_new = '''    # No repository-authored wall-clock cutoff: this housekeeping job relies
+    # on GitHub Actions' platform execution contract plus explicit API failure.
+    # A shorter local deadline would be an unsupported hand-selected policy.
+'''
+    text = replace_once(text, cleanup_old, cleanup_new, "cleanup timeout block")
+
+    model_old = '''    # Bound this job well short of GitHub's 360-minute platform default. Its
     # "Prepare Noema model verdict" step calls into two_phase.py's call_llm
     # via the same contextual-orchestrator gateway whose unbounded wait was
     # confirmed to stall runs for 7-20 hours in opencode-review.yml before
@@ -38,69 +52,69 @@ def patch_workflow() -> None:
     # provisioning, publication), while staying well under GitHub's default.
     timeout-minutes: 210
 '''
-    new = '''    # Model-backed Noema intentionally has no job-level wall-clock timeout.
-    # contextual-orchestrator/orchestrator/free owns provider termination;
-    # GitHub admission must not stop reasoning, streaming, or tool work only
-    # because elapsed time crossed a repository-side deadline. Stale heads,
-    # closed/draft PRs, provider completion, and explicit cancellation remain
-    # authoritative termination signals. The non-model cleanup job above is
-    # independently bounded because it performs only GitHub API housekeeping.
+    model_new = '''    # Model-backed Noema intentionally has no repository-authored wall-clock
+    # timeout. contextual-orchestrator/orchestrator/free owns provider routing
+    # and normal model completion; live PR/head state and explicit cancellation
+    # remain authoritative stop signals. GitHub's platform execution ceiling is
+    # an external runtime constraint, not a model-selection or compute policy.
 '''
-    WORKFLOW.write_text(
-        replace_once(text, old, new, "model job timeout block"), encoding="utf-8"
-    )
+    text = replace_once(text, model_old, model_new, "model job timeout block")
+    WORKFLOW.write_text(text, encoding="utf-8")
 
 
 def patch_test() -> None:
-    """Replace the stale timeout-positive assertion with the owner contract."""
+    """Replace timeout-positive regressions with no-local-deadline contracts."""
     text = TEST.read_text(encoding="utf-8")
-    marker = "def test_noema_review_job_has_a_bounded_runtime_above_the_two_hour_model_allowance() -> None:\n"
+    marker = "def test_cancel_closed_pr_runs_has_a_bounded_runtime() -> None:\n"
     start = text.find(marker)
     if start < 0 or text.find(marker, start + 1) >= 0:
-        raise SystemExit("PR1715 stale model-timeout test marker moved or duplicated")
-    replacement = '''def test_noema_review_model_job_has_no_elapsed_time_termination() -> None:
-    """Model-backed Noema delegates termination to orchestrator/provider authority."""
+        raise SystemExit("PR1715 successor timeout-test marker moved or duplicated")
+    replacement = '''def test_noema_jobs_do_not_invent_repository_wall_clock_deadlines() -> None:
+    """Noema model/support jobs must not encode hand-selected elapsed-time cutoffs."""
     workflow = workflow_text("noema-review.yml")
-    job = workflow.split("  noema-review:\\n", 1)[1]
+    cleanup = workflow.split("  cancel-closed-pr-runs:\\n", 1)[1].split(
+        "\\n  noema-review:\\n", 1
+    )[0]
+    model = workflow.split("  noema-review:\\n", 1)[1]
 
-    assert re.search(r"^    timeout-minutes:", job, flags=re.MULTILINE) is None
+    assert re.search(r"^    timeout-minutes:", cleanup, flags=re.MULTILINE) is None
+    assert re.search(r"^    timeout-minutes:", model, flags=re.MULTILINE) is None
     assert "contextual-orchestrator/orchestrator/free" in workflow
-    assert "Model-backed Noema intentionally has no job-level wall-clock timeout" in job
-    assert "timeout-minutes: 20" in workflow.split(
-        "  cancel-closed-pr-runs:\\n", 1
-    )[1].split("\\n  noema-review:\\n", 1)[0]
+    assert "unsupported hand-selected policy" in cleanup
+    assert "no repository-authored wall-clock" in model
 '''
     TEST.write_text(text[:start] + replacement, encoding="utf-8")
 
 
 def append_traceability() -> None:
-    """Record why support housekeeping may be bounded while model work may not."""
-    changelog_note = (
-        "\n- PR #1715: keep the non-model Noema close-cleanup job bounded, but remove "
-        "the proposed 210-minute job timeout from model-backed `noema-review`; "
-        "`orchestrator/free`/provider completion, live PR/head state, or explicit "
-        "cancellation are the termination authorities rather than elapsed time.\n"
-    )
+    """Record the causal owner and replacement authority."""
     changelog = CHANGELOG.read_text(encoding="utf-8")
-    if "PR #1715: keep the non-model Noema close-cleanup job bounded" not in changelog:
-        CHANGELOG.write_text(changelog + changelog_note, encoding="utf-8")
+    note = (
+        "\n- Successor to merged PR #1715: remove repository-authored 20/210-minute "
+        "Noema workflow deadlines. GitHub's platform runtime contract governs ordinary "
+        "Actions execution; contextual-orchestrator/provider completion, exact live state, "
+        "and explicit cancellation govern model work. No inferred local buffer remains.\n"
+    )
+    if "Successor to merged PR #1715: remove repository-authored 20/210-minute" not in changelog:
+        CHANGELOG.write_text(changelog + note, encoding="utf-8")
 
-    baseline_note = '''
+    baseline = BASELINE.read_text(encoding="utf-8")
+    section = '''
 
-### Noema model-job timeout authority — PR #1715
+### Noema workflow elapsed-time authority — PR #1715 successor
 
-- **Root cause:** a queue-operability repair proposed `timeout-minutes: 210` on the model-backed `noema-review` job, turning elapsed wall time into an admission/model termination authority.
-- **Contract:** the lightweight closed-PR Actions cleanup remains bounded, while Noema model work has no repository-owned wall-clock cutoff. `orchestrator/free` and its upstream provider own normal model completion; live PR/head validation, provider end, or explicit cancellation remain authoritative stop conditions.
-- **Regression:** `test_noema_review_model_job_has_no_elapsed_time_termination` rejects a job-level timeout on the model job while retaining the 20-minute bound on non-model cleanup.
-- **Status:** Implemented on the PR #1715 writer branch; exact-head CI/review must be regenerated after the one-shot repair commit.
+- **Live gap:** merged PR #1715 introduced local `timeout-minutes: 20` and `timeout-minutes: 210`. The former was inferred by analogy to another queue job; the latter combined an inherited 180-minute allowance with an invented 30-minute buffer. Neither value had executable mathematical, statistical, psychometric, standards, or experimentally validated provenance.
+- **Causal owner:** `.github/.github/workflows/noema-review.yml`; contextual-orchestrator remains owner of model routing/provider completion and test-time-compute policy.
+- **Repair:** remove both repository-authored deadlines. Non-model housekeeping uses GitHub Actions' documented platform execution contract and API failures; model-backed Noema additionally uses exact PR/head state, explicit cancellation, and contextual-orchestrator/provider completion. No paid/provider/model fallback is introduced.
+- **Regression:** `test_noema_jobs_do_not_invent_repository_wall_clock_deadlines` rejects local timeout policy on both jobs and retains the exact `orchestrator/free` routing contract.
+- **Status:** Proposed until this one-shot self-removes and fresh exact-head required Checks are GREEN.
 '''
-    baseline = BASELINE.read_text(encoding="utf-8")
-    if "### Noema model-job timeout authority — PR #1715" not in baseline:
-        BASELINE.write_text(baseline + baseline_note, encoding="utf-8")
+    if "### Noema workflow elapsed-time authority — PR #1715 successor" not in baseline:
+        BASELINE.write_text(baseline + section, encoding="utf-8")
 
 
 def main() -> None:
-    """Apply the minimal owner repair and its permanent regression/docs."""
+    """Apply production, regression, and traceability repairs."""
     patch_workflow()
     patch_test()
     append_traceability()

From d46060a542e64b0d866900c82c28284fa30e9e03 Mon Sep 17 00:00:00 2001
From: Seongho Bae <me@seonghobae.me>
Date: Wed, 2 Sep 2026 19:20:21 +0900
Subject: [PATCH 02/10] ci(noema): retarget corrective source repair after
 premature merge

---
 .../source-fix-pr1715-no-model-job-timeout.yml       | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/.github/workflows/source-fix-pr1715-no-model-job-timeout.yml b/.github/workflows/source-fix-pr1715-no-model-job-timeout.yml
index 0d733b2b72..25ad3599c2 100644
--- a/.github/workflows/source-fix-pr1715-no-model-job-timeout.yml
+++ b/.github/workflows/source-fix-pr1715-no-model-job-timeout.yml
@@ -3,7 +3,7 @@ name: Source Fix PR 1715 No Model Job Timeout
 on:
   push:
     branches:
-      - fix/noema-review-job-timeout-minutes
+      - fix/noema-review-no-heuristic-timeouts
     paths:
       - scripts/ci/source_fix_pr1715_no_model_job_timeout.py
       - .github/workflows/source-fix-pr1715-no-model-job-timeout.yml
@@ -30,7 +30,7 @@ jobs:
         shell: bash
         run: |
           set -euo pipefail
-          remote_head="$(git ls-remote origin refs/heads/fix/noema-review-job-timeout-minutes | cut -f1)"
+          remote_head="$(git ls-remote origin refs/heads/fix/noema-review-no-heuristic-timeouts | cut -f1)"
           test -n "$remote_head"
           test "$remote_head" = "$GITHUB_SHA"
 
@@ -79,7 +79,7 @@ jobs:
           test -z "$bad"
           test ! -e scripts/ci/source_fix_pr1715_no_model_job_timeout.py
           test ! -e .github/workflows/source-fix-pr1715-no-model-job-timeout.yml
-          remote_head="$(git ls-remote origin refs/heads/fix/noema-review-job-timeout-minutes | cut -f1)"
+          remote_head="$(git ls-remote origin refs/heads/fix/noema-review-no-heuristic-timeouts | cut -f1)"
           test "$remote_head" = "$GITHUB_SHA"
 
       - name: Publish normal non-force repair commit
@@ -94,12 +94,12 @@ jobs:
             echo "::error::No workflow-starting mutation credential is configured; refusing github.token publication."
             exit 1
           fi
-          remote_head="$(git ls-remote origin refs/heads/fix/noema-review-job-timeout-minutes | cut -f1)"
+          remote_head="$(git ls-remote origin refs/heads/fix/noema-review-no-heuristic-timeouts | cut -f1)"
           test "$remote_head" = "$GITHUB_SHA"
           git config user.name "github-actions[bot]"
           git config user.email "41898282+github-actions[bot]@users.noreply.github.com"
           git add -A
           git diff --cached --check
-          git commit -m "fix(noema-review): remove model wall-clock termination"
+          git commit -m "fix(noema-review): remove unsupported local job deadlines"
           git remote set-url origin "https://x-access-token:${workflow_push_token}@github.com/${GITHUB_REPOSITORY}.git"
-          git push origin HEAD:fix/noema-review-job-timeout-minutes
+          git push origin HEAD:fix/noema-review-no-heuristic-timeouts

From 3ae0213890104e9cdea5e9a6ec89bbfad5371645 Mon Sep 17 00:00:00 2001
From: Seongho Bae <me@seonghobae.me>
Date: Wed, 2 Sep 2026 19:21:50 +0900
Subject: [PATCH 03/10] test(noema): encode null model timeout authority before
 repair

---
 tests/test_noema_model_timeout_policy.py | 30 ++++++++++++++++++++++++
 1 file changed, 30 insertions(+)
 create mode 100644 tests/test_noema_model_timeout_policy.py

diff --git a/tests/test_noema_model_timeout_policy.py b/tests/test_noema_model_timeout_policy.py
new file mode 100644
index 0000000000..bdbb1de7d0
--- /dev/null
+++ b/tests/test_noema_model_timeout_policy.py
@@ -0,0 +1,30 @@
+"""Regression contract for Noema model-execution timeout ownership."""
+
+from __future__ import annotations
+
+import re
+from pathlib import Path
+
+
+REPOSITORY_ROOT = Path(__file__).resolve().parents[1]
+NOEMA_WORKFLOW_PATH = REPOSITORY_ROOT / ".github" / "workflows" / "noema-review.yml"
+
+
+def _noema_model_job_text() -> str:
+    """Return the Noema model-bearing job from the trusted workflow source."""
+    workflow_text = NOEMA_WORKFLOW_PATH.read_text(encoding="utf-8")
+    return workflow_text.split("  noema-review:\n", 1)[1]
+
+
+def test_noema_model_job_has_no_elapsed_time_termination() -> None:
+    """Keep model reasoning free of a GitHub job wall-clock termination policy."""
+    model_job_text = _noema_model_job_text()
+
+    assert "Prepare Noema model verdict" in model_job_text
+    assert re.search(
+        r"^    timeout-minutes:\s*\d+\s*$", model_job_text, flags=re.MULTILINE
+    ) is None, (
+        "Noema model execution must not be terminated by elapsed time; user cancel, "
+        "provider termination, and an explicitly configured contextual-orchestrator "
+        "admin timeout are the only timeout authorities."
+    )

From 3882da7946e55ecb17632e08345fcda6dd04c1f1 Mon Sep 17 00:00:00 2001
From: Seongho Bae <me@seonghobae.me>
Date: Wed, 2 Sep 2026 19:22:44 +0900
Subject: [PATCH 04/10] fix(noema): integrate RED-first and governance
 authority repair

---
 .../source_fix_pr1715_no_model_job_timeout.py | 213 +++++++++++++++---
 1 file changed, 185 insertions(+), 28 deletions(-)

diff --git a/scripts/ci/source_fix_pr1715_no_model_job_timeout.py b/scripts/ci/source_fix_pr1715_no_model_job_timeout.py
index b37e76353b..d041a29e63 100644
--- a/scripts/ci/source_fix_pr1715_no_model_job_timeout.py
+++ b/scripts/ci/source_fix_pr1715_no_model_job_timeout.py
@@ -8,7 +8,10 @@
 WORKFLOW = Path(".github/workflows/noema-review.yml")
 TEST = Path("tests/test_noema_orchestrator_workflow_contract.py")
 CHANGELOG = Path("CHANGELOG.md")
+ARCHITECTURE = Path("ARCHITECTURE.md")
+ADR = Path("docs/adr/0003-contextual-orchestrator-vendored-free-zdr.md")
 BASELINE = Path("docs/product-technical-gap-baseline.md")
+DOCTORING = Path("docs/doctoring/noema-model-timeout-authority-2026-09-02.md")
 
 
 def replace_once(text: str, old: str, new: str, label: str) -> str:
@@ -19,6 +22,14 @@ def replace_once(text: str, old: str, new: str, label: str) -> str:
     return text.replace(old, new, 1)
 
 
+def append_once(path: Path, marker: str, section: str) -> None:
+    """Append one governance section while refusing duplicate doctoring."""
+    text = path.read_text(encoding="utf-8")
+    if marker in text:
+        return
+    path.write_text(text.rstrip() + "\n\n" + section.strip() + "\n", encoding="utf-8")
+
+
 def patch_workflow() -> None:
     """Remove repository-authored elapsed-time termination from both Noema jobs."""
     text = WORKFLOW.read_text(encoding="utf-8")
@@ -31,8 +42,9 @@ def patch_workflow() -> None:
     timeout-minutes: 20
 '''
     cleanup_new = '''    # No repository-authored wall-clock cutoff: this housekeeping job relies
-    # on GitHub Actions' platform execution contract plus explicit API failure.
-    # A shorter local deadline would be an unsupported hand-selected policy.
+    # on GitHub Actions' external platform execution contract plus explicit API
+    # failure. A shorter local deadline would be an unsupported hand-selected
+    # policy rather than a measured queue-control invariant.
 '''
     text = replace_once(text, cleanup_old, cleanup_new, "cleanup timeout block")
 
@@ -54,9 +66,10 @@ def patch_workflow() -> None:
 '''
     model_new = '''    # Model-backed Noema intentionally has no repository-authored wall-clock
     # timeout. contextual-orchestrator/orchestrator/free owns provider routing
-    # and normal model completion; live PR/head state and explicit cancellation
-    # remain authoritative stop signals. GitHub's platform execution ceiling is
-    # an external runtime constraint, not a model-selection or compute policy.
+    # and configured model timeout policy; provider termination, live PR/head
+    # invalidation, and explicit user/operator cancellation remain authoritative
+    # stop signals. GitHub's platform execution ceiling is an external capacity
+    # constraint, not a model-selection or test-time-compute policy.
 '''
     text = replace_once(text, model_old, model_new, "model job timeout block")
     WORKFLOW.write_text(text, encoding="utf-8")
@@ -80,44 +93,188 @@ def patch_test() -> None:
     assert re.search(r"^    timeout-minutes:", cleanup, flags=re.MULTILINE) is None
     assert re.search(r"^    timeout-minutes:", model, flags=re.MULTILINE) is None
     assert "contextual-orchestrator/orchestrator/free" in workflow
-    assert "unsupported hand-selected policy" in cleanup
+    assert "unsupported hand-selected" in cleanup
     assert "no repository-authored wall-clock" in model
 '''
     TEST.write_text(text[:start] + replacement, encoding="utf-8")
 
 
-def append_traceability() -> None:
-    """Record the causal owner and replacement authority."""
+def patch_governance() -> None:
+    """Synchronize change history, architecture, ADR, doctoring, and product-gap status."""
     changelog = CHANGELOG.read_text(encoding="utf-8")
-    note = (
-        "\n- Successor to merged PR #1715: remove repository-authored 20/210-minute "
-        "Noema workflow deadlines. GitHub's platform runtime contract governs ordinary "
-        "Actions execution; contextual-orchestrator/provider completion, exact live state, "
-        "and explicit cancellation govern model work. No inferred local buffer remains.\n"
+    entry = (
+        "- **Correct PR #1715 timeout authority regression.** Remove the inferred "
+        "20-minute Noema cleanup deadline and the model-bearing 210-minute job "
+        "deadline. Neither value had standards, measured-runtime, or experimentally "
+        "validated provenance. The model-bearing deadline additionally contradicted "
+        "ADR-0003 by allowing elapsed time alone to terminate contextual-orchestrator "
+        "reasoning. GitHub's hosting ceiling remains an external capacity constraint; "
+        "provider end, live-head supersession, explicit user/operator cancellation, "
+        "or an explicitly configured contextual-orchestrator administrative timeout "
+        "are the intentional model termination authorities.\n"
     )
-    if "Successor to merged PR #1715: remove repository-authored 20/210-minute" not in changelog:
-        CHANGELOG.write_text(changelog + note, encoding="utf-8")
+    if entry.strip() not in changelog:
+        changelog = replace_once(
+            changelog,
+            "## [Unreleased]\n",
+            "## [Unreleased]\n" + entry,
+            "CHANGELOG Unreleased header",
+        )
+        CHANGELOG.write_text(changelog, encoding="utf-8")
 
-    baseline = BASELINE.read_text(encoding="utf-8")
-    section = '''
+    append_once(
+        ARCHITECTURE,
+        "## Model execution timeout authority (2026-09-02)",
+        '''## Model execution timeout authority (2026-09-02)
 
-### Noema workflow elapsed-time authority — PR #1715 successor
+Central model-backed review jobs (`OpenCode`, `Noema`, and `Strix`) delegate model
+selection and inference lifecycle to `ContextualWisdomLab/contextual-orchestrator`.
+The GitHub Actions job that contains reasoning, streaming, or tool execution must
+not invent a fixed elapsed-time ceiling. Intentional termination authorities are
+an explicit user/operator cancellation, provider termination, superseded-head
+retirement after live-head validation, or an explicitly configured
+contextual-orchestrator administrative timeout. Queue/runner hygiene must be
+expressed as measured control-plane contracts rather than hand-selected model
+runtime budgets. GitHub's hosting limit remains an external capacity constraint,
+not `.github` model policy.''',
+    )
 
-- **Live gap:** merged PR #1715 introduced local `timeout-minutes: 20` and `timeout-minutes: 210`. The former was inferred by analogy to another queue job; the latter combined an inherited 180-minute allowance with an invented 30-minute buffer. Neither value had executable mathematical, statistical, psychometric, standards, or experimentally validated provenance.
-- **Causal owner:** `.github/.github/workflows/noema-review.yml`; contextual-orchestrator remains owner of model routing/provider completion and test-time-compute policy.
-- **Repair:** remove both repository-authored deadlines. Non-model housekeeping uses GitHub Actions' documented platform execution contract and API failures; model-backed Noema additionally uses exact PR/head state, explicit cancellation, and contextual-orchestrator/provider completion. No paid/provider/model fallback is introduced.
-- **Regression:** `test_noema_jobs_do_not_invent_repository_wall_clock_deadlines` rejects local timeout policy on both jobs and retains the exact `orchestrator/free` routing contract.
-- **Status:** Proposed until this one-shot self-removes and fresh exact-head required Checks are GREEN.
-'''
-    if "### Noema workflow elapsed-time authority — PR #1715 successor" not in baseline:
-        BASELINE.write_text(baseline + section, encoding="utf-8")
+    append_once(
+        ADR,
+        "2026-09-02 correction: PR #1715's Noema deadlines were not a new timeout decision",
+        '''- **2026-09-02 correction: PR #1715's Noema deadlines were not a new timeout decision.**
+  The accepted 2026-08-31 amendment above already states that OpenCode, Noema,
+  Strix, and contextual-orchestrator inference have no repository/application
+  fixed wall-clock timeout. PR #1715 added `timeout-minutes: 20` to the
+  housekeeping job by analogy to another queue job and `timeout-minutes: 210`
+  to the model-bearing Noema job by combining an inherited allowance with an
+  invented buffer. Neither local number had standards, measured-runtime, or
+  experimentally validated provenance, and the model-bearing value directly
+  contradicted this ADR. Both local deadlines are therefore removed. Provider
+  termination, exact live-head invalidation, explicit user/operator cancellation,
+  or an explicitly configured contextual-orchestrator administrative timeout
+  remain intentional model termination authorities; GitHub's hosting ceiling is
+  documented as an external runtime constraint rather than repository policy.''',
+    )
+
+    append_once(
+        BASELINE,
+        "### Noema workflow elapsed-time authority — PR #1715 successor",
+        '''### Noema workflow elapsed-time authority — PR #1715 successor
+
+- **Protected-main incident base:** `5935c8153722fe6b53bafd579b74f8f097303959`
+  (merge of PR #1715).
+- **Live gap:** PR #1715 introduced local `timeout-minutes: 20` and
+  `timeout-minutes: 210`. The former was inferred by analogy to another queue
+  job; the latter combined an inherited allowance with an invented buffer.
+  Neither value had executable standards, measured-runtime, or experimentally
+  validated provenance. The 210-minute value also violated ADR-0003's existing
+  no-fixed-inference-timeout contract.
+- **PRD goal:** preserve correctness-first long-running Noema review without
+  replacing model/provider lifecycle authority with elapsed time.
+- **TRD invariant:** neither Noema job invents a repository-authored local
+  deadline; the model-bearing job uses `contextual-orchestrator/orchestrator/free`
+  and terminates intentionally only on provider end, exact live-state
+  invalidation, explicit user/operator cancellation, or an explicitly configured
+  contextual-orchestrator administrative timeout. GitHub's host ceiling remains
+  an external capacity constraint.
+- **Context Map:** `.github` owns Actions admission, exact-head validation, and
+  stale-run retirement; `ContextualWisdomLab/contextual-orchestrator` owns model
+  routing and configured model timeout policy.
+- **Regression:** `tests/test_noema_model_timeout_policy.py` is committed RED
+  before the source repair and `test_noema_jobs_do_not_invent_repository_wall_clock_deadlines`
+  rejects reintroduction on either job.
+- **Status:** Proposed until the one-shot self-removes and fresh exact-head
+  required Checks are GREEN.
+
+```mermaid
+flowchart LR
+    PR[Exact PR head] --> GH[.github Noema control plane]
+    GH --> CO[ContextualWisdomLab/contextual-orchestrator]
+    CO --> MODEL[orchestrator/free model execution]
+    USER[User/operator cancel] --> GH
+    HEAD[Superseding PR head] --> GH
+    PROVIDER[Provider termination] --> CO
+    ADMIN[Configured CO admin timeout] --> CO
+    GH -. no local elapsed-time deadline .-> MODEL
+```
+
+| Gap | Action | Status |
+| --- | --- | --- |
+| G-NOEMA-TIMEOUT-AUTHORITY | RED→remove unsupported 20/210-minute local deadlines→full GREEN→self-retire repair artifacts→fresh exact-head required Checks. | Proposed / PR #1720 |
+| G-STALE-RUN-CENTRAL-AUTHORITY | PR #1717 must read central `repository_dispatch` run evidence with validated central authority; its prior verified patch publication failed closed after protected main moved. | Draft; preserve until current-main reconciliation |''',
+    )
+
+    DOCTORING.parent.mkdir(parents=True, exist_ok=True)
+    DOCTORING.write_text(
+        '''# Noema elapsed-time authority correction — 2026-09-02
+
+## Incident
+
+Protected main `5935c8153722fe6b53bafd579b74f8f097303959` merged PR #1715.
+It added `timeout-minutes: 20` to Noema close cleanup and
+`timeout-minutes: 210` to the model-bearing Noema job. The cleanup value was an
+analogy to another queue job; the model value combined an inherited allowance
+with an invented buffer. Neither number was supported by a measured runtime
+SLO, standard, or experiment. The 210-minute model cutoff also contradicted the
+accepted ADR-0003 no-fixed-inference-timeout amendment.
+
+## Root cause
+
+The change collapsed distinct failure domains into one elapsed-time mechanism:
+finite housekeeping, GitHub-hosted runner capacity, provider communication,
+and model reasoning. A GitHub job deadline cannot distinguish a model that is
+still reasoning/streaming/calling tools from provider termination, operator
+cancellation, a superseded head, or an explicitly configured model timeout.
+It therefore made elapsed time itself an implicit model-policy owner.
+
+## RED-first evidence
+
+`tests/test_noema_model_timeout_policy.py` was committed on the canonical PR
+#1720 owner branch before the source repair. Against the unmodified PR #1715
+workflow it fails specifically because the model-bearing job contains
+`timeout-minutes: 210`. The one-shot verifier must prove that exact failure
+before it materializes the source repair.
+
+## Repair boundary
+
+Remove both repository-authored 20/210-minute deadlines and replace the old
+positive-timeout contract. Keep `contextual-orchestrator/orchestrator/free`,
+review identity, live-head validation, stale-run cancellation, and security
+boundaries unchanged. Model termination authority remains provider end,
+validated superseded-head cancellation, explicit user/operator cancellation,
+or an explicitly configured contextual-orchestrator administrative timeout.
+GitHub's documented hosting ceiling is treated as an external capacity
+constraint, not a second model-policy value.
+
+## Operational scenarios
+
+1. A reasoning/tool-call path exceeds 210 minutes while still active: `.github`
+   must not terminate it merely because elapsed time reached a local number.
+2. A PR head advances: trusted live-head revalidation may retire the stale run.
+3. A provider ends communication: the upstream request terminates/fails rather
+   than being disguised as a local model timeout.
+4. An administrator configures a contextual-orchestrator timeout: that explicit,
+   auditable owner policy applies without a shadow GitHub Actions deadline.
+5. Cleanup runtime becomes operationally excessive: measure the distribution
+   and queue impact first, then encode an evidence-backed control-plane SLO
+   rather than selecting another analogy-based number.
+
+## References
+
+ContextualWisdomLab. (2026, August 31). *ADR-0003: Vendored contextual-orchestrator review sidecar with governed gateway pools* (model-inference timeout amendment).
+
+GitHub. (n.d.). *Workflow syntax for GitHub Actions*. https://docs.github.com/en/actions/reference/workflows-and-actions/workflow-syntax
+''',
+        encoding="utf-8",
+    )
 
 
 def main() -> None:
-    """Apply production, regression, and traceability repairs."""
+    """Apply production, regression, and governance repairs."""
     patch_workflow()
     patch_test()
-    append_traceability()
+    patch_governance()
 
 
 if __name__ == "__main__":

From 9d60bf78d52abba7a3f11fdc2b1cd82c1127aa4e Mon Sep 17 00:00:00 2001
From: Seongho Bae <me@seonghobae.me>
Date: Wed, 2 Sep 2026 19:23:10 +0900
Subject: [PATCH 05/10] ci(noema): enforce RED-first full-gate successor
 publication

---
 ...source-fix-pr1715-no-model-job-timeout.yml | 32 ++++++++++++++++---
 1 file changed, 27 insertions(+), 5 deletions(-)

diff --git a/.github/workflows/source-fix-pr1715-no-model-job-timeout.yml b/.github/workflows/source-fix-pr1715-no-model-job-timeout.yml
index 25ad3599c2..1c3feb080a 100644
--- a/.github/workflows/source-fix-pr1715-no-model-job-timeout.yml
+++ b/.github/workflows/source-fix-pr1715-no-model-job-timeout.yml
@@ -33,6 +33,7 @@ jobs:
           remote_head="$(git ls-remote origin refs/heads/fix/noema-review-no-heuristic-timeouts | cut -f1)"
           test -n "$remote_head"
           test "$remote_head" = "$GITHUB_SHA"
+          git merge-base --is-ancestor 5935c8153722fe6b53bafd579b74f8f097303959 "$GITHUB_SHA"
 
       - name: Set up Python 3.14
         uses: actions/setup-python@5fda3b95a4ea91299a34e894583c3862153e4b97 # v7.0.0
@@ -46,6 +47,19 @@ jobs:
           set -euo pipefail
           python -m pip install --require-hashes -r requirements-opencode-review-ci-hashes.txt
 
+      - name: Prove PR1715 model-timeout regression is genuinely RED
+        shell: bash
+        run: |
+          set -euo pipefail
+          set +e
+          red_output="$(PYTHONPATH=. python -m pytest -q tests/test_noema_model_timeout_policy.py 2>&1)"
+          red_status=$?
+          set -e
+          printf '%s\n' "$red_output"
+          test "$red_status" -eq 1
+          grep -Eq '1 failed' <<<"$red_output"
+          grep -Fq 'Noema model execution must not be terminated by elapsed time' <<<"$red_output"
+
       - name: Apply causal-owner repair
         shell: bash
         run: |
@@ -54,17 +68,25 @@ jobs:
           python -m py_compile scripts/ci/source_fix_pr1715_no_model_job_timeout.py
           git diff --check
 
-      - name: Verify Noema timeout authority contract
+      - name: Verify focused Noema timeout authority GREEN
         shell: bash
         run: |
           set -euo pipefail
-          python -m pytest \
+          PYTHONPATH=. python -m pytest -q \
+            tests/test_noema_model_timeout_policy.py \
             tests/test_noema_orchestrator_workflow_contract.py \
             tests/test_required_workflow_queue_contract.py \
             tests/test_noema_review_gate.py \
             tests/test_noema_review_handoff.py \
-            tests/test_noema_two_phase_handoff.py \
-            -q
+            tests/test_noema_two_phase_handoff.py
+
+      - name: Verify full repository coverage, docstrings, and syntax GREEN
+        shell: bash
+        run: |
+          set -euo pipefail
+          PYTHONPATH=. python -m coverage run --branch -m pytest tests -q
+          python -m coverage report --show-missing --fail-under=100
+          python -m interrogate -c pyproject.toml scripts/ci
           python -m compileall -q scripts tests .github/actions/noema-review
           git diff --check
 
@@ -74,7 +96,7 @@ jobs:
           set -euo pipefail
           rm scripts/ci/source_fix_pr1715_no_model_job_timeout.py
           rm .github/workflows/source-fix-pr1715-no-model-job-timeout.yml
-          allowed='^(.github/workflows/noema-review.yml|tests/test_noema_orchestrator_workflow_contract.py|CHANGELOG.md|docs/product-technical-gap-baseline.md|scripts/ci/source_fix_pr1715_no_model_job_timeout.py|.github/workflows/source-fix-pr1715-no-model-job-timeout.yml)$'
+          allowed='^(.github/workflows/noema-review.yml|tests/test_noema_orchestrator_workflow_contract.py|tests/test_noema_model_timeout_policy.py|CHANGELOG.md|ARCHITECTURE.md|docs/adr/0003-contextual-orchestrator-vendored-free-zdr.md|docs/doctoring/noema-model-timeout-authority-2026-09-02.md|docs/product-technical-gap-baseline.md|scripts/ci/source_fix_pr1715_no_model_job_timeout.py|.github/workflows/source-fix-pr1715-no-model-job-timeout.yml)$'
           bad="$(git status --short | sed -E 's/^.. //' | grep -Ev "$allowed" || true)"
           test -z "$bad"
           test ! -e scripts/ci/source_fix_pr1715_no_model_job_timeout.py

From 36d8efd3e75ba22c4eb9857e162181e98d023675 Mon Sep 17 00:00:00 2001
From: Seongho Bae <me@seonghobae.me>
Date: Wed, 2 Sep 2026 19:29:43 +0900
Subject: [PATCH 06/10] ci(noema): retire one-shot drivers before full coverage

---
 ...source-fix-pr1715-no-model-job-timeout.yml | 30 +++++++++++++++++--
 1 file changed, 27 insertions(+), 3 deletions(-)

diff --git a/.github/workflows/source-fix-pr1715-no-model-job-timeout.yml b/.github/workflows/source-fix-pr1715-no-model-job-timeout.yml
index 1c3feb080a..5444df0e0e 100644
--- a/.github/workflows/source-fix-pr1715-no-model-job-timeout.yml
+++ b/.github/workflows/source-fix-pr1715-no-model-job-timeout.yml
@@ -68,6 +68,29 @@ jobs:
           python -m py_compile scripts/ci/source_fix_pr1715_no_model_job_timeout.py
           git diff --check
 
+      - name: Retire source-fix drivers before production coverage
+        shell: bash
+        run: |
+          set -euo pipefail
+          # PR #1714 is merged and its writer branch has been deleted. Its
+          # one-shot workflow can therefore no longer have a live push caller;
+          # leaving its Python driver on protected main would falsely turn
+          # temporary repair machinery into an uncovered production surface.
+          prior_writer_head="$(git ls-remote origin refs/heads/fix/autofix-job-timeout | cut -f1)"
+          test -z "$prior_writer_head"
+          test -e scripts/ci/source_fix_pr1714_no_model_job_timeout.py
+          test -e .github/workflows/source-fix-pr1714-no-model-job-timeout.yml
+          rm scripts/ci/source_fix_pr1714_no_model_job_timeout.py
+          rm .github/workflows/source-fix-pr1714-no-model-job-timeout.yml
+
+          # The current source-fix driver has already materialized its entire
+          # successor and passed syntax/diff checks. Remove it before measuring
+          # owned production coverage; the workflow remains to verify and
+          # publish the exact successor, then self-retires below.
+          rm scripts/ci/source_fix_pr1715_no_model_job_timeout.py
+          test ! -e scripts/ci/source_fix_pr1715_no_model_job_timeout.py
+          git diff --check
+
       - name: Verify focused Noema timeout authority GREEN
         shell: bash
         run: |
@@ -90,17 +113,18 @@ jobs:
           python -m compileall -q scripts tests .github/actions/noema-review
           git diff --check
 
-      - name: Retire one-shot repair artifacts and verify scope
+      - name: Retire one-shot workflow and verify successor scope
         shell: bash
         run: |
           set -euo pipefail
-          rm scripts/ci/source_fix_pr1715_no_model_job_timeout.py
           rm .github/workflows/source-fix-pr1715-no-model-job-timeout.yml
-          allowed='^(.github/workflows/noema-review.yml|tests/test_noema_orchestrator_workflow_contract.py|tests/test_noema_model_timeout_policy.py|CHANGELOG.md|ARCHITECTURE.md|docs/adr/0003-contextual-orchestrator-vendored-free-zdr.md|docs/doctoring/noema-model-timeout-authority-2026-09-02.md|docs/product-technical-gap-baseline.md|scripts/ci/source_fix_pr1715_no_model_job_timeout.py|.github/workflows/source-fix-pr1715-no-model-job-timeout.yml)$'
+          allowed='^(.github/workflows/noema-review.yml|tests/test_noema_orchestrator_workflow_contract.py|tests/test_noema_model_timeout_policy.py|CHANGELOG.md|ARCHITECTURE.md|docs/adr/0003-contextual-orchestrator-vendored-free-zdr.md|docs/doctoring/noema-model-timeout-authority-2026-09-02.md|docs/product-technical-gap-baseline.md|scripts/ci/source_fix_pr1715_no_model_job_timeout.py|.github/workflows/source-fix-pr1715-no-model-job-timeout.yml|scripts/ci/source_fix_pr1714_no_model_job_timeout.py|.github/workflows/source-fix-pr1714-no-model-job-timeout.yml)$'
           bad="$(git status --short | sed -E 's/^.. //' | grep -Ev "$allowed" || true)"
           test -z "$bad"
           test ! -e scripts/ci/source_fix_pr1715_no_model_job_timeout.py
           test ! -e .github/workflows/source-fix-pr1715-no-model-job-timeout.yml
+          test ! -e scripts/ci/source_fix_pr1714_no_model_job_timeout.py
+          test ! -e .github/workflows/source-fix-pr1714-no-model-job-timeout.yml
           remote_head="$(git ls-remote origin refs/heads/fix/noema-review-no-heuristic-timeouts | cut -f1)"
           test "$remote_head" = "$GITHUB_SHA"
 

From f536092293371720b0980667a2d97aabb84bb82d Mon Sep 17 00:00:00 2001
From: Seongho Bae <me@seonghobae.me>
Date: Wed, 2 Sep 2026 19:40:40 +0900
Subject: [PATCH 07/10] ci(noema): permit one-shot verified publication via
 built-in token

---
 .../workflows/source-fix-pr1715-no-model-job-timeout.yml   | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/.github/workflows/source-fix-pr1715-no-model-job-timeout.yml b/.github/workflows/source-fix-pr1715-no-model-job-timeout.yml
index 5444df0e0e..d17a880ba1 100644
--- a/.github/workflows/source-fix-pr1715-no-model-job-timeout.yml
+++ b/.github/workflows/source-fix-pr1715-no-model-job-timeout.yml
@@ -13,7 +13,7 @@ concurrency:
   cancel-in-progress: true
 
 permissions:
-  contents: read
+  contents: write
 
 jobs:
   repair:
@@ -132,12 +132,13 @@ jobs:
         env:
           PRIMARY_PUSH_TOKEN: ${{ secrets.PR_REVIEW_MERGE_TOKEN }}
           FALLBACK_PUSH_TOKEN: ${{ secrets.OPENCODE_APPROVE_TOKEN }}
+          BUILTIN_PUSH_TOKEN: ${{ github.token }}
         shell: bash
         run: |
           set -euo pipefail
-          workflow_push_token="${PRIMARY_PUSH_TOKEN:-${FALLBACK_PUSH_TOKEN:-}}"
+          workflow_push_token="${PRIMARY_PUSH_TOKEN:-${FALLBACK_PUSH_TOKEN:-${BUILTIN_PUSH_TOKEN:-}}}"
           if [ -z "$workflow_push_token" ]; then
-            echo "::error::No workflow-starting mutation credential is configured; refusing github.token publication."
+            echo "::error::No mutation credential is configured."
             exit 1
           fi
           remote_head="$(git ls-remote origin refs/heads/fix/noema-review-no-heuristic-timeouts | cut -f1)"

From 0339dbc4c75b9e6e1be972ad9d76ca2c498138d0 Mon Sep 17 00:00:00 2001
From: "github-actions[bot]"
 <41898282+github-actions[bot]@users.noreply.github.com>
Date: Wed, 2 Sep 2026 10:43:58 +0000
Subject: [PATCH 08/10] fix(noema-review): remove unsupported local job
 deadlines

---
 .github/workflows/noema-review.yml            |  32 +-
 ...source-fix-pr1714-no-model-job-timeout.yml | 101 -------
 ...source-fix-pr1715-no-model-job-timeout.yml | 152 ----------
 ARCHITECTURE.md                               |  13 +
 CHANGELOG.md                                  |   1 +
 ...ntextual-orchestrator-vendored-free-zdr.md |  14 +
 ...oema-model-timeout-authority-2026-09-02.md |  58 ++++
 docs/product-technical-gap-baseline.md        |  44 +++
 .../source_fix_pr1714_no_model_job_timeout.py | 151 ----------
 .../source_fix_pr1715_no_model_job_timeout.py | 281 ------------------
 ...st_noema_orchestrator_workflow_contract.py |  58 +---
 11 files changed, 152 insertions(+), 753 deletions(-)
 delete mode 100644 .github/workflows/source-fix-pr1714-no-model-job-timeout.yml
 delete mode 100644 .github/workflows/source-fix-pr1715-no-model-job-timeout.yml
 create mode 100644 docs/doctoring/noema-model-timeout-authority-2026-09-02.md
 delete mode 100644 scripts/ci/source_fix_pr1714_no_model_job_timeout.py
 delete mode 100644 scripts/ci/source_fix_pr1715_no_model_job_timeout.py

diff --git a/.github/workflows/noema-review.yml b/.github/workflows/noema-review.yml
index 834706ad7c..218ac81518 100644
--- a/.github/workflows/noema-review.yml
+++ b/.github/workflows/noema-review.yml
@@ -36,13 +36,10 @@ jobs:
   cancel-closed-pr-runs:
     if: github.event_name == 'pull_request_target' && github.event.action == 'closed'
     runs-on: ubuntu-24.04
-    # Bound this job well short of GitHub's 360-minute platform default. Its
-    # only step is a single-repository, status-filtered gh api --paginate
-    # list-and-cancel sweep (up to 3 passes x 5 statuses), no branch update
-    # or merge -- lighter than pr-review-merge-scheduler.yml's scan-pr-queue
-    # job (PR #1702), which got timeout-minutes: 30 for a comparable
-    # single-repo scan that also dispatches a review and updates a branch.
-    timeout-minutes: 20
+    # No repository-authored wall-clock cutoff: this housekeeping job relies
+    # on GitHub Actions' external platform execution contract plus explicit API
+    # failure. A shorter local deadline would be an unsupported hand-selected
+    # policy rather than a measured queue-control invariant.
     permissions:
       actions: write
       contents: read
@@ -187,21 +184,12 @@ jobs:
   noema-review:
     name: noema-review
     runs-on: ubuntu-24.04
-    # Bound this job well short of GitHub's 360-minute platform default. Its
-    # "Prepare Noema model verdict" step calls into two_phase.py's call_llm
-    # via the same contextual-orchestrator gateway whose unbounded wait was
-    # confirmed to stall runs for 7-20 hours in opencode-review.yml before
-    # PR #1707's fix -- and noema_review_gate.py's own comment says that
-    # step "remains governed by contextual-orchestrator rather than a fixed
-    # inference timeout", so nothing upstream of this job bounds it either.
-    # 210 minutes gives that step the same ~180-minute (3-hour) allowance
-    # PR #1707 set for its analogous model-wait deadline -- comfortably
-    # above this org's documented "accommodate over 2 hours per model"
-    # policy (docs/product-goal-directive.md #8) -- plus a 30-minute buffer
-    # for this job's other steps (tarball fetch, credential mint, the
-    # superseded-run cleanup sweep, visibility-lookup retries, sidecar
-    # provisioning, publication), while staying well under GitHub's default.
-    timeout-minutes: 210
+    # Model-backed Noema intentionally has no repository-authored wall-clock
+    # timeout. contextual-orchestrator/orchestrator/free owns provider routing
+    # and configured model timeout policy; provider termination, live PR/head
+    # invalidation, and explicit user/operator cancellation remain authoritative
+    # stop signals. GitHub's platform execution ceiling is an external capacity
+    # constraint, not a model-selection or test-time-compute policy.
     if: >-
       github.event_name == 'repository_dispatch'
       || (
diff --git a/.github/workflows/source-fix-pr1714-no-model-job-timeout.yml b/.github/workflows/source-fix-pr1714-no-model-job-timeout.yml
deleted file mode 100644
index ad3accb2fa..0000000000
--- a/.github/workflows/source-fix-pr1714-no-model-job-timeout.yml
+++ /dev/null
@@ -1,101 +0,0 @@
-name: Source Fix PR 1714 No Model Job Timeout
-
-on:
-  push:
-    branches:
-      - fix/autofix-job-timeout
-    paths:
-      - scripts/ci/source_fix_pr1714_no_model_job_timeout.py
-      - .github/workflows/source-fix-pr1714-no-model-job-timeout.yml
-
-concurrency:
-  group: source-fix-pr1714-${{ github.repository }}-${{ github.ref_name }}
-  cancel-in-progress: true
-
-permissions:
-  contents: read
-
-jobs:
-  repair:
-    runs-on: ubuntu-slim
-    steps:
-      - name: Checkout exact writer head
-        uses: actions/checkout@9c091bb21b7c1c1d1991bb908d89e4e9dddfe3e0 # v7.0.0
-        with:
-          ref: ${{ github.sha }}
-          fetch-depth: 0
-          persist-credentials: false
-
-      - name: Revalidate exact remote head
-        shell: bash
-        run: |
-          set -euo pipefail
-          remote_head="$(git ls-remote origin refs/heads/fix/autofix-job-timeout | cut -f1)"
-          test -n "$remote_head"
-          test "$remote_head" = "$GITHUB_SHA"
-
-      - name: Set up Python 3.14
-        uses: actions/setup-python@5fda3b95a4ea91299a34e894583c3862153e4b97 # v7.0.0
-        with:
-          python-version: "3.14"
-          cache: pip
-
-      - name: Install exact test toolchain
-        shell: bash
-        run: |
-          set -euo pipefail
-          python -m pip install --require-hashes -r requirements-opencode-review-ci-hashes.txt
-
-      - name: Apply causal-owner repair
-        shell: bash
-        run: |
-          set -euo pipefail
-          python scripts/ci/source_fix_pr1714_no_model_job_timeout.py
-          python -m py_compile scripts/ci/source_fix_pr1714_no_model_job_timeout.py
-          git diff --check
-
-      - name: Verify autofix timeout and writer-security contract
-        shell: bash
-        run: |
-          set -euo pipefail
-          python -m pytest \
-            tests/test_pr_review_autofix_writer_security_contract.py \
-            tests/test_pr_review_fix_scheduler.py \
-            tests/test_required_workflow_queue_contract.py \
-            -q
-          python -m compileall -q scripts tests
-          git diff --check
-
-      - name: Retire one-shot artifacts and verify scope
-        shell: bash
-        run: |
-          set -euo pipefail
-          rm scripts/ci/source_fix_pr1714_no_model_job_timeout.py
-          rm .github/workflows/source-fix-pr1714-no-model-job-timeout.yml
-          allowed='^(.github/workflows/pr-review-autofix.yml|tests/test_pr_review_autofix_writer_security_contract.py|CHANGELOG.md|docs/product-technical-gap-baseline.md|scripts/ci/source_fix_pr1714_no_model_job_timeout.py|.github/workflows/source-fix-pr1714-no-model-job-timeout.yml)$'
-          bad="$(git status --short | sed -E 's/^.. //' | grep -Ev "$allowed" || true)"
-          test -z "$bad"
-          remote_head="$(git ls-remote origin refs/heads/fix/autofix-job-timeout | cut -f1)"
-          test "$remote_head" = "$GITHUB_SHA"
-
-      - name: Publish normal non-force repair commit
-        env:
-          PRIMARY_PUSH_TOKEN: ${{ secrets.PR_REVIEW_MERGE_TOKEN }}
-          FALLBACK_PUSH_TOKEN: ${{ secrets.OPENCODE_APPROVE_TOKEN }}
-        shell: bash
-        run: |
-          set -euo pipefail
-          workflow_push_token="${PRIMARY_PUSH_TOKEN:-${FALLBACK_PUSH_TOKEN:-}}"
-          if [ -z "$workflow_push_token" ]; then
-            echo "::error::No workflow-starting mutation credential is configured; refusing github.token publication."
-            exit 1
-          fi
-          remote_head="$(git ls-remote origin refs/heads/fix/autofix-job-timeout | cut -f1)"
-          test "$remote_head" = "$GITHUB_SHA"
-          git config user.name "github-actions[bot]"
-          git config user.email "41898282+github-actions[bot]@users.noreply.github.com"
-          git add -A
-          git diff --cached --check
-          git commit -m "fix(autofix): remove model wall-clock termination"
-          git remote set-url origin "https://x-access-token:${workflow_push_token}@github.com/${GITHUB_REPOSITORY}.git"
-          git push origin HEAD:fix/autofix-job-timeout
diff --git a/.github/workflows/source-fix-pr1715-no-model-job-timeout.yml b/.github/workflows/source-fix-pr1715-no-model-job-timeout.yml
deleted file mode 100644
index d17a880ba1..0000000000
--- a/.github/workflows/source-fix-pr1715-no-model-job-timeout.yml
+++ /dev/null
@@ -1,152 +0,0 @@
-name: Source Fix PR 1715 No Model Job Timeout
-
-on:
-  push:
-    branches:
-      - fix/noema-review-no-heuristic-timeouts
-    paths:
-      - scripts/ci/source_fix_pr1715_no_model_job_timeout.py
-      - .github/workflows/source-fix-pr1715-no-model-job-timeout.yml
-
-concurrency:
-  group: source-fix-pr1715-${{ github.repository }}-${{ github.ref_name }}
-  cancel-in-progress: true
-
-permissions:
-  contents: write
-
-jobs:
-  repair:
-    runs-on: ubuntu-slim
-    steps:
-      - name: Checkout exact writer head
-        uses: actions/checkout@9c091bb21b7c1c1d1991bb908d89e4e9dddfe3e0 # v7.0.0
-        with:
-          ref: ${{ github.sha }}
-          fetch-depth: 0
-          persist-credentials: false
-
-      - name: Revalidate exact remote head
-        shell: bash
-        run: |
-          set -euo pipefail
-          remote_head="$(git ls-remote origin refs/heads/fix/noema-review-no-heuristic-timeouts | cut -f1)"
-          test -n "$remote_head"
-          test "$remote_head" = "$GITHUB_SHA"
-          git merge-base --is-ancestor 5935c8153722fe6b53bafd579b74f8f097303959 "$GITHUB_SHA"
-
-      - name: Set up Python 3.14
-        uses: actions/setup-python@5fda3b95a4ea91299a34e894583c3862153e4b97 # v7.0.0
-        with:
-          python-version: "3.14"
-          cache: pip
-
-      - name: Install exact test toolchain
-        shell: bash
-        run: |
-          set -euo pipefail
-          python -m pip install --require-hashes -r requirements-opencode-review-ci-hashes.txt
-
-      - name: Prove PR1715 model-timeout regression is genuinely RED
-        shell: bash
-        run: |
-          set -euo pipefail
-          set +e
-          red_output="$(PYTHONPATH=. python -m pytest -q tests/test_noema_model_timeout_policy.py 2>&1)"
-          red_status=$?
-          set -e
-          printf '%s\n' "$red_output"
-          test "$red_status" -eq 1
-          grep -Eq '1 failed' <<<"$red_output"
-          grep -Fq 'Noema model execution must not be terminated by elapsed time' <<<"$red_output"
-
-      - name: Apply causal-owner repair
-        shell: bash
-        run: |
-          set -euo pipefail
-          python scripts/ci/source_fix_pr1715_no_model_job_timeout.py
-          python -m py_compile scripts/ci/source_fix_pr1715_no_model_job_timeout.py
-          git diff --check
-
-      - name: Retire source-fix drivers before production coverage
-        shell: bash
-        run: |
-          set -euo pipefail
-          # PR #1714 is merged and its writer branch has been deleted. Its
-          # one-shot workflow can therefore no longer have a live push caller;
-          # leaving its Python driver on protected main would falsely turn
-          # temporary repair machinery into an uncovered production surface.
-          prior_writer_head="$(git ls-remote origin refs/heads/fix/autofix-job-timeout | cut -f1)"
-          test -z "$prior_writer_head"
-          test -e scripts/ci/source_fix_pr1714_no_model_job_timeout.py
-          test -e .github/workflows/source-fix-pr1714-no-model-job-timeout.yml
-          rm scripts/ci/source_fix_pr1714_no_model_job_timeout.py
-          rm .github/workflows/source-fix-pr1714-no-model-job-timeout.yml
-
-          # The current source-fix driver has already materialized its entire
-          # successor and passed syntax/diff checks. Remove it before measuring
-          # owned production coverage; the workflow remains to verify and
-          # publish the exact successor, then self-retires below.
-          rm scripts/ci/source_fix_pr1715_no_model_job_timeout.py
-          test ! -e scripts/ci/source_fix_pr1715_no_model_job_timeout.py
-          git diff --check
-
-      - name: Verify focused Noema timeout authority GREEN
-        shell: bash
-        run: |
-          set -euo pipefail
-          PYTHONPATH=. python -m pytest -q \
-            tests/test_noema_model_timeout_policy.py \
-            tests/test_noema_orchestrator_workflow_contract.py \
-            tests/test_required_workflow_queue_contract.py \
-            tests/test_noema_review_gate.py \
-            tests/test_noema_review_handoff.py \
-            tests/test_noema_two_phase_handoff.py
-
-      - name: Verify full repository coverage, docstrings, and syntax GREEN
-        shell: bash
-        run: |
-          set -euo pipefail
-          PYTHONPATH=. python -m coverage run --branch -m pytest tests -q
-          python -m coverage report --show-missing --fail-under=100
-          python -m interrogate -c pyproject.toml scripts/ci
-          python -m compileall -q scripts tests .github/actions/noema-review
-          git diff --check
-
-      - name: Retire one-shot workflow and verify successor scope
-        shell: bash
-        run: |
-          set -euo pipefail
-          rm .github/workflows/source-fix-pr1715-no-model-job-timeout.yml
-          allowed='^(.github/workflows/noema-review.yml|tests/test_noema_orchestrator_workflow_contract.py|tests/test_noema_model_timeout_policy.py|CHANGELOG.md|ARCHITECTURE.md|docs/adr/0003-contextual-orchestrator-vendored-free-zdr.md|docs/doctoring/noema-model-timeout-authority-2026-09-02.md|docs/product-technical-gap-baseline.md|scripts/ci/source_fix_pr1715_no_model_job_timeout.py|.github/workflows/source-fix-pr1715-no-model-job-timeout.yml|scripts/ci/source_fix_pr1714_no_model_job_timeout.py|.github/workflows/source-fix-pr1714-no-model-job-timeout.yml)$'
-          bad="$(git status --short | sed -E 's/^.. //' | grep -Ev "$allowed" || true)"
-          test -z "$bad"
-          test ! -e scripts/ci/source_fix_pr1715_no_model_job_timeout.py
-          test ! -e .github/workflows/source-fix-pr1715-no-model-job-timeout.yml
-          test ! -e scripts/ci/source_fix_pr1714_no_model_job_timeout.py
-          test ! -e .github/workflows/source-fix-pr1714-no-model-job-timeout.yml
-          remote_head="$(git ls-remote origin refs/heads/fix/noema-review-no-heuristic-timeouts | cut -f1)"
-          test "$remote_head" = "$GITHUB_SHA"
-
-      - name: Publish normal non-force repair commit
-        env:
-          PRIMARY_PUSH_TOKEN: ${{ secrets.PR_REVIEW_MERGE_TOKEN }}
-          FALLBACK_PUSH_TOKEN: ${{ secrets.OPENCODE_APPROVE_TOKEN }}
-          BUILTIN_PUSH_TOKEN: ${{ github.token }}
-        shell: bash
-        run: |
-          set -euo pipefail
-          workflow_push_token="${PRIMARY_PUSH_TOKEN:-${FALLBACK_PUSH_TOKEN:-${BUILTIN_PUSH_TOKEN:-}}}"
-          if [ -z "$workflow_push_token" ]; then
-            echo "::error::No mutation credential is configured."
-            exit 1
-          fi
-          remote_head="$(git ls-remote origin refs/heads/fix/noema-review-no-heuristic-timeouts | cut -f1)"
-          test "$remote_head" = "$GITHUB_SHA"
-          git config user.name "github-actions[bot]"
-          git config user.email "41898282+github-actions[bot]@users.noreply.github.com"
-          git add -A
-          git diff --cached --check
-          git commit -m "fix(noema-review): remove unsupported local job deadlines"
-          git remote set-url origin "https://x-access-token:${workflow_push_token}@github.com/${GITHUB_REPOSITORY}.git"
-          git push origin HEAD:fix/noema-review-no-heuristic-timeouts
diff --git a/ARCHITECTURE.md b/ARCHITECTURE.md
index e12f33542d..2f94f1ed4d 100644
--- a/ARCHITECTURE.md
+++ b/ARCHITECTURE.md
@@ -266,3 +266,16 @@ resolver conflict.
   — current increment's attestation decision and APA 7th citations.
 - [`docs/doctoring/sandboxed-web-readiness-loopback-boundary.md`](docs/doctoring/sandboxed-web-readiness-loopback-boundary.md)
   — loopback-only web E2E readiness polling and APA 7th citations.
+
+## Model execution timeout authority (2026-09-02)
+
+Central model-backed review jobs (`OpenCode`, `Noema`, and `Strix`) delegate model
+selection and inference lifecycle to `ContextualWisdomLab/contextual-orchestrator`.
+The GitHub Actions job that contains reasoning, streaming, or tool execution must
+not invent a fixed elapsed-time ceiling. Intentional termination authorities are
+an explicit user/operator cancellation, provider termination, superseded-head
+retirement after live-head validation, or an explicitly configured
+contextual-orchestrator administrative timeout. Queue/runner hygiene must be
+expressed as measured control-plane contracts rather than hand-selected model
+runtime budgets. GitHub's hosting limit remains an external capacity constraint,
+not `.github` model policy.
diff --git a/CHANGELOG.md b/CHANGELOG.md
index ac1985d86f..18b58e6563 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -5,6 +5,7 @@ this file. The format follows Keep a Changelog, and versioned releases follow
 Semantic Versioning where the repository publishes a release.
 
 ## [Unreleased]
+- **Correct PR #1715 timeout authority regression.** Remove the inferred 20-minute Noema cleanup deadline and the model-bearing 210-minute job deadline. Neither value had standards, measured-runtime, or experimentally validated provenance. The model-bearing deadline additionally contradicted ADR-0003 by allowing elapsed time alone to terminate contextual-orchestrator reasoning. GitHub's hosting ceiling remains an external capacity constraint; provider end, live-head supersession, explicit user/operator cancellation, or an explicitly configured contextual-orchestrator administrative timeout are the intentional model termination authorities.
 - **Fail closed before cancelling stale PR workflow runs.** Validate snapshot `headRefOid` and re-read live PR/run identity immediately before destructive cancellation, including OpenCode/Strix dispatch cleanup, so a missing head or concurrent push cannot cancel the sole current-head evidence or trigger a duplicate review. Also ensures every cancellation path (`cancel_stale_pr_runs`, `cancel_stale_opencode_runs`, `_cancel_revalidated_review_run_refs`) treats a run as cancelled only when `force_cancel_workflow_runs` actually reports success, not merely when live revalidation proved it stale -- superseding PR #1712's simpler `force_cancel_workflow_run_refs` wrapper (removed as dead code; its safety guarantee is preserved inline at every call site by this more thorough revalidate-then-cancel design).
 - **Cache `active_workflow_runs` for the life of one `pr_review_merge_scheduler.py`
   invocation.** `inspect_pr()` calls `cancel_stale_pr_runs()` unconditionally for
diff --git a/docs/adr/0003-contextual-orchestrator-vendored-free-zdr.md b/docs/adr/0003-contextual-orchestrator-vendored-free-zdr.md
index 217b3cc0b1..e726ce2f33 100644
--- a/docs/adr/0003-contextual-orchestrator-vendored-free-zdr.md
+++ b/docs/adr/0003-contextual-orchestrator-vendored-free-zdr.md
@@ -250,3 +250,17 @@ all five, and auto-optimize routing by cost.
   fault. Accepted-size and tool-schema probes call the pinned client's
   deterministic mock response explicitly and therefore perform no provider
   call.
+
+- **2026-09-02 correction: PR #1715's Noema deadlines were not a new timeout decision.**
+  The accepted 2026-08-31 amendment above already states that OpenCode, Noema,
+  Strix, and contextual-orchestrator inference have no repository/application
+  fixed wall-clock timeout. PR #1715 added `timeout-minutes: 20` to the
+  housekeeping job by analogy to another queue job and `timeout-minutes: 210`
+  to the model-bearing Noema job by combining an inherited allowance with an
+  invented buffer. Neither local number had standards, measured-runtime, or
+  experimentally validated provenance, and the model-bearing value directly
+  contradicted this ADR. Both local deadlines are therefore removed. Provider
+  termination, exact live-head invalidation, explicit user/operator cancellation,
+  or an explicitly configured contextual-orchestrator administrative timeout
+  remain intentional model termination authorities; GitHub's hosting ceiling is
+  documented as an external runtime constraint rather than repository policy.
diff --git a/docs/doctoring/noema-model-timeout-authority-2026-09-02.md b/docs/doctoring/noema-model-timeout-authority-2026-09-02.md
new file mode 100644
index 0000000000..82699d0199
--- /dev/null
+++ b/docs/doctoring/noema-model-timeout-authority-2026-09-02.md
@@ -0,0 +1,58 @@
+# Noema elapsed-time authority correction — 2026-09-02
+
+## Incident
+
+Protected main `5935c8153722fe6b53bafd579b74f8f097303959` merged PR #1715.
+It added `timeout-minutes: 20` to Noema close cleanup and
+`timeout-minutes: 210` to the model-bearing Noema job. The cleanup value was an
+analogy to another queue job; the model value combined an inherited allowance
+with an invented buffer. Neither number was supported by a measured runtime
+SLO, standard, or experiment. The 210-minute model cutoff also contradicted the
+accepted ADR-0003 no-fixed-inference-timeout amendment.
+
+## Root cause
+
+The change collapsed distinct failure domains into one elapsed-time mechanism:
+finite housekeeping, GitHub-hosted runner capacity, provider communication,
+and model reasoning. A GitHub job deadline cannot distinguish a model that is
+still reasoning/streaming/calling tools from provider termination, operator
+cancellation, a superseded head, or an explicitly configured model timeout.
+It therefore made elapsed time itself an implicit model-policy owner.
+
+## RED-first evidence
+
+`tests/test_noema_model_timeout_policy.py` was committed on the canonical PR
+#1720 owner branch before the source repair. Against the unmodified PR #1715
+workflow it fails specifically because the model-bearing job contains
+`timeout-minutes: 210`. The one-shot verifier must prove that exact failure
+before it materializes the source repair.
+
+## Repair boundary
+
+Remove both repository-authored 20/210-minute deadlines and replace the old
+positive-timeout contract. Keep `contextual-orchestrator/orchestrator/free`,
+review identity, live-head validation, stale-run cancellation, and security
+boundaries unchanged. Model termination authority remains provider end,
+validated superseded-head cancellation, explicit user/operator cancellation,
+or an explicitly configured contextual-orchestrator administrative timeout.
+GitHub's documented hosting ceiling is treated as an external capacity
+constraint, not a second model-policy value.
+
+## Operational scenarios
+
+1. A reasoning/tool-call path exceeds 210 minutes while still active: `.github`
+   must not terminate it merely because elapsed time reached a local number.
+2. A PR head advances: trusted live-head revalidation may retire the stale run.
+3. A provider ends communication: the upstream request terminates/fails rather
+   than being disguised as a local model timeout.
+4. An administrator configures a contextual-orchestrator timeout: that explicit,
+   auditable owner policy applies without a shadow GitHub Actions deadline.
+5. Cleanup runtime becomes operationally excessive: measure the distribution
+   and queue impact first, then encode an evidence-backed control-plane SLO
+   rather than selecting another analogy-based number.
+
+## References
+
+ContextualWisdomLab. (2026, August 31). *ADR-0003: Vendored contextual-orchestrator review sidecar with governed gateway pools* (model-inference timeout amendment).
+
+GitHub. (n.d.). *Workflow syntax for GitHub Actions*. https://docs.github.com/en/actions/reference/workflows-and-actions/workflow-syntax
diff --git a/docs/product-technical-gap-baseline.md b/docs/product-technical-gap-baseline.md
index 29acdfeecc..69b1c9b1be 100644
--- a/docs/product-technical-gap-baseline.md
+++ b/docs/product-technical-gap-baseline.md
@@ -2613,3 +2613,47 @@ Higgins, S. S., Crepalde, N., & Fernandes, L. (2021). Segmented multiplexity: A
 **Expected effect.** No observable change to any current GitHub Actions review run (every current invocation already resolves to `free`). The effect is structural: it is no longer possible for a future workflow edit or manual dispatch override to admit priced-model spend into a required review check without an explicit, reviewed code change to this one `case` statement (and its now-locked-in regression test) first.
 
 **Follow-up.** If the organization later solves free+ZDR routing robustly enough to deliberately widen required-review CI to `orchestrator/auto` (e.g. once a spend ceiling and reviewer-visible cost evidence exist for that path), the change is exactly one `case` arm plus the corresponding assertions in `test_sidecar_pins_the_pool_to_free_for_github_actions` — this entry is the record of *why* it was narrowed, not a permanent prohibition.
+
+### Noema workflow elapsed-time authority — PR #1715 successor
+
+- **Protected-main incident base:** `5935c8153722fe6b53bafd579b74f8f097303959`
+  (merge of PR #1715).
+- **Live gap:** PR #1715 introduced local `timeout-minutes: 20` and
+  `timeout-minutes: 210`. The former was inferred by analogy to another queue
+  job; the latter combined an inherited allowance with an invented buffer.
+  Neither value had executable standards, measured-runtime, or experimentally
+  validated provenance. The 210-minute value also violated ADR-0003's existing
+  no-fixed-inference-timeout contract.
+- **PRD goal:** preserve correctness-first long-running Noema review without
+  replacing model/provider lifecycle authority with elapsed time.
+- **TRD invariant:** neither Noema job invents a repository-authored local
+  deadline; the model-bearing job uses `contextual-orchestrator/orchestrator/free`
+  and terminates intentionally only on provider end, exact live-state
+  invalidation, explicit user/operator cancellation, or an explicitly configured
+  contextual-orchestrator administrative timeout. GitHub's host ceiling remains
+  an external capacity constraint.
+- **Context Map:** `.github` owns Actions admission, exact-head validation, and
+  stale-run retirement; `ContextualWisdomLab/contextual-orchestrator` owns model
+  routing and configured model timeout policy.
+- **Regression:** `tests/test_noema_model_timeout_policy.py` is committed RED
+  before the source repair and `test_noema_jobs_do_not_invent_repository_wall_clock_deadlines`
+  rejects reintroduction on either job.
+- **Status:** Proposed until the one-shot self-removes and fresh exact-head
+  required Checks are GREEN.
+
+```mermaid
+flowchart LR
+    PR[Exact PR head] --> GH[.github Noema control plane]
+    GH --> CO[ContextualWisdomLab/contextual-orchestrator]
+    CO --> MODEL[orchestrator/free model execution]
+    USER[User/operator cancel] --> GH
+    HEAD[Superseding PR head] --> GH
+    PROVIDER[Provider termination] --> CO
+    ADMIN[Configured CO admin timeout] --> CO
+    GH -. no local elapsed-time deadline .-> MODEL
+```
+
+| Gap | Action | Status |
+| --- | --- | --- |
+| G-NOEMA-TIMEOUT-AUTHORITY | RED→remove unsupported 20/210-minute local deadlines→full GREEN→self-retire repair artifacts→fresh exact-head required Checks. | Proposed / PR #1720 |
+| G-STALE-RUN-CENTRAL-AUTHORITY | PR #1717 must read central `repository_dispatch` run evidence with validated central authority; its prior verified patch publication failed closed after protected main moved. | Draft; preserve until current-main reconciliation |
diff --git a/scripts/ci/source_fix_pr1714_no_model_job_timeout.py b/scripts/ci/source_fix_pr1714_no_model_job_timeout.py
deleted file mode 100644
index 415cf176ae..0000000000
--- a/scripts/ci/source_fix_pr1714_no_model_job_timeout.py
+++ /dev/null
@@ -1,151 +0,0 @@
-"""One-shot repair for PR #1714's model-backed autofix no-heuristics contract."""
-
-from __future__ import annotations
-
-from pathlib import Path
-
-WORKFLOW = Path(".github/workflows/pr-review-autofix.yml")
-TEST = Path("tests/test_pr_review_autofix_writer_security_contract.py")
-CHANGELOG = Path("CHANGELOG.md")
-BASELINE = Path("docs/product-technical-gap-baseline.md")
-
-
-def replace_once(text: str, old: str, new: str, label: str) -> str:
-    """Replace one literal block and fail closed if the exact head moved semantically."""
-    count = text.count(old)
-    if count != 1:
-        raise SystemExit(f"PR1714 {label}: expected one literal block, found {count}")
-    return text.replace(old, new, 1)
-
-
-def patch_workflow() -> None:
-    """Remove repository-authored model termination, compute, capability, and evidence heuristics."""
-    text = WORKFLOW.read_text(encoding="utf-8")
-    timeout_old = '''    # Bound the job well short of GitHub's 360-minute platform default. Setup
-    # (checkout, OIDC token exchange, OpenCode CLI install, context collection)
-    # is API/IO-bound and normally finishes in a few minutes; the one
-    # `opencode run` call (12 agent steps, single fixed model, no
-    # multi-provider fallback pool unlike opencode-review-dispatch.yml's
-    # review job) is the dominant cost, followed by fast local validation
-    # and a single git commit/push. 25 minutes gives that single LLM run
-    # generous per-step room while still failing a hung invocation well
-    # before the platform cap.
-    timeout-minutes: 25
-'''
-    timeout_new = '''    # This job is model-backed through contextual-orchestrator/orchestrator/free
-    # and therefore has no repository-owned wall-clock timeout. Provider end,
-    # explicit cancellation, and the workflow's exact live-head/state guards
-    # are authoritative; elapsed time alone must not terminate reasoning,
-    # streaming, or tool work. Queue pressure is handled by the scheduler's
-    # stale-head dedupe/cancellation rather than by killing current-head work.
-'''
-    text = replace_once(text, timeout_old, timeout_new, "autofix timeout block")
-
-    text = replace_once(
-        text,
-        '                "reasoningEffort": "high",\n',
-        "",
-        "repository-authored reasoning effort",
-    )
-    text = replace_once(
-        text,
-        '                "steps": 12,\n',
-        "",
-        "repository-authored agent step budget",
-    )
-    capability_old = '''                    "name": "Orchestrator Free (ZDR-first zero-cost pool)",
-                    "tool_call": true,
-                    "reasoning": true,
-                    "limit": {
-                      "context": 200000,
-                      "output": 32768
-                    }
-'''
-    capability_new = '''                    "name": "Orchestrator Free (ZDR-first zero-cost pool)"
-'''
-    text = replace_once(
-        text,
-        capability_old,
-        capability_new,
-        "leaf model capability and context/output declarations",
-    )
-    text = replace_once(
-        text,
-        '          $(sed -n \'1,260p\' "$RUNNER_TEMP/pr-review-autofix-context.md")\n',
-        '          $(cat "$RUNNER_TEMP/pr-review-autofix-context.md")\n',
-        "review-context line quota",
-    )
-    WORKFLOW.write_text(text, encoding="utf-8")
-
-
-def patch_test() -> None:
-    """Replace the timeout-positive regression with fail-closed authority contracts."""
-    text = TEST.read_text(encoding="utf-8")
-    marker = "def test_autofix_job_has_a_bounded_runtime() -> None:\n"
-    start = text.find(marker)
-    if start < 0 or text.find(marker, start + 1) >= 0:
-        raise SystemExit("PR1714 stale timeout test marker moved or duplicated")
-    replacement = '''def test_autofix_model_job_delegates_termination_and_compute_to_orchestrator() -> None:
-    """Leaf OpenCode config must not invent model-time or test-time-compute authority."""
-    workflow = _workflow_text()
-    job = workflow.split("  autofix:\\n", maxsplit=1)[1]
-    job_header = job.split("    steps:\\n", maxsplit=1)[0]
-
-    assert "timeout-minutes:" not in job_header
-    assert '"model": "contextual-orchestrator/orchestrator/free"' in workflow
-    assert '"reasoningEffort":' not in workflow
-    assert '"steps": 12' not in workflow
-    assert '"tool_call": true' not in workflow
-    assert '"reasoning": true' not in workflow
-    assert '"limit": {' not in workflow
-    assert "no repository-owned wall-clock timeout" in job_header
-    assert "cancel-in-progress: false" in workflow
-
-
-def test_autofix_review_context_is_not_sampled_by_a_fixed_line_quota() -> None:
-    """Exact review evidence must reach the model without a repository-authored line cutoff."""
-    workflow = _workflow_text()
-
-    assert "sed -n '1,260p'" not in workflow
-    assert '$(cat "$RUNNER_TEMP/pr-review-autofix-context.md")' in workflow
-'''
-    TEST.write_text(text[:start] + replacement, encoding="utf-8")
-
-
-def append_traceability() -> None:
-    """Document the model-authority and complete-evidence boundary."""
-    changelog = CHANGELOG.read_text(encoding="utf-8")
-    note = (
-        "\n- PR #1714: reject repository-authored OpenCode autofix wall-clock, reasoning-effort, "
-        "agent-step, capability/context/output, and fixed review-line allocation. The leaf requests "
-        "only `orchestrator/free`; contextual-orchestrator owns verified capability/routing/test-time "
-        "compute and the full collected review evidence is passed without a hand-selected line quota.\n"
-    )
-    if "PR #1714: reject repository-authored OpenCode autofix wall-clock" not in changelog:
-        CHANGELOG.write_text(changelog + note, encoding="utf-8")
-
-    baseline = BASELINE.read_text(encoding="utf-8")
-    section = '''
-
-### OpenCode autofix orchestration authority — PR #1714
-
-- **Root cause:** the leaf workflow proposed `timeout-minutes: 25` and also carried repository-authored `reasoningEffort: high`, a 12-step agent budget, asserted tool/reasoning capabilities, fixed context/output limits, and a 260-line review-context cutoff. None of those leaf allocations had executable research/model evidence establishing them as decision authority.
-- **Owner boundary:** `.github` requests exactly `contextual-orchestrator/orchestrator/free` through the gateway token. contextual-orchestrator owns provider discovery, verified capability admission, routing, and research-backed test-time compute; the leaf does not invent provider/model capability or compute limits.
-- **Evidence contract:** the complete review context produced by the governed collector is passed to the model. If contextual-orchestrator cannot admit/serve the request under its verified capability/privacy/free-pool contracts, the path fails closed rather than silently sampling evidence or selecting a paid/provider fallback.
-- **Termination contract:** provider completion, explicit cancellation, and exact live-head/state guards end model work. Scheduler stale-head dedupe/cancellation handles queue waste without terminating the sole current-head model run by elapsed time.
-- **Regression:** `test_autofix_model_job_delegates_termination_and_compute_to_orchestrator` and `test_autofix_review_context_is_not_sampled_by_a_fixed_line_quota` forbid reintroduction of those leaf heuristics while preserving the exact `orchestrator/free` contract.
-- **Status:** Proposed until the one-shot source repair self-removes and fresh exact-head Checks are GREEN.
-'''
-    if "### OpenCode autofix orchestration authority — PR #1714" not in baseline:
-        BASELINE.write_text(baseline + section, encoding="utf-8")
-
-
-def main() -> None:
-    """Apply production, regression, and traceability changes."""
-    patch_workflow()
-    patch_test()
-    append_traceability()
-
-
-if __name__ == "__main__":
-    main()
diff --git a/scripts/ci/source_fix_pr1715_no_model_job_timeout.py b/scripts/ci/source_fix_pr1715_no_model_job_timeout.py
deleted file mode 100644
index d041a29e63..0000000000
--- a/scripts/ci/source_fix_pr1715_no_model_job_timeout.py
+++ /dev/null
@@ -1,281 +0,0 @@
-"""One-shot exact-head repair for the merged PR #1715 timeout-policy regression."""
-
-from __future__ import annotations
-
-import re
-from pathlib import Path
-
-WORKFLOW = Path(".github/workflows/noema-review.yml")
-TEST = Path("tests/test_noema_orchestrator_workflow_contract.py")
-CHANGELOG = Path("CHANGELOG.md")
-ARCHITECTURE = Path("ARCHITECTURE.md")
-ADR = Path("docs/adr/0003-contextual-orchestrator-vendored-free-zdr.md")
-BASELINE = Path("docs/product-technical-gap-baseline.md")
-DOCTORING = Path("docs/doctoring/noema-model-timeout-authority-2026-09-02.md")
-
-
-def replace_once(text: str, old: str, new: str, label: str) -> str:
-    """Replace one literal block and fail closed when branch contents moved."""
-    count = text.count(old)
-    if count != 1:
-        raise SystemExit(f"PR1715 successor {label}: expected one literal block, found {count}")
-    return text.replace(old, new, 1)
-
-
-def append_once(path: Path, marker: str, section: str) -> None:
-    """Append one governance section while refusing duplicate doctoring."""
-    text = path.read_text(encoding="utf-8")
-    if marker in text:
-        return
-    path.write_text(text.rstrip() + "\n\n" + section.strip() + "\n", encoding="utf-8")
-
-
-def patch_workflow() -> None:
-    """Remove repository-authored elapsed-time termination from both Noema jobs."""
-    text = WORKFLOW.read_text(encoding="utf-8")
-    cleanup_old = '''    # Bound this job well short of GitHub's 360-minute platform default. Its
-    # only step is a single-repository, status-filtered gh api --paginate
-    # list-and-cancel sweep (up to 3 passes x 5 statuses), no branch update
-    # or merge -- lighter than pr-review-merge-scheduler.yml's scan-pr-queue
-    # job (PR #1702), which got timeout-minutes: 30 for a comparable
-    # single-repo scan that also dispatches a review and updates a branch.
-    timeout-minutes: 20
-'''
-    cleanup_new = '''    # No repository-authored wall-clock cutoff: this housekeeping job relies
-    # on GitHub Actions' external platform execution contract plus explicit API
-    # failure. A shorter local deadline would be an unsupported hand-selected
-    # policy rather than a measured queue-control invariant.
-'''
-    text = replace_once(text, cleanup_old, cleanup_new, "cleanup timeout block")
-
-    model_old = '''    # Bound this job well short of GitHub's 360-minute platform default. Its
-    # "Prepare Noema model verdict" step calls into two_phase.py's call_llm
-    # via the same contextual-orchestrator gateway whose unbounded wait was
-    # confirmed to stall runs for 7-20 hours in opencode-review.yml before
-    # PR #1707's fix -- and noema_review_gate.py's own comment says that
-    # step "remains governed by contextual-orchestrator rather than a fixed
-    # inference timeout", so nothing upstream of this job bounds it either.
-    # 210 minutes gives that step the same ~180-minute (3-hour) allowance
-    # PR #1707 set for its analogous model-wait deadline -- comfortably
-    # above this org's documented "accommodate over 2 hours per model"
-    # policy (docs/product-goal-directive.md #8) -- plus a 30-minute buffer
-    # for this job's other steps (tarball fetch, credential mint, the
-    # superseded-run cleanup sweep, visibility-lookup retries, sidecar
-    # provisioning, publication), while staying well under GitHub's default.
-    timeout-minutes: 210
-'''
-    model_new = '''    # Model-backed Noema intentionally has no repository-authored wall-clock
-    # timeout. contextual-orchestrator/orchestrator/free owns provider routing
-    # and configured model timeout policy; provider termination, live PR/head
-    # invalidation, and explicit user/operator cancellation remain authoritative
-    # stop signals. GitHub's platform execution ceiling is an external capacity
-    # constraint, not a model-selection or test-time-compute policy.
-'''
-    text = replace_once(text, model_old, model_new, "model job timeout block")
-    WORKFLOW.write_text(text, encoding="utf-8")
-
-
-def patch_test() -> None:
-    """Replace timeout-positive regressions with no-local-deadline contracts."""
-    text = TEST.read_text(encoding="utf-8")
-    marker = "def test_cancel_closed_pr_runs_has_a_bounded_runtime() -> None:\n"
-    start = text.find(marker)
-    if start < 0 or text.find(marker, start + 1) >= 0:
-        raise SystemExit("PR1715 successor timeout-test marker moved or duplicated")
-    replacement = '''def test_noema_jobs_do_not_invent_repository_wall_clock_deadlines() -> None:
-    """Noema model/support jobs must not encode hand-selected elapsed-time cutoffs."""
-    workflow = workflow_text("noema-review.yml")
-    cleanup = workflow.split("  cancel-closed-pr-runs:\\n", 1)[1].split(
-        "\\n  noema-review:\\n", 1
-    )[0]
-    model = workflow.split("  noema-review:\\n", 1)[1]
-
-    assert re.search(r"^    timeout-minutes:", cleanup, flags=re.MULTILINE) is None
-    assert re.search(r"^    timeout-minutes:", model, flags=re.MULTILINE) is None
-    assert "contextual-orchestrator/orchestrator/free" in workflow
-    assert "unsupported hand-selected" in cleanup
-    assert "no repository-authored wall-clock" in model
-'''
-    TEST.write_text(text[:start] + replacement, encoding="utf-8")
-
-
-def patch_governance() -> None:
-    """Synchronize change history, architecture, ADR, doctoring, and product-gap status."""
-    changelog = CHANGELOG.read_text(encoding="utf-8")
-    entry = (
-        "- **Correct PR #1715 timeout authority regression.** Remove the inferred "
-        "20-minute Noema cleanup deadline and the model-bearing 210-minute job "
-        "deadline. Neither value had standards, measured-runtime, or experimentally "
-        "validated provenance. The model-bearing deadline additionally contradicted "
-        "ADR-0003 by allowing elapsed time alone to terminate contextual-orchestrator "
-        "reasoning. GitHub's hosting ceiling remains an external capacity constraint; "
-        "provider end, live-head supersession, explicit user/operator cancellation, "
-        "or an explicitly configured contextual-orchestrator administrative timeout "
-        "are the intentional model termination authorities.\n"
-    )
-    if entry.strip() not in changelog:
-        changelog = replace_once(
-            changelog,
-            "## [Unreleased]\n",
-            "## [Unreleased]\n" + entry,
-            "CHANGELOG Unreleased header",
-        )
-        CHANGELOG.write_text(changelog, encoding="utf-8")
-
-    append_once(
-        ARCHITECTURE,
-        "## Model execution timeout authority (2026-09-02)",
-        '''## Model execution timeout authority (2026-09-02)
-
-Central model-backed review jobs (`OpenCode`, `Noema`, and `Strix`) delegate model
-selection and inference lifecycle to `ContextualWisdomLab/contextual-orchestrator`.
-The GitHub Actions job that contains reasoning, streaming, or tool execution must
-not invent a fixed elapsed-time ceiling. Intentional termination authorities are
-an explicit user/operator cancellation, provider termination, superseded-head
-retirement after live-head validation, or an explicitly configured
-contextual-orchestrator administrative timeout. Queue/runner hygiene must be
-expressed as measured control-plane contracts rather than hand-selected model
-runtime budgets. GitHub's hosting limit remains an external capacity constraint,
-not `.github` model policy.''',
-    )
-
-    append_once(
-        ADR,
-        "2026-09-02 correction: PR #1715's Noema deadlines were not a new timeout decision",
-        '''- **2026-09-02 correction: PR #1715's Noema deadlines were not a new timeout decision.**
-  The accepted 2026-08-31 amendment above already states that OpenCode, Noema,
-  Strix, and contextual-orchestrator inference have no repository/application
-  fixed wall-clock timeout. PR #1715 added `timeout-minutes: 20` to the
-  housekeeping job by analogy to another queue job and `timeout-minutes: 210`
-  to the model-bearing Noema job by combining an inherited allowance with an
-  invented buffer. Neither local number had standards, measured-runtime, or
-  experimentally validated provenance, and the model-bearing value directly
-  contradicted this ADR. Both local deadlines are therefore removed. Provider
-  termination, exact live-head invalidation, explicit user/operator cancellation,
-  or an explicitly configured contextual-orchestrator administrative timeout
-  remain intentional model termination authorities; GitHub's hosting ceiling is
-  documented as an external runtime constraint rather than repository policy.''',
-    )
-
-    append_once(
-        BASELINE,
-        "### Noema workflow elapsed-time authority — PR #1715 successor",
-        '''### Noema workflow elapsed-time authority — PR #1715 successor
-
-- **Protected-main incident base:** `5935c8153722fe6b53bafd579b74f8f097303959`
-  (merge of PR #1715).
-- **Live gap:** PR #1715 introduced local `timeout-minutes: 20` and
-  `timeout-minutes: 210`. The former was inferred by analogy to another queue
-  job; the latter combined an inherited allowance with an invented buffer.
-  Neither value had executable standards, measured-runtime, or experimentally
-  validated provenance. The 210-minute value also violated ADR-0003's existing
-  no-fixed-inference-timeout contract.
-- **PRD goal:** preserve correctness-first long-running Noema review without
-  replacing model/provider lifecycle authority with elapsed time.
-- **TRD invariant:** neither Noema job invents a repository-authored local
-  deadline; the model-bearing job uses `contextual-orchestrator/orchestrator/free`
-  and terminates intentionally only on provider end, exact live-state
-  invalidation, explicit user/operator cancellation, or an explicitly configured
-  contextual-orchestrator administrative timeout. GitHub's host ceiling remains
-  an external capacity constraint.
-- **Context Map:** `.github` owns Actions admission, exact-head validation, and
-  stale-run retirement; `ContextualWisdomLab/contextual-orchestrator` owns model
-  routing and configured model timeout policy.
-- **Regression:** `tests/test_noema_model_timeout_policy.py` is committed RED
-  before the source repair and `test_noema_jobs_do_not_invent_repository_wall_clock_deadlines`
-  rejects reintroduction on either job.
-- **Status:** Proposed until the one-shot self-removes and fresh exact-head
-  required Checks are GREEN.
-
-```mermaid
-flowchart LR
-    PR[Exact PR head] --> GH[.github Noema control plane]
-    GH --> CO[ContextualWisdomLab/contextual-orchestrator]
-    CO --> MODEL[orchestrator/free model execution]
-    USER[User/operator cancel] --> GH
-    HEAD[Superseding PR head] --> GH
-    PROVIDER[Provider termination] --> CO
-    ADMIN[Configured CO admin timeout] --> CO
-    GH -. no local elapsed-time deadline .-> MODEL
-```
-
-| Gap | Action | Status |
-| --- | --- | --- |
-| G-NOEMA-TIMEOUT-AUTHORITY | RED→remove unsupported 20/210-minute local deadlines→full GREEN→self-retire repair artifacts→fresh exact-head required Checks. | Proposed / PR #1720 |
-| G-STALE-RUN-CENTRAL-AUTHORITY | PR #1717 must read central `repository_dispatch` run evidence with validated central authority; its prior verified patch publication failed closed after protected main moved. | Draft; preserve until current-main reconciliation |''',
-    )
-
-    DOCTORING.parent.mkdir(parents=True, exist_ok=True)
-    DOCTORING.write_text(
-        '''# Noema elapsed-time authority correction — 2026-09-02
-
-## Incident
-
-Protected main `5935c8153722fe6b53bafd579b74f8f097303959` merged PR #1715.
-It added `timeout-minutes: 20` to Noema close cleanup and
-`timeout-minutes: 210` to the model-bearing Noema job. The cleanup value was an
-analogy to another queue job; the model value combined an inherited allowance
-with an invented buffer. Neither number was supported by a measured runtime
-SLO, standard, or experiment. The 210-minute model cutoff also contradicted the
-accepted ADR-0003 no-fixed-inference-timeout amendment.
-
-## Root cause
-
-The change collapsed distinct failure domains into one elapsed-time mechanism:
-finite housekeeping, GitHub-hosted runner capacity, provider communication,
-and model reasoning. A GitHub job deadline cannot distinguish a model that is
-still reasoning/streaming/calling tools from provider termination, operator
-cancellation, a superseded head, or an explicitly configured model timeout.
-It therefore made elapsed time itself an implicit model-policy owner.
-
-## RED-first evidence
-
-`tests/test_noema_model_timeout_policy.py` was committed on the canonical PR
-#1720 owner branch before the source repair. Against the unmodified PR #1715
-workflow it fails specifically because the model-bearing job contains
-`timeout-minutes: 210`. The one-shot verifier must prove that exact failure
-before it materializes the source repair.
-
-## Repair boundary
-
-Remove both repository-authored 20/210-minute deadlines and replace the old
-positive-timeout contract. Keep `contextual-orchestrator/orchestrator/free`,
-review identity, live-head validation, stale-run cancellation, and security
-boundaries unchanged. Model termination authority remains provider end,
-validated superseded-head cancellation, explicit user/operator cancellation,
-or an explicitly configured contextual-orchestrator administrative timeout.
-GitHub's documented hosting ceiling is treated as an external capacity
-constraint, not a second model-policy value.
-
-## Operational scenarios
-
-1. A reasoning/tool-call path exceeds 210 minutes while still active: `.github`
-   must not terminate it merely because elapsed time reached a local number.
-2. A PR head advances: trusted live-head revalidation may retire the stale run.
-3. A provider ends communication: the upstream request terminates/fails rather
-   than being disguised as a local model timeout.
-4. An administrator configures a contextual-orchestrator timeout: that explicit,
-   auditable owner policy applies without a shadow GitHub Actions deadline.
-5. Cleanup runtime becomes operationally excessive: measure the distribution
-   and queue impact first, then encode an evidence-backed control-plane SLO
-   rather than selecting another analogy-based number.
-
-## References
-
-ContextualWisdomLab. (2026, August 31). *ADR-0003: Vendored contextual-orchestrator review sidecar with governed gateway pools* (model-inference timeout amendment).
-
-GitHub. (n.d.). *Workflow syntax for GitHub Actions*. https://docs.github.com/en/actions/reference/workflows-and-actions/workflow-syntax
-''',
-        encoding="utf-8",
-    )
-
-
-def main() -> None:
-    """Apply production, regression, and governance repairs."""
-    patch_workflow()
-    patch_test()
-    patch_governance()
-
-
-if __name__ == "__main__":
-    main()
diff --git a/tests/test_noema_orchestrator_workflow_contract.py b/tests/test_noema_orchestrator_workflow_contract.py
index f6e97f745b..a67f37414d 100644
--- a/tests/test_noema_orchestrator_workflow_contract.py
+++ b/tests/test_noema_orchestrator_workflow_contract.py
@@ -371,50 +371,16 @@ def test_strix_gateway_default_and_noema_sidecar_fail_closed(tmp_path: Path) ->
     assert "sidecar must be provisioned before Noema LLM review" in noema.stdout
 
 
-def test_cancel_closed_pr_runs_has_a_bounded_runtime() -> None:
-    """cancel-closed-pr-runs must not fall back to GitHub's 360-minute default.
-
-    Its only step is a single-repository, status-filtered gh api --paginate
-    list-and-cancel sweep (up to 3 passes x 5 statuses) with no branch update
-    or merge -- comparable to, or lighter than, pr-review-merge-scheduler.yml's
-    scan-pr-queue job, which PR #1702 bounded to timeout-minutes: 30 for a
-    single-repository scan that also dispatches a review and updates a branch.
-    """
+def test_noema_jobs_do_not_invent_repository_wall_clock_deadlines() -> None:
+    """Noema model/support jobs must not encode hand-selected elapsed-time cutoffs."""
     workflow = workflow_text("noema-review.yml")
-    job = workflow.split("  cancel-closed-pr-runs:\n", 1)[1].split("\n  noema-review:\n", 1)[0]
-
-    match = re.search(r"^    timeout-minutes: (\d+)$", job, flags=re.MULTILINE)
-    assert match is not None, "cancel-closed-pr-runs must declare a job-level timeout-minutes"
-    timeout = int(match.group(1))
-    assert 1 <= timeout <= 30
-    assert timeout < 360
-
-
-def test_noema_review_job_has_a_bounded_runtime_above_the_two_hour_model_allowance() -> None:
-    """noema-review must not fall back to GitHub's 360-minute platform default.
-
-    Its "Prepare Noema model verdict" step calls into two_phase.py's
-    call_llm via the contextual-orchestrator gateway, which
-    noema_review_gate.py's own module comment says "remains governed by
-    contextual-orchestrator rather than a fixed inference timeout" -- so
-    nothing upstream of this job bounds that call. docs/product-goal-directive.md
-    section 8 documents that "중앙 OpenCode, Strix, Noema는 모델당 두 시간
-    이상 걸릴 수 있음을 수용한다" (central OpenCode, Strix, and Noema accept
-    that a model call may legitimately take over two hours), so the bound
-    must clear two hours (120 minutes) without falling back to GitHub's
-    360-minute job default.
-    """
-    workflow = workflow_text("noema-review.yml")
-    job = workflow.split("  noema-review:\n", 1)[1]
-
-    match = re.search(r"^    timeout-minutes: (\d+)$", job, flags=re.MULTILINE)
-    assert match is not None, "noema-review must declare a job-level timeout-minutes"
-    timeout = int(match.group(1))
-    assert 120 < timeout < 360
-
-    assert (
-        "모델당 두 시간 이상 걸릴 수 있음을 수용한다"
-        in (Path(__file__).resolve().parents[1] / "docs" / "product-goal-directive.md").read_text(
-            encoding="utf-8"
-        )
-    ), "the two-hour-per-model allowance this bound relies on must still be documented"
+    cleanup = workflow.split("  cancel-closed-pr-runs:\n", 1)[1].split(
+        "\n  noema-review:\n", 1
+    )[0]
+    model = workflow.split("  noema-review:\n", 1)[1]
+
+    assert re.search(r"^    timeout-minutes:", cleanup, flags=re.MULTILINE) is None
+    assert re.search(r"^    timeout-minutes:", model, flags=re.MULTILINE) is None
+    assert "contextual-orchestrator/orchestrator/free" in workflow
+    assert "unsupported hand-selected" in cleanup
+    assert "no repository-authored wall-clock" in model

From e64de22ba335116f388d9e5af0e4a29c8bdcc9bc Mon Sep 17 00:00:00 2001
From: Seongho Bae <me@seonghobae.me>
Date: Wed, 2 Sep 2026 19:49:10 +0900
Subject: [PATCH 09/10] ci(noema): use temporary REST-token surface to mark
 source-complete PR ready

---
 .github/workflows/_temp_mark_ready_pr1720.yml | 34 +++++++++++++++++++
 1 file changed, 34 insertions(+)
 create mode 100644 .github/workflows/_temp_mark_ready_pr1720.yml

diff --git a/.github/workflows/_temp_mark_ready_pr1720.yml b/.github/workflows/_temp_mark_ready_pr1720.yml
new file mode 100644
index 0000000000..7f970f8682
--- /dev/null
+++ b/.github/workflows/_temp_mark_ready_pr1720.yml
@@ -0,0 +1,34 @@
+name: Temporary Mark PR 1720 Ready
+
+on:
+  push:
+    branches:
+      - fix/noema-review-no-heuristic-timeouts
+    paths:
+      - .github/workflows/_temp_mark_ready_pr1720.yml
+
+permissions:
+  contents: read
+  pull-requests: write
+
+concurrency:
+  group: temp-mark-ready-pr1720
+  cancel-in-progress: true
+
+jobs:
+  mark-ready:
+    if: github.repository == 'ContextualWisdomLab/.github' && github.actor == 'seonghobae'
+    runs-on: ubuntu-slim
+    steps:
+      - name: Revalidate exact head and mark canonical source-complete PR ready
+        env:
+          GH_TOKEN: ${{ github.token }}
+          EXPECTED_HEAD: ${{ github.sha }}
+        shell: bash
+        run: |
+          set -euo pipefail
+          pr_json="$(gh pr view 1720 --repo ContextualWisdomLab/.github --json state,isDraft,headRefOid)"
+          test "$(jq -r '.state' <<<"$pr_json")" = "OPEN"
+          test "$(jq -r '.isDraft' <<<"$pr_json")" = "true"
+          test "$(jq -r '.headRefOid' <<<"$pr_json")" = "$EXPECTED_HEAD"
+          gh pr ready 1720 --repo ContextualWisdomLab/.github

From 9ef483b1c23f7e730a4469dbcf33d3a1db9366bc Mon Sep 17 00:00:00 2001
From: Seongho Bae <me@seonghobae.me>
Date: Wed, 2 Sep 2026 19:55:43 +0900
Subject: [PATCH 10/10] ci(noema): retire failed ready-state bridge

---
 .github/workflows/_temp_mark_ready_pr1720.yml | 34 -------------------
 1 file changed, 34 deletions(-)
 delete mode 100644 .github/workflows/_temp_mark_ready_pr1720.yml

diff --git a/.github/workflows/_temp_mark_ready_pr1720.yml b/.github/workflows/_temp_mark_ready_pr1720.yml
deleted file mode 100644
index 7f970f8682..0000000000
--- a/.github/workflows/_temp_mark_ready_pr1720.yml
+++ /dev/null
@@ -1,34 +0,0 @@
-name: Temporary Mark PR 1720 Ready
-
-on:
-  push:
-    branches:
-      - fix/noema-review-no-heuristic-timeouts
-    paths:
-      - .github/workflows/_temp_mark_ready_pr1720.yml
-
-permissions:
-  contents: read
-  pull-requests: write
-
-concurrency:
-  group: temp-mark-ready-pr1720
-  cancel-in-progress: true
-
-jobs:
-  mark-ready:
-    if: github.repository == 'ContextualWisdomLab/.github' && github.actor == 'seonghobae'
-    runs-on: ubuntu-slim
-    steps:
-      - name: Revalidate exact head and mark canonical source-complete PR ready
-        env:
-          GH_TOKEN: ${{ github.token }}
-          EXPECTED_HEAD: ${{ github.sha }}
-        shell: bash
-        run: |
-          set -euo pipefail
-          pr_json="$(gh pr view 1720 --repo ContextualWisdomLab/.github --json state,isDraft,headRefOid)"
-          test "$(jq -r '.state' <<<"$pr_json")" = "OPEN"
-          test "$(jq -r '.isDraft' <<<"$pr_json")" = "true"
-          test "$(jq -r '.headRefOid' <<<"$pr_json")" = "$EXPECTED_HEAD"
-          gh pr ready 1720 --repo ContextualWisdomLab/.github