fix: preserve unknown observed model evidence

Assistant: codex
Assistant-Model: gpt-6.1-sol
Assistant-Session: 01a10238-7ecc-78e2-903d-075f50e913cc

docs/evidence/2026-10-03-emergent-session-readiness-review.md

@@ -0,0 +1,38 @@
+# EMERGENT-WP-0002: session proof and readiness review
+
+Knowledge 0.2.1; package digest d8ef41bacd2c6551ae3b87628004a5df6943928be1c93eb01e09efc6ad782767; emergent-security commit f9cba6fb35425f74e34401b1baf09f6057ace493.
+Scoped technical review by delegated Codex worker, 2026-10-03. No profile or
+credential activated and no production request made.
+
+GLAS-WP-0008 is finished: blocked profiles refuse before sandbox creation;
+unverified means a labeled bounded attempt; ready requires positive evidence.
+GLAS-WP-0012 remains blocked on the owner delivery and real-provider acceptance
+already recorded in T02–T05. A direct Codex/Claude/Grok CLI comprehension run
+cannot close those gates or establish credential isolation.
+
+Current contract supports start_session, dispatch_tool, end_session and cleanup;
+it supplies a generated task file and opaque bounded rein run. There is no
+provider resume, compaction or provider-switch event API in this boundary.
+The generic task description can request orientation loading, but it is no
+attestation that the provider loaded or retained knowledge. Session state in
+the Python adapter is not a provider conversation/session identifier.
+
+Local correction: configured model pins no longer fill missing resolved_model
+in gateway or either adapter. Provider/rein-reported model remains evidence;
+absence remains null. A rein report may still be an alias, not an exact model
+revision. No session identifier or exact revision is fabricated from a profile.
+Tests check missing reports and distinct observed models on both adapters.
+
+Acceptance requirements retained in GLAS-WP-0012 T04: after owner readiness,
+record profile/runtime/rein versions, provider-reported exact model and session
+ID (unknown if unavailable), package digest and source status, startup retrieval
+and comprehension, then provider-observed resume/compaction and repository /
+provider transitions with post-event reload receipts and scoped comprehension.
+Do not simulate an event and label it actual continuity. Preserve failed runs
+and proof scope. Until native event evidence exists, EMERGENT T05/T07 remains
+incomplete. These conditions extend existing proof requirements, not readiness.
+
+Validation: focused gateway/adapter tests 68 passed; full `tests/` suite
+139 passed with `.venv/bin/python -m pytest tests/ -q -p no:cacheprovider`.
+`git diff --check` passed. These are deterministic regression tests, not
+native provider continuity, readiness or isolation proof.

docs/harness-contract.md

@@ -178,3 +178,12 @@ of the returned sandbox. Errors preparing its reachability or SandboxHandle
 fail at session_start, with destruction still attempted. Readiness here is the
 owner lifecycle state; it does not promote a Glas profile's operational
 readiness or independently prove runtime/provider health.
+
+
+## Observed model evidence (2026-10-03)
+
+`resolved_model` is supplied only by a rein execution report. A configured
+model pin is a requested route, and does not fill absent observed metadata.
+Unknown remains null, including failure before provider execution. A reported
+alias does not attest an exact provider revision or conversation identifier.
+See `docs/evidence/2026-10-03-emergent-session-readiness-review.md`.

src/glas_harness/gateway.py

@@ -308,7 +308,6 @@ def _build_result(
     resolved_model = (
         (summary.resolved_model if summary else None)
         or (tool_result.resolved_model if tool_result else None)
-        or (profile.model.model if profile else None)
     )
     tokens_spent = (
         summary.tokens_spent if summary and summary.tokens_spent is not None

src/glas_harness/reins/rein_aharness.py

@@ -122,7 +122,7 @@ class ReinAharness(Rein):
             tokens_spent=last_result.get("tokens_spent"),
             cost_usd=last_result.get("cost_usd"),
             duration_s=last_result.get("execution_time_s"),
-            resolved_model=last_result.get("model") or self.model,
+            resolved_model=last_result.get("model"),
             metadata={
                 "tool_profile": last_result.get("tool_profile") or self.tool_profile,
             },
@@ -167,7 +167,7 @@ class ReinAharness(Rein):
             tokens_spent=last_result.get("tokens_spent"),
             cost_usd=last_result.get("cost_usd"),
             duration_s=last_result.get("execution_time_s"),
-            resolved_model=last_result.get("model") or self.model,
+            resolved_model=last_result.get("model"),
             artifacts=[head_after] if committed and head_after else [],
             metadata={
                 "tool_profile": last_result.get("tool_profile") or self.tool_profile,

src/glas_harness/reins/rein_openweights.py

@@ -101,7 +101,7 @@ class ReinOpenWeights(Rein):
             events_completeness="unavailable",
             tokens_spent=last_result.get("tokens_spent"),
             duration_s=last_result.get("execution_time_s"),
-            resolved_model=last_result.get("model") or self.model,
+            resolved_model=last_result.get("model"),
             metadata={
                 "turns": last_result.get("turns"),
                 "tool_profile": last_result.get("tool_profile") or self.tool_profile,
@@ -125,7 +125,7 @@ class ReinOpenWeights(Rein):
             reason=reason,
             tokens_spent=last_result.get("tokens_spent"),
             duration_s=last_result.get("execution_time_s"),
-            resolved_model=last_result.get("model") or self.model,
+            resolved_model=last_result.get("model"),
             artifacts=[head_after] if committed and head_after else [],
             metadata={
                 "turns": last_result.get("turns"),

tests/test_gateway.py

@@ -597,3 +597,21 @@ def test_cleanup_failure_never_captures_artifact():
     assert not result.ok
     capture.assert_not_called()
     manager.destroy.assert_called_once()
+
+
+def test_missing_provider_model_does_not_become_configured_pin() -> None:
+    class UnknownModelRein(_FakeRein):
+        def dispatch_tool(self, session, tool_call):
+            return ToolResult(ok=True)
+
+        def end_session(self, session):
+            return ExecutionSummary(outcome="succeeded")
+
+    manager = MagicMock()
+    manager.create.return_value = _fake_status()
+    result = run_execution(
+        _request(), catalog=_catalog_with_readiness(),
+        rein=UnknownModelRein(), manager=manager,
+    )
+    assert result.evidence.resolved_model is None
+    assert result.evidence.model_route is not None

tests/test_rein_aharness.py

@@ -226,3 +226,20 @@ def test_profile_native_limits_are_carried_in_owned_task_file():
         assert transport.write_task_file.call_args.args[0]["max_turns"] == 3
         with pytest.raises(ValueError,match="Glas-generated"):
             rein.start_session(profile,{"task_file":"/untrusted.json"},object())
+
+
+@pytest.mark.parametrize("reported", [None, "provider-observed-model"])
+def test_resolved_model_requires_rein_report(reported) -> None:
+    rein = ReinAharness(model="configured-pin")
+    transport = MagicMock(spec=ExecutionTransport)
+    payload = {"ok": True}
+    if reported is not None:
+        payload["model"] = reported
+    transport.run.return_value = MagicMock(returncode=0, stdout=json.dumps(payload), stderr="")
+    transport.git_head.return_value = "new-head"
+    session = {"task_file": "/tmp/task.json", "transport": transport,
+               "timeout_seconds": 30, "head_before": "old-head"}
+    with patch.object(rein, "_bin", return_value="/usr/bin/rein_aharness"):
+        result = rein.dispatch_tool(session, ToolCall(name="run_task"))
+    assert result.resolved_model == reported
+    assert rein.end_session(session).resolved_model == reported

tests/test_rein_openweights.py

@@ -174,3 +174,20 @@ def test_cleanup_session_removes_generated_task_file() -> None:
         {"transport": transport, "task_file": "/sandbox/task.json", "generated_task_file": True}
     )
     transport.remove_file.assert_called_once_with("/sandbox/task.json")
+
+
+@pytest.mark.parametrize("reported", [None, "provider-observed-model"])
+def test_resolved_model_requires_rein_report(reported) -> None:
+    rein = ReinOpenWeights(model="configured-pin")
+    transport = MagicMock(spec=ExecutionTransport)
+    payload = {"ok": True}
+    if reported is not None:
+        payload["model"] = reported
+    transport.run.return_value = MagicMock(returncode=0, stdout=json.dumps(payload), stderr="")
+    transport.git_head.return_value = "new-head"
+    session = {"task_file": "/tmp/task.json", "transport": transport,
+               "timeout_seconds": 30, "head_before": "old-head"}
+    with patch.object(rein, "_bin", return_value="/usr/bin/rein_openweights"):
+        result = rein.dispatch_tool(session, ToolCall(name="run_task"))
+    assert result.resolved_model == reported
+    assert rein.end_session(session).resolved_model == reported

workplans/GLAS-WP-0012-first-local-profile-production-proof.md

@@ -412,3 +412,14 @@ started through its owner on railiance01, SECRETS-WP-0009-T03 step B2) and
 T04/T05 all require that external owner delivery. Nothing in this repo can
 advance them today; `status: blocked` stands unchanged. Re-review when the
 runtime rebuild/placement or owner authentication lands.
+
+
+## EMERGENT-WP-0002 proof requirements — 2026-10-03
+
+Scoped review and local observed-model correction:
+`docs/evidence/2026-10-03-emergent-session-readiness-review.md`.
+T04 retains actual native-provider startup and continuity acceptance, including
+reload after real resume/compaction, provider/repository changes and exact
+available model/session metadata. The current lifecycle exposes no continuity
+event API. Existing owner-delivery blockers and all task/readiness states remain;
+local deterministic tests are not governed provider or isolation acceptance.