test: add EdgeEnv replay warning fixtures

README.md

@@ -94,6 +94,8 @@ edgeenv report compare <run_id_a> <run_id_b>
 ```
 
 For the full flow, see [Compare Workflow Guide](docs/compare-workflow-guide.md).
+The guide also links small committed runtime regression replay fixtures under
+`examples/regression/` for downstream AIGuard/Lab handoff checks.
 
 ### 4. Optional Resource And Sampler Evidence
 

docs/compare-workflow-guide.md

@@ -116,6 +116,20 @@ If a runtime telemetry history artifact is available, pass
 regression report. Telemetry context remains supplemental evidence; it never
 bypasses the same-condition comparability gate.
 
+Committed replay-context examples are available when downstream tools need a
+small EdgeEnv-owned fixture without running a benchmark:
+
+- `examples/regression/edgeenv_candidate_telemetry_gap.json` shows a comparable
+  same-condition report where the candidate run is missing runtime telemetry in
+  both the result artifact and telemetry history.
+- `examples/regression/edgeenv_sequence_inversion.json` shows a comparable
+  same-condition report where baseline/candidate `execution_sequence_id` order
+  is inverted in the replay context.
+
+These examples intentionally do not include `guard_analysis` or a deployment
+decision. EdgeEnv owns the registry, replay context, comparability judgement,
+and regression evidence; AIGuard and Lab consume the artifact later.
+
 Default starter thresholds:
 
 | Signal | Threshold | Meaning |

examples/regression/edgeenv_candidate_telemetry_gap.json

@@ -0,0 +1,92 @@
+{
+  "baseline_run_id": "baseline",
+  "candidate_run_id": "candidate",
+  "regression_detected": false,
+  "regression_type": "none",
+  "severity": "none",
+  "comparable": true,
+  "mode": "same-condition",
+  "evidence": {
+    "mean_delta_pct": 8.0,
+    "p95_delta_pct": 10.0,
+    "p99_delta_pct": 12.0,
+    "fps_delta_pct": -5.0,
+    "memory_peak_delta_pct": 4.0,
+    "baseline": {
+      "latency_mean_ms": 100.0,
+      "latency_p95_ms": 120.0,
+      "latency_p99_ms": 130.0,
+      "throughput_fps": 50.0,
+      "memory_peak_mb": 512.0
+    },
+    "candidate": {
+      "latency_mean_ms": 108.0,
+      "latency_p95_ms": 132.0,
+      "latency_p99_ms": 145.6,
+      "throughput_fps": 47.5,
+      "memory_peak_mb": 532.48
+    },
+    "triggered_thresholds": []
+  },
+  "recommendation": "no_action_required",
+  "comparability": {
+    "comparable": "Yes",
+    "mode": "same-condition",
+    "reasons": [
+      "Required comparison fields match.",
+      "Runtime/provider/target fields match."
+    ]
+  },
+  "runtime_telemetry_context": {
+    "role": "supplemental_runtime_telemetry_context",
+    "source": "result_artifacts+runtime_telemetry_history",
+    "baseline": {
+      "run_id": "baseline",
+      "result_telemetry_present": true,
+      "history_entry_present": true,
+      "history_missing_recorded": false,
+      "schema_version": "edgeenv.runtime-telemetry.v1",
+      "telemetry_timestamp": "2026-05-22T00:00:01Z",
+      "execution_sequence_id": 1,
+      "telemetry_source": "synthetic_local_fixture",
+      "available_sections": [
+        "execution_sequence_id",
+        "latency",
+        "resource",
+        "telemetry_timestamp"
+      ],
+      "history_telemetry_timestamp": "2026-05-22T00:00:01Z",
+      "history_execution_sequence_id": 1
+    },
+    "candidate": {
+      "run_id": "candidate",
+      "result_telemetry_present": false,
+      "history_entry_present": false,
+      "history_missing_recorded": true,
+      "history_missing_reason": "runtime_telemetry_missing"
+    },
+    "evidence_gaps": [
+      {
+        "run_id": "candidate",
+        "reason": "runtime_telemetry_missing_in_result"
+      },
+      {
+        "run_id": "candidate",
+        "reason": "runtime_telemetry_missing"
+      }
+    ],
+    "notes": [
+      "Runtime telemetry context is supplemental evidence, not a comparability gate.",
+      "Missing telemetry is an evidence gap, not a failed benchmark run.",
+      "Regression deltas are still gated by same-condition comparability."
+    ],
+    "history": {
+      "schema_version": "edgeenv.runtime-telemetry-history.v1",
+      "summary": {
+        "registered_runs": 3,
+        "telemetry_runs": 2,
+        "missing_telemetry_runs": 1
+      }
+    }
+  }
+}

examples/regression/edgeenv_sequence_inversion.json

@@ -0,0 +1,94 @@
+{
+  "baseline_run_id": "baseline",
+  "candidate_run_id": "candidate",
+  "regression_detected": false,
+  "regression_type": "none",
+  "severity": "none",
+  "comparable": true,
+  "mode": "same-condition",
+  "evidence": {
+    "mean_delta_pct": 12.0,
+    "p95_delta_pct": 4.166667,
+    "p99_delta_pct": 3.846154,
+    "fps_delta_pct": -4.0,
+    "memory_peak_delta_pct": 5.0,
+    "baseline": {
+      "latency_mean_ms": 100.0,
+      "latency_p95_ms": 120.0,
+      "latency_p99_ms": 130.0,
+      "throughput_fps": 50.0,
+      "memory_peak_mb": 512.0
+    },
+    "candidate": {
+      "latency_mean_ms": 112.0,
+      "latency_p95_ms": 125.0,
+      "latency_p99_ms": 135.0,
+      "throughput_fps": 48.0,
+      "memory_peak_mb": 537.6
+    },
+    "triggered_thresholds": []
+  },
+  "recommendation": "no_action_required",
+  "comparability": {
+    "comparable": "Yes",
+    "mode": "same-condition",
+    "reasons": [
+      "Required comparison fields match.",
+      "Runtime/provider/target fields match."
+    ]
+  },
+  "runtime_telemetry_context": {
+    "role": "supplemental_runtime_telemetry_context",
+    "source": "result_artifacts+runtime_telemetry_history",
+    "baseline": {
+      "run_id": "baseline",
+      "result_telemetry_present": true,
+      "history_entry_present": true,
+      "history_missing_recorded": false,
+      "schema_version": "edgeenv.runtime-telemetry.v1",
+      "telemetry_timestamp": "2026-05-22T00:00:05Z",
+      "execution_sequence_id": 5,
+      "telemetry_source": "synthetic_local_fixture",
+      "available_sections": [
+        "execution_sequence_id",
+        "latency",
+        "resource",
+        "telemetry_timestamp"
+      ],
+      "history_telemetry_timestamp": "2026-05-22T00:00:05Z",
+      "history_execution_sequence_id": 5
+    },
+    "candidate": {
+      "run_id": "candidate",
+      "result_telemetry_present": true,
+      "history_entry_present": true,
+      "history_missing_recorded": false,
+      "schema_version": "edgeenv.runtime-telemetry.v1",
+      "telemetry_timestamp": "2026-05-22T00:00:02Z",
+      "execution_sequence_id": 2,
+      "telemetry_source": "synthetic_local_fixture",
+      "available_sections": [
+        "execution_sequence_id",
+        "latency",
+        "resource",
+        "telemetry_timestamp"
+      ],
+      "history_telemetry_timestamp": "2026-05-22T00:00:02Z",
+      "history_execution_sequence_id": 2
+    },
+    "evidence_gaps": [],
+    "notes": [
+      "Runtime telemetry context is supplemental evidence, not a comparability gate.",
+      "Missing telemetry is an evidence gap, not a failed benchmark run.",
+      "Regression deltas are still gated by same-condition comparability."
+    ],
+    "history": {
+      "schema_version": "edgeenv.runtime-telemetry-history.v1",
+      "summary": {
+        "registered_runs": 2,
+        "telemetry_runs": 2,
+        "missing_telemetry_runs": 0
+      }
+    }
+  }
+}

tests/test_regression.py

@@ -16,6 +16,9 @@
 from helpers import make_result
 
 
+EXAMPLE_REGRESSION_DIR = Path("examples/regression")
+
+
 def test_regression_detects_same_condition_latency_and_resource_regression(
     bench_config,
     target_profile,
@@ -629,6 +632,58 @@ def test_cli_telemetry_replay_candidate_gap_to_regression_smoke(
     assert "runtime_telemetry_missing" in markdown
 
 
+def test_committed_replay_warning_fixtures_preserve_edgeenv_owned_context():
+    candidate_gap = json.loads(
+        (EXAMPLE_REGRESSION_DIR / "edgeenv_candidate_telemetry_gap.json").read_text(
+            encoding="utf-8"
+        )
+    )
+    sequence_inversion = json.loads(
+        (EXAMPLE_REGRESSION_DIR / "edgeenv_sequence_inversion.json").read_text(
+            encoding="utf-8"
+        )
+    )
+
+    assert "guard_analysis" not in candidate_gap
+    assert "guard_analysis" not in sequence_inversion
+    assert candidate_gap["comparable"] is True
+    assert sequence_inversion["mode"] == "same-condition"
+    assert candidate_gap["regression_detected"] is False
+    assert sequence_inversion["regression_detected"] is False
+
+    candidate_context = candidate_gap["runtime_telemetry_context"]
+    assert candidate_context["source"] == "result_artifacts+runtime_telemetry_history"
+    assert candidate_context["history"]["summary"]["missing_telemetry_runs"] == 1
+    assert candidate_context["candidate"]["result_telemetry_present"] is False
+    assert candidate_context["candidate"]["history_entry_present"] is False
+    assert candidate_context["candidate"]["history_missing_recorded"] is True
+    assert candidate_context["candidate"]["history_missing_reason"] == (
+        "runtime_telemetry_missing"
+    )
+    assert candidate_context["evidence_gaps"] == [
+        {
+            "run_id": "candidate",
+            "reason": "runtime_telemetry_missing_in_result",
+        },
+        {
+            "run_id": "candidate",
+            "reason": "runtime_telemetry_missing",
+        },
+    ]
+
+    sequence_context = sequence_inversion["runtime_telemetry_context"]
+    assert sequence_context["history"]["summary"]["missing_telemetry_runs"] == 0
+    assert sequence_context["baseline"]["execution_sequence_id"] == 5
+    assert sequence_context["baseline"]["history_execution_sequence_id"] == 5
+    assert sequence_context["candidate"]["execution_sequence_id"] == 2
+    assert sequence_context["candidate"]["history_execution_sequence_id"] == 2
+    assert sequence_context["evidence_gaps"] == []
+    assert any(
+        "not a comparability gate" in note
+        for note in sequence_context["notes"]
+    )
+
+
 def test_regression_cli_marks_runtime_comparison_not_evaluated(
     tmp_path,
     bench_config,