diff --git a/docs/design/iter-slm252-lot2-01-not-authorized-20260725.json b/docs/design/iter-slm252-lot2-01-not-authorized-20260725.json
new file mode 100644
index 000000000..e07ebc2a2
--- /dev/null
+++ b/docs/design/iter-slm252-lot2-01-not-authorized-20260725.json
@@ -0,0 +1,26 @@
+{
+  "contract_hash": "b1a3a36449cad0aead49593a1894907ab283d44c6c692ce62c549231eacf8ce9",
+  "contract_id": "causal-supervision-routing-gate-v1",
+  "issue": {
+    "activation_requirement": "SLM-251 (LOT1-02) authorizes LOT2 and provides the selected parent, faithful treatment checkpoint/recipe, continued explicit control, curriculum, and fairness manifests.",
+    "alias": "LOT2-01",
+    "gate_contract_id": "causal-supervision-routing-gate-v1",
+    "linear_issue": "SLM-252",
+    "slug": "lot2-01-supervision-routing",
+    "title": "Run the causal supervision routing x timing x grounding factorial"
+  },
+  "lot1_01_disposition": "not_authorized",
+  "lot1_02_disposition": "not_authorized",
+  "schema_version": "lot_downstream_gate/v1",
+  "verdict": "not_authorized",
+  "verdict_rationale": "LOT2-01 (SLM-252) activation requires: SLM-251 (LOT1-02) authorizes LOT2 and provides the selected parent, faithful treatment checkpoint/recipe, continued explicit control, curriculum, and fairness manifests. The LOT1-02 launch gate reports 'not_authorized' (LOT1-01 disposition 'not_authorized'): no faithful K x c model path, curriculum, Stage 0 parent, or continued-explicit control exists, and SLM-249's oracle ceiling is not positive. Closing not_authorized in plan-only/bounded-diagnostic mode; no training, factorial, intervention, or systems code is added by this disposition.",
+  "version_stamp": {
+    "code_commit": "e484978344f756a606cc08e0cb5d7961d7447ad8",
+    "code_dirty": true,
+    "components": {
+      "harness.experiments": "v111"
+    },
+    "stamp_schema": "version_stamp/v1",
+    "stamped_at": "2026-07-25T16:05:28.682401+00:00"
+  }
+}
diff --git a/docs/design/iter-slm252-lot2-01-not-authorized-20260725.md b/docs/design/iter-slm252-lot2-01-not-authorized-20260725.md
new file mode 100644
index 000000000..88ac57d6a
--- /dev/null
+++ b/docs/design/iter-slm252-lot2-01-not-authorized-20260725.md
@@ -0,0 +1,15 @@
+# SLM-252 LOT2-01 — Downstream-gate disposition (causal-supervision-routing-gate-v1)
+
+Verdict: **not_authorized**
+
+LOT2-01 (SLM-252) activation requires: SLM-251 (LOT1-02) authorizes LOT2 and provides the selected parent, faithful treatment checkpoint/recipe, continued explicit control, curriculum, and fairness manifests. The LOT1-02 launch gate reports 'not_authorized' (LOT1-01 disposition 'not_authorized'): no faithful K x c model path, curriculum, Stage 0 parent, or continued-explicit control exists, and SLM-249's oracle ceiling is not positive. Closing not_authorized in plan-only/bounded-diagnostic mode; no training, factorial, intervention, or systems code is added by this disposition.
+
+## Upstream chain
+
+- LOT1-02 launch gate (SLM-251): `not_authorized`
+- LOT1-01 activation gate (SLM-250): `not_authorized`
+- Activation requirement: SLM-251 (LOT1-02) authorizes LOT2 and provides the selected parent, faithful treatment checkpoint/recipe, continued explicit control, curriculum, and fairness manifests.
+
+## Non-goals honored
+
+No training campaign, factorial, readout/decoder, intervention, systems measurement, or production default change. This disposition is itself the deliverable while the upstream activation gate is unmet.
diff --git a/scripts/evaluate_lot_downstream_gate.py b/scripts/evaluate_lot_downstream_gate.py
new file mode 100644
index 000000000..2d8970ed0
--- /dev/null
+++ b/scripts/evaluate_lot_downstream_gate.py
@@ -0,0 +1,81 @@
+#!/usr/bin/env python3
+"""Evaluate a LOT2+ issue's downstream activation gate in plan-only mode.
+
+No training, factorial, intervention, or systems code is loaded or run.
+Reads the real committed upstream contract artifacts and emits a per-issue
+``LotDownstreamGateV1`` disposition.
+
+Example:
+  python -m scripts.evaluate_lot_downstream_gate --issue SLM-252 \
+      --out outputs/runs/slm252_downstream_gate
+"""
+
+from __future__ import annotations
+
+import argparse
+import json
+from pathlib import Path
+
+from slm_training.harnesses.experiments.lot_downstream_gate import (
+    DOWNSTREAM_ISSUES,
+    evaluate_downstream_gate,
+    render_markdown,
+)
+from slm_training.harnesses.experiments.lot1_01_activation_gate import (
+    load_upstream_contract,
+)
+
+
+def main(argv: list[str] | None = None) -> int:
+    parser = argparse.ArgumentParser(
+        description="LOT2+ downstream launch-gate evaluator (plan-only, no campaign code)"
+    )
+    parser.add_argument(
+        "--issue",
+        required=True,
+        choices=sorted(DOWNSTREAM_ISSUES),
+        help="Linear issue id to evaluate (registry: DOWNSTREAM_ISSUES)",
+    )
+    parser.add_argument(
+        "--fidelity-contract",
+        type=Path,
+        default=Path("docs/design/lotus-openui-fidelity-contract-v1.json"),
+        help="SLM-248 LotusOpenUIFidelityContractV1 JSON artifact",
+    )
+    parser.add_argument(
+        "--trace-gate-contract",
+        type=Path,
+        default=Path("docs/design/compiler-reasoning-trace-v1.json"),
+        help="SLM-249 CompilerReasoningTraceGateV1 JSON artifact",
+    )
+    parser.add_argument(
+        "--out",
+        type=Path,
+        default=None,
+        help="Output directory (default outputs/runs/<issue>_downstream_gate)",
+    )
+    args = parser.parse_args(argv)
+
+    spec = DOWNSTREAM_ISSUES[args.issue]
+    out = args.out or Path("outputs/runs") / f"{args.issue.lower()}_downstream_gate"
+
+    contract = evaluate_downstream_gate(
+        spec,
+        load_upstream_contract(args.fidelity_contract),
+        load_upstream_contract(args.trace_gate_contract),
+    )
+
+    out.mkdir(parents=True, exist_ok=True)
+    stem = spec.slug.replace("-", "_")
+    (out / f"{stem}_gate.json").write_text(
+        json.dumps(contract.to_dict(), indent=2, sort_keys=True, default=str) + "\n",
+        encoding="utf-8",
+    )
+    markdown = render_markdown(contract)
+    (out / f"{stem}_gate.md").write_text(markdown, encoding="utf-8")
+    print(markdown)
+    return 0
+
+
+if __name__ == "__main__":
+    raise SystemExit(main())
diff --git a/src/slm_training/harnesses/experiments/lot_downstream_gate.py b/src/slm_training/harnesses/experiments/lot_downstream_gate.py
new file mode 100644
index 000000000..3b96ccfd5
--- /dev/null
+++ b/src/slm_training/harnesses/experiments/lot_downstream_gate.py
@@ -0,0 +1,186 @@
+"""Generic LOT-project downstream launch-gate evaluator (plan-only closeout).
+
+Every LOT2/LOT3/LOT4 issue (SLM-252 … SLM-258) declares the same shape of
+activation gate: the LOT1 chain must first authorize it — concretely, the
+SLM-251 (LOT1-02) launch gate must be met, which itself requires the
+SLM-248 transfer authorization and the SLM-249 oracle ceiling (evaluated
+by :mod:`slm_training.harnesses.experiments.lot1_02_activation_gate`).
+When the upstream chain is unmet, the downstream issue closes
+``not_authorized`` in plan-only/bounded-diagnostic mode with no training,
+factorial, intervention, or systems code.
+
+This module evaluates that shared upstream gate once and stamps a
+per-issue ``LotDownstreamGateV1`` disposition. It contains no model or
+campaign code — evaluating the gate honestly *is* the deliverable while
+the prerequisites are unmet.
+"""
+
+from __future__ import annotations
+
+import hashlib
+import json
+from dataclasses import asdict, dataclass, field
+from typing import Any
+
+from slm_training.harness_core.versioning import build_version_stamp
+from slm_training.harnesses.experiments.lot1_01_activation_gate import (
+    AUTHORIZED_WIRING_ONLY,
+    NOT_AUTHORIZED,
+)
+from slm_training.harnesses.experiments.lot1_02_activation_gate import (
+    evaluate_lot1_02_launch_gate,
+)
+
+__all__ = [
+    "DOWNSTREAM_ISSUES",
+    "GATE_SCHEMA_VERSION",
+    "PENDING_CAMPAIGN_GATE",
+    "DownstreamIssueSpec",
+    "LotDownstreamGateV1",
+    "evaluate_downstream_gate",
+    "render_markdown",
+]
+
+GATE_SCHEMA_VERSION = "lot_downstream_gate/v1"
+
+#: Verdict when the upstream contract chain is met: the issue's own
+#: activation still requires the LOT1-02 campaign's published gate
+#: (``FaithfulLatentTransferGateV1``), which does not exist until the
+#: campaign actually runs. This is never a quality or launch claim.
+PENDING_CAMPAIGN_GATE = "upstream_authorized_pending_campaign_gate"
+
+
+@dataclass(frozen=True)
+class DownstreamIssueSpec:
+    """Identity of one LOT2+ issue whose activation rides on the LOT1 chain."""
+
+    linear_issue: str
+    alias: str
+    title: str
+    gate_contract_id: str
+    slug: str
+    activation_requirement: str
+
+    def to_dict(self) -> dict[str, Any]:
+        return dict(asdict(self))
+
+
+DOWNSTREAM_ISSUES: dict[str, DownstreamIssueSpec] = {
+    "SLM-252": DownstreamIssueSpec(
+        linear_issue="SLM-252",
+        alias="LOT2-01",
+        title="Run the causal supervision routing x timing x grounding factorial",
+        gate_contract_id="causal-supervision-routing-gate-v1",
+        slug="lot2-01-supervision-routing",
+        activation_requirement=(
+            "SLM-251 (LOT1-02) authorizes LOT2 and provides the selected "
+            "parent, faithful treatment checkpoint/recipe, continued explicit "
+            "control, curriculum, and fairness manifests."
+        ),
+    ),
+}
+
+
+@dataclass(frozen=True)
+class LotDownstreamGateV1:
+    """Per-issue downstream disposition derived from the LOT1 launch gate."""
+
+    schema_version: str = GATE_SCHEMA_VERSION
+    contract_id: str = ""
+    issue: DownstreamIssueSpec | None = None
+    lot1_02_disposition: str = NOT_AUTHORIZED
+    lot1_01_disposition: str = NOT_AUTHORIZED
+    verdict: str = NOT_AUTHORIZED
+    verdict_rationale: str = ""
+    version_stamp: dict[str, Any] = field(default_factory=dict)
+
+    def to_dict(self) -> dict[str, Any]:
+        data = dict(asdict(self))
+        data["issue"] = self.issue.to_dict() if self.issue is not None else None
+        data["contract_hash"] = self.contract_hash()
+        return data
+
+    def contract_hash(self) -> str:
+        payload = {
+            "schema_version": self.schema_version,
+            "contract_id": self.contract_id,
+            "issue": self.issue.to_dict() if self.issue is not None else None,
+            "lot1_02_disposition": self.lot1_02_disposition,
+            "verdict": self.verdict,
+        }
+        return hashlib.sha256(
+            json.dumps(payload, sort_keys=True, default=str).encode("utf-8")
+        ).hexdigest()
+
+
+def evaluate_downstream_gate(
+    spec: DownstreamIssueSpec,
+    fidelity_contract: dict[str, Any],
+    trace_gate_contract: dict[str, Any],
+) -> LotDownstreamGateV1:
+    """Evaluate one LOT2+ issue's activation against real upstream contracts.
+
+    The verdict is ``not_authorized`` unless the LOT1-02 launch gate is
+    met; even then it is only ``upstream_authorized_pending_campaign_gate``
+    — the issue's own campaign gate must still be published by a real run
+    before any factorial/intervention/systems work is authorized.
+    """
+    lot1_02 = evaluate_lot1_02_launch_gate(fidelity_contract, trace_gate_contract)
+
+    if lot1_02.verdict == AUTHORIZED_WIRING_ONLY:
+        verdict = PENDING_CAMPAIGN_GATE
+        rationale = (
+            "The LOT1 contract-level prerequisites are met, but "
+            f"{spec.alias} ({spec.linear_issue}) still requires the LOT1-02 "
+            "campaign's published FaithfulLatentTransferGateV1 with a "
+            "positive authorization before any objective/architecture work "
+            "is authorized. No campaign has run; this is not a launch claim."
+        )
+    else:
+        verdict = NOT_AUTHORIZED
+        rationale = (
+            f"{spec.alias} ({spec.linear_issue}) activation requires: "
+            f"{spec.activation_requirement} The LOT1-02 launch gate reports "
+            f"'{lot1_02.verdict}' (LOT1-01 disposition "
+            f"'{lot1_02.lot1_01_disposition}'): no faithful K x c model "
+            "path, curriculum, Stage 0 parent, or continued-explicit control "
+            "exists, and SLM-249's oracle ceiling is not positive. Closing "
+            "not_authorized in plan-only/bounded-diagnostic mode; no "
+            "training, factorial, intervention, or systems code is added by "
+            "this disposition."
+        )
+
+    return LotDownstreamGateV1(
+        contract_id=spec.gate_contract_id,
+        issue=spec,
+        lot1_02_disposition=lot1_02.verdict,
+        lot1_01_disposition=lot1_02.lot1_01_disposition,
+        verdict=verdict,
+        verdict_rationale=rationale,
+        version_stamp=build_version_stamp("harness.experiments"),
+    )
+
+
+def render_markdown(contract: LotDownstreamGateV1) -> str:
+    spec = contract.issue
+    lines = [
+        f"# {spec.linear_issue} {spec.alias} — Downstream-gate disposition ({contract.contract_id})",
+        "",
+        f"Verdict: **{contract.verdict}**",
+        "",
+        contract.verdict_rationale,
+        "",
+        "## Upstream chain",
+        "",
+        f"- LOT1-02 launch gate (SLM-251): `{contract.lot1_02_disposition}`",
+        f"- LOT1-01 activation gate (SLM-250): `{contract.lot1_01_disposition}`",
+        f"- Activation requirement: {spec.activation_requirement}",
+        "",
+        "## Non-goals honored",
+        "",
+        "No training campaign, factorial, readout/decoder, intervention, "
+        "systems measurement, or production default change. This disposition "
+        "is itself the deliverable while the upstream activation gate is unmet.",
+        "",
+    ]
+    return "\n".join(lines)
diff --git a/src/slm_training/resources/versions.json b/src/slm_training/resources/versions.json
index 7e05f6716..ab75aa7b0 100644
--- a/src/slm_training/resources/versions.json
+++ b/src/slm_training/resources/versions.json
@@ -1912,7 +1912,7 @@
       ]
     },
     "harness.experiments": {
-      "version": "v117",
+      "version": "v118",
       "kind": "harness",
       "paths": [
         "scripts/run_slm298_capacity_context_curriculum.py",
@@ -1921,6 +1921,11 @@
         "tests/test_scripts/test_run_slm298_capacity_context_curriculum.py"
       ],
       "history": [
+        {
+          "version": "v118",
+          "date": "2026-07-25",
+          "note": "add generic LOT2+ downstream launch-gate evaluator (lot_downstream_gate.py) closing SLM-252 LOT2-01 not_authorized off the unmet LOT1-02 launch gate; no campaign code"
+        },
         {
           "version": "v117",
           "date": "2026-07-25",
@@ -1976,6 +1981,11 @@
           "date": "2026-07-25",
           "note": "add SLM-300 AP-015 self-context exposure-bias curriculum manifest, policy-origin-mixture wiring/fixture harness, and mixture-zero legacy-equivalence invariant"
         },
+        {
+          "version": "v111",
+          "date": "2026-07-25",
+          "note": "add generic LOT2+ downstream launch-gate evaluator (lot_downstream_gate.py) closing SLM-252 LOT2-01 not_authorized off the unmet LOT1-02 launch gate; no campaign code"
+        },
         {
           "version": "v110",
           "date": "2026-07-25",
@@ -9784,6 +9794,25 @@
           "note": "initial registration; SLM-251 LOT1-02 launch-gate evaluator closes not_authorized against the real SLM-248/SLM-249 upstream contracts and the LOT1-01 not_authorized disposition"
         }
       ]
+    },
+    "harness.experiments.lot_downstream_gate": {
+      "version": "v1",
+      "kind": "gate",
+      "paths": [
+        "src/slm_training/harnesses/experiments/lot_downstream_gate.py",
+        "scripts/evaluate_lot_downstream_gate.py",
+        "tests/test_harnesses/experiments/test_lot_downstream_gate.py",
+        "tests/test_scripts/test_evaluate_lot_downstream_gate.py",
+        "docs/design/iter-slm252-lot2-01-not-authorized-20260725.json",
+        "docs/design/iter-slm252-lot2-01-not-authorized-20260725.md"
+      ],
+      "history": [
+        {
+          "version": "v1",
+          "date": "2026-07-25",
+          "note": "initial registration; generic LOT2+ downstream gate closes SLM-252 LOT2-01 not_authorized against the real upstream contracts"
+        }
+      ]
     }
   }
 }
diff --git a/tests/test_harnesses/experiments/test_lot_downstream_gate.py b/tests/test_harnesses/experiments/test_lot_downstream_gate.py
new file mode 100644
index 000000000..8ccfca088
--- /dev/null
+++ b/tests/test_harnesses/experiments/test_lot_downstream_gate.py
@@ -0,0 +1,99 @@
+"""Tests for slm_training.harnesses.experiments.lot_downstream_gate (SLM-252+)."""
+
+from __future__ import annotations
+
+from pathlib import Path
+
+from slm_training.harnesses.experiments.lot1_01_activation_gate import (
+    AUTHORIZED_WIRING_ONLY,
+    NOT_AUTHORIZED,
+    REQUIRED_FIDELITY_VERDICT,
+    REQUIRED_TRACE_VERDICT,
+    load_upstream_contract,
+)
+from slm_training.harnesses.experiments.lot_downstream_gate import (
+    DOWNSTREAM_ISSUES,
+    PENDING_CAMPAIGN_GATE,
+    evaluate_downstream_gate,
+    render_markdown,
+)
+
+REPO_ROOT = Path(__file__).resolve().parents[3]
+FIDELITY_CONTRACT_PATH = REPO_ROOT / "docs/design/lotus-openui-fidelity-contract-v1.json"
+TRACE_GATE_CONTRACT_PATH = REPO_ROOT / "docs/design/compiler-reasoning-trace-v1.json"
+
+
+def _real_contracts() -> tuple[dict, dict]:
+    return (
+        load_upstream_contract(FIDELITY_CONTRACT_PATH),
+        load_upstream_contract(TRACE_GATE_CONTRACT_PATH),
+    )
+
+
+def _positive_contracts() -> tuple[dict, dict]:
+    fidelity = {
+        "contract_id": "synthetic-fidelity",
+        "contract_hash": "deadbeef",
+        "authorization": {
+            "linear_issue": "SLM-248",
+            "verdict": REQUIRED_FIDELITY_VERDICT,
+            "verdict_rationale": "synthetic positive",
+        },
+    }
+    trace = {
+        "contract_id": "synthetic-trace",
+        "contract_hash": "cafef00d",
+        "gate": {
+            "linear_issue": "SLM-249",
+            "verdict": REQUIRED_TRACE_VERDICT,
+            "verdict_rationale": "synthetic positive",
+            "allowed_lot1_implementation": [
+                "kxc_bounded_implementation: K=6, c=479, stages per contract"
+            ],
+        },
+    }
+    return fidelity, trace
+
+
+def test_slm252_real_contracts_are_not_authorized() -> None:
+    """The currently-committed SLM-248/SLM-249 artifacts must yield
+    not_authorized for LOT2-01: LOT1-02's launch gate is unmet, so no
+    parent checkpoint, treatment recipe, control, or curriculum exists.
+    """
+    fidelity, trace = _real_contracts()
+    contract = evaluate_downstream_gate(DOWNSTREAM_ISSUES["SLM-252"], fidelity, trace)
+
+    assert contract.verdict == NOT_AUTHORIZED
+    assert contract.lot1_02_disposition == NOT_AUTHORIZED
+    assert contract.lot1_01_disposition == NOT_AUTHORIZED
+    assert contract.contract_id == "causal-supervision-routing-gate-v1"
+
+
+def test_positive_upstream_is_pending_not_authorized() -> None:
+    """Even with a synthetic positive LOT1 chain, a LOT2+ issue is only
+    upstream_authorized_pending_campaign_gate — its own campaign gate must
+    still be published by a real run. The evaluator never jumps straight
+    to a launch/quality authorization.
+    """
+    fidelity, trace = _positive_contracts()
+    contract = evaluate_downstream_gate(DOWNSTREAM_ISSUES["SLM-252"], fidelity, trace)
+
+    assert contract.verdict == PENDING_CAMPAIGN_GATE
+    assert contract.lot1_02_disposition == AUTHORIZED_WIRING_ONLY
+
+
+def test_registry_entries_are_well_formed() -> None:
+    for issue_id, spec in DOWNSTREAM_ISSUES.items():
+        assert spec.linear_issue == issue_id
+        assert spec.gate_contract_id
+        assert spec.slug
+        assert spec.activation_requirement
+
+
+def test_render_markdown_covers_chain() -> None:
+    fidelity, trace = _real_contracts()
+    contract = evaluate_downstream_gate(DOWNSTREAM_ISSUES["SLM-252"], fidelity, trace)
+    markdown = render_markdown(contract)
+    assert "SLM-252" in markdown
+    assert "not_authorized" in markdown
+    assert "LOT1-02" in markdown
diff --git a/tests/test_scripts/test_evaluate_lot_downstream_gate.py b/tests/test_scripts/test_evaluate_lot_downstream_gate.py
new file mode 100644
index 000000000..b7d26517e
--- /dev/null
+++ b/tests/test_scripts/test_evaluate_lot_downstream_gate.py
@@ -0,0 +1,39 @@
+"""Tests for scripts/evaluate_lot_downstream_gate.py (SLM-252+)."""
+
+from __future__ import annotations
+
+import json
+from pathlib import Path
+
+from scripts import evaluate_lot_downstream_gate
+
+
+def test_slm252_default_run_reports_not_authorized(tmp_path: Path) -> None:
+    out = tmp_path / "gate"
+    rc = evaluate_lot_downstream_gate.main(["--issue", "SLM-252", "--out", str(out)])
+    assert rc == 0
+
+    data = json.loads((out / "lot2_01_supervision_routing_gate.json").read_text())
+    assert data["verdict"] == "not_authorized"
+    assert data["issue"]["linear_issue"] == "SLM-252"
+    assert (out / "lot2_01_supervision_routing_gate.md").exists()
+
+
+def test_run_against_explicit_contract_paths(tmp_path: Path) -> None:
+    out = tmp_path / "gate_explicit"
+    rc = evaluate_lot_downstream_gate.main(
+        [
+            "--issue",
+            "SLM-252",
+            "--fidelity-contract",
+            "docs/design/lotus-openui-fidelity-contract-v1.json",
+            "--trace-gate-contract",
+            "docs/design/compiler-reasoning-trace-v1.json",
+            "--out",
+            str(out),
+        ]
+    )
+    assert rc == 0
+    markdown = (out / "lot2_01_supervision_routing_gate.md").read_text()
+    assert "SLM-252" in markdown
+    assert "not_authorized" in markdown