QuantStrategyLab · Pigbibi · May 10, 2026 · May 10, 2026
@@ -6,13 +6,6 @@ name: Monthly Optimization Planner
       upstream_run_id:
         description: "AI review run id from CryptoSnapshotPipelines"
         required: true
-      downstream_run_id:
-        description: "AI review run id from BinancePlatform"
-        required: true
-      downstream_repo:
-        description: "Downstream execution repo"
-        required: true
-        default: "QuantStrategyLab/BinancePlatform"
 
 jobs:
   planner:
@@ -50,7 +43,6 @@ jobs:
           owner: ${{ github.repository_owner }}
           repositories: |
             CryptoSnapshotPipelines
-            BinancePlatform
             CryptoStrategies
           permission-actions: write
           permission-issues: write
@@ -81,27 +73,16 @@ jobs:
             --repo "${GITHUB_REPOSITORY}" \
             --dir data/input/upstream
 
-      - name: Download downstream AI review artifact
-        env:
-          GH_TOKEN: ${{ steps.cross_repo_token.outputs.token }}
-        run: |
-          mkdir -p data/input/downstream
-          gh run download "${{ inputs.downstream_run_id }}" \
-            --repo "${{ inputs.downstream_repo }}" \
-            --dir data/input/downstream
-
       - name: Resolve downloaded artifact paths
         id: artifact_paths
         run: |
           UPSTREAM_DIR=$(find data/input/upstream -mindepth 1 -maxdepth 1 -type d | head -1)
-          DOWNSTREAM_DIR=$(find data/input/downstream -mindepth 1 -maxdepth 1 -type d | head -1)
-          if [ -z "${UPSTREAM_DIR}" ] || [ -z "${DOWNSTREAM_DIR}" ]; then
-            echo "Failed to resolve downloaded artifact directories" >&2
+          if [ -z "${UPSTREAM_DIR}" ]; then
+            echo "Failed to resolve downloaded upstream artifact directory" >&2
             exit 1
           fi
           {
             echo "upstream_dir=${UPSTREAM_DIR}"
-            echo "downstream_dir=${DOWNSTREAM_DIR}"
           } >> "${GITHUB_OUTPUT}"
 
       - name: Prepare upstream review payload
@@ -114,21 +95,10 @@ jobs:
             --run-url "${GITHUB_SERVER_URL}/${GITHUB_REPOSITORY}/actions/runs/${{ inputs.upstream_run_id }}" \
             --output-file data/output/prepared/upstream_review_payload.json
 
-      - name: Prepare downstream review payload
-        run: |
-          python3 scripts/build_ai_review_payload.py \
-            --source-repo "${{ inputs.downstream_repo }}" \
-            --review-kind execution_runtime \
-            --issue-context-file "${{ steps.artifact_paths.outputs.downstream_dir }}/issue_context.json" \
-            --secondary-review-file "${{ steps.artifact_paths.outputs.downstream_dir }}/secondary_review.json" \
-            --run-url "${GITHUB_SERVER_URL}/${{ inputs.downstream_repo }}/actions/runs/${{ inputs.downstream_run_id }}" \
-            --output-file data/output/prepared/downstream_review_payload.json
-
       - name: Build monthly optimization plan
         run: |
           python3 scripts/build_monthly_optimization_plan.py \
             --upstream-review-file data/output/prepared/upstream_review_payload.json \
-            --downstream-review-file data/output/prepared/downstream_review_payload.json \
             --output-dir data/output/monthly_optimization
 
       - name: Append optimization summary
@@ -174,18 +144,6 @@ jobs:
         env:
           GITHUB_TOKEN: ${{ steps.cross_repo_token.outputs.token }}
 
-      - name: Fan out BinancePlatform task issue
-        run: |
-          python3 scripts/fanout_monthly_optimization_tasks.py \
-            --plan-file data/output/monthly_optimization/optimization_plan.json \
-            --owner-repo BinancePlatform \
-            --repo "${{ inputs.downstream_repo }}" \
-            --planner-issue-url "${GITHUB_SERVER_URL}/${GITHUB_REPOSITORY}/issues/${{ steps.optimization_issue.outputs.issue_number }}" \
-            --output-file data/output/monthly_optimization/fanout/binance_platform.json \
-            --allow-permission-skip
-        env:
-          GITHUB_TOKEN: ${{ steps.cross_repo_token.outputs.token }}
-
       - name: Append fanout summary
         run: |
           python3 - <<'PY' >> "$GITHUB_STEP_SUMMARY"
@@ -235,63 +193,8 @@ jobs:
           gh workflow run experiment_validation.yml \
             -f issue_number="${{ steps.upstream_experiment_target.outputs.issue_number }}"
 
-      - name: Resolve downstream experiment validation target
-        id: downstream_experiment_target
-        run: |
-          python3 - <<'PY'
-          import json
-          import os
-          from pathlib import Path
-
-          fanout = json.loads(
-              Path("data/output/monthly_optimization/fanout/binance_platform.json").read_text(encoding="utf-8")
-          )
-          plan = json.loads(
-              Path("data/output/monthly_optimization/optimization_plan.json").read_text(encoding="utf-8")
-          )
-          actions = plan.get("repo_action_summary", {}).get("BinancePlatform", {}).get("actions", [])
-          should_dispatch = bool(fanout.get("issue_number")) and fanout.get("status") in {"created", "updated"} and any(
-              action.get("experiment_only") for action in actions
-          )
-          with open(os.environ["GITHUB_OUTPUT"], "a", encoding="utf-8") as output:
-              print(f"should_dispatch={'true' if should_dispatch else 'false'}", file=output)
-              print(f"issue_number={fanout.get('issue_number') or ''}", file=output)
-          PY
-
-      - name: Best-effort label BinancePlatform issue for experiment validation
-        if: steps.downstream_experiment_target.outputs.should_dispatch == 'true'
-        env:
-          GH_TOKEN: ${{ steps.cross_repo_token.outputs.token }}
-          TARGET_REPO: ${{ inputs.downstream_repo }}
-          ISSUE_NUMBER: ${{ steps.downstream_experiment_target.outputs.issue_number }}
-        run: |
-          set +e
-          gh label create experiment-validation             --repo "$TARGET_REPO"             --color 1D76DB             --description "Trigger experiment validation for monthly optimization tasks"             --force
-          label_status=$?
-          gh issue edit "$ISSUE_NUMBER"             --repo "$TARGET_REPO"             --add-label experiment-validation
-          issue_status=$?
-          set -e
-          if [ "$label_status" -ne 0 ] || [ "$issue_status" -ne 0 ]; then
-            echo "Downstream experiment-validation label update skipped for $TARGET_REPO#$ISSUE_NUMBER." >> "$GITHUB_STEP_SUMMARY"
-          fi
-
-      - name: Dispatch BinancePlatform experiment validation
-        if: steps.downstream_experiment_target.outputs.should_dispatch == 'true'
-        env:
-          GH_TOKEN: ${{ steps.cross_repo_token.outputs.token }}
-        run: |
-          set +e
-          gh workflow run experiment_validation.yml \
-            -R "${{ inputs.downstream_repo }}" \
-            -f issue_number="${{ steps.downstream_experiment_target.outputs.issue_number }}"
-          status=$?
-          set -e
-          if [ "$status" -ne 0 ]; then
-            echo "Downstream experiment validation dispatch skipped: ${{ steps.cross_repo_token.outputs.source }} token likely needs Actions write on ${{ inputs.downstream_repo }}." >> "$GITHUB_STEP_SUMMARY"
-          fi
-
       - name: Upload planner artifact
         uses: actions/upload-artifact@v7
         with:
-          name: monthly-optimization-plan-${{ inputs.upstream_run_id }}-${{ inputs.downstream_run_id }}
+          name: monthly-optimization-plan-${{ inputs.upstream_run_id }}
           path: data/output/monthly_optimization/
@@ -9,7 +9,6 @@
 SCHEMA_VERSION = "2026-04-02"
 REPO_ROLE_BY_KIND = {
     "upstream_selector": "upstream_selector_review",
-    "execution_runtime": "execution_runtime_review",
 }
 
 

@@ -10,7 +10,7 @@
 
 RISK_ORDER = {"low": 0, "medium": 1, "high": 2}
 SCHEMA_VERSION = "2026-04-02"
-REPO_ORDER = ["CryptoSnapshotPipelines", "CryptoStrategies", "BinancePlatform"]
+REPO_ORDER = ["CryptoSnapshotPipelines", "CryptoStrategies"]
 MANUAL_REVIEW_PREFIXES = (
     "check ",
     "review ",
@@ -30,26 +30,6 @@ def _combined_action_text(action: dict[str, Any]) -> str:
 def _resolve_owner_repo(source_review: dict[str, Any], action: dict[str, Any]) -> str:
     text = _combined_action_text(action)
 
-    if any(
-        marker in text
-        for marker in (
-            "monthly report",
-            "cash-flow",
-            "cash flow",
-            "withdrawal",
-            "deposit",
-            "realized pnl",
-            "unrealized pnl",
-            "open positions",
-            "no-trade",
-            "gating",
-            "free usdt",
-            "dca",
-            "rotation",
-        )
-    ):
-        return "BinancePlatform"
-
     if any(
         marker in text
         for marker in (
@@ -127,15 +107,24 @@ def normalize_action(source_review: dict[str, Any], action: dict[str, Any]) -> d
     }
 
 
-def build_plan(upstream_review: dict[str, Any], downstream_review: dict[str, Any]) -> dict[str, Any]:
-    source_reviews = [upstream_review, downstream_review]
+def build_plan(*source_reviews: dict[str, Any]) -> dict[str, Any]:
+    if not source_reviews:
+        raise ValueError("at least one source review is required")
+    source_review_list = list(source_reviews)
+
     normalized_actions = [
         normalize_action(review, action)
-        for review in source_reviews
+        for review in source_review_list
         for action in review.get("recommended_actions", [])
     ]
+    in_scope_actions = [
+        action for action in normalized_actions if action["owner_repo"] in REPO_ORDER
+    ]
+    out_of_scope_actions = [
+        action for action in normalized_actions if action["owner_repo"] not in REPO_ORDER
+    ]
     repo_groups: dict[str, list[dict[str, Any]]] = defaultdict(list)
-    for action in normalized_actions:
+    for action in in_scope_actions:
         repo_groups[action["owner_repo"]].append(action)
 
     repo_action_summary = {
@@ -148,30 +137,31 @@ def build_plan(upstream_review: dict[str, Any], downstream_review: dict[str, Any
         if repo_groups.get(repo)
     }
 
-    safe_auto_pr_candidates = [action for action in normalized_actions if action["auto_pr_safe"] and action["risk_level"] == "low"]
-    experiment_candidates = [action for action in normalized_actions if action["experiment_only"]]
+    safe_auto_pr_candidates = [action for action in in_scope_actions if action["auto_pr_safe"] and action["risk_level"] == "low"]
+    experiment_candidates = [action for action in in_scope_actions if action["experiment_only"]]
     human_review_required = [
-        action for action in normalized_actions if (not action["auto_pr_safe"]) or action["risk_level"] != "low"
+        action for action in in_scope_actions if (not action["auto_pr_safe"]) or action["risk_level"] != "low"
     ]
     operator_focus = [
         f"{review['source_repo']}: {review['summary']}"
-        for review in source_reviews
+        for review in source_review_list
     ]
 
     highest_review_risk = highest_risk([
-        {"risk_level": upstream_review["risk_level"]},
-        {"risk_level": downstream_review["risk_level"]},
+        {"risk_level": review["risk_level"]}
+        for review in source_review_list
     ])
 
     return {
         "schema_version": SCHEMA_VERSION,
         "generated_at": datetime.now(UTC).isoformat().replace("+00:00", "Z"),
-        "source_reviews": source_reviews,
+        "source_reviews": source_review_list,
         "highest_review_risk": highest_review_risk,
         "repo_action_summary": repo_action_summary,
         "safe_auto_pr_candidates": sort_actions(safe_auto_pr_candidates),
         "experiment_candidates": sort_actions(experiment_candidates),
         "human_review_required": sort_actions(human_review_required),
+        "out_of_scope_actions": sort_actions(out_of_scope_actions),
         "operator_focus": operator_focus,
     }
 
@@ -184,6 +174,7 @@ def render_summary_markdown(plan: dict[str, Any]) -> str:
         f"- Safe auto-PR candidates: `{len(plan['safe_auto_pr_candidates'])}`",
         f"- Experiment candidates: `{len(plan['experiment_candidates'])}`",
         f"- Human review required: `{len(plan['human_review_required'])}`",
+        f"- Out-of-scope downstream actions: `{len(plan.get('out_of_scope_actions', []))}`",
         "",
         "## Source Reviews",
     ]
@@ -218,24 +209,33 @@ def render_summary_markdown(plan: dict[str, Any]) -> str:
         lines.extend(["", "## Operator Focus"])
         lines.extend(f"- {item}" for item in plan["operator_focus"])
 
+    if plan.get("out_of_scope_actions"):
+        lines.extend(["", "## Out-of-Scope Actions"])
+        for action in plan["out_of_scope_actions"]:
+            lines.append(
+                f"- `{action['risk_level']}` {action['owner_repo']}: {action['title']} "
+                f"(from {action['source_repo']} #{action['source_issue_number']})"
+            )
+
     return "\n".join(lines).strip() + "\n"
 
 
 def parse_args() -> argparse.Namespace:
     parser = argparse.ArgumentParser(
-        description="Build the monthly optimization plan by combining upstream and downstream AI review payloads.",
+        description="Build the monthly optimization plan from upstream selector AI review payloads.",
     )
     parser.add_argument("--upstream-review-file", required=True, type=Path)
-    parser.add_argument("--downstream-review-file", required=True, type=Path)
+    parser.add_argument("--downstream-review-file", type=Path)
     parser.add_argument("--output-dir", required=True, type=Path)
     return parser.parse_args()
 
 
 def main() -> int:
     args = parse_args()
-    upstream_review = json.loads(args.upstream_review_file.read_text(encoding="utf-8"))
-    downstream_review = json.loads(args.downstream_review_file.read_text(encoding="utf-8"))
-    plan = build_plan(upstream_review, downstream_review)
+    source_reviews = [json.loads(args.upstream_review_file.read_text(encoding="utf-8"))]
+    if args.downstream_review_file:
+        source_reviews.append(json.loads(args.downstream_review_file.read_text(encoding="utf-8")))
+    plan = build_plan(*source_reviews)
     args.output_dir.mkdir(parents=True, exist_ok=True)
     (args.output_dir / "optimization_plan.json").write_text(
         json.dumps(plan, ensure_ascii=False, indent=2) + "\n",

@@ -48,26 +48,6 @@
         "walk-forward",
         "walk forward",
     ),
-    "BinancePlatform": (
-        "dca",
-        "rotation",
-        "eligibility gate",
-        "free usdt",
-        "cash flow",
-        "withdrawal",
-        "deposit",
-        "open position",
-        "execution",
-        "live trading",
-        "threshold",
-        "circuit breaker",
-        "capital threshold",
-        "allocation",
-        "sizing",
-        "liquidity",
-        "spread",
-        "adv",
-    ),
     "CryptoStrategies": (
         "strategy",
         "signal",
@@ -85,22 +65,6 @@
         r"^src/",
         r"^config/",
     ),
-    "BinancePlatform": (
-        r"^application/",
-        r"^infra/",
-        r"^strategy/",
-        r"^entrypoints/",
-        r"^main\.py$",
-        r"^runtime_support\.py$",
-        r"^live_services\.py$",
-        r"^degraded_mode_support\.py$",
-        r"^market_snapshot_support\.py$",
-        r"^trade_state_support\.py$",
-        r"^trend_pool_support\.py$",
-        r"^strategy_core\.py$",
-        r"^strategy_loader\.py$",
-        r"^strategy_registry\.py$",
-    ),
     "CryptoStrategies": (
         r"^src/",
     ),
@@ -182,17 +146,6 @@ def _is_completed_low_risk_task(action: dict[str, Any], repo_root: Path) -> bool
                 and "deterministic tie-break" in runbook
             )
 
-    if repo_name == "BinancePlatform" and (
-        "zero-trade diagnostics" in title
-        or "diagnostic reporting for no-trade months" in title
-        or "no-trade months" in title
-    ):
-        monthly_report = _read_text(repo_root / "scripts" / "run_monthly_report_bundle.py")
-        return (
-            "## Zero-Trade Diagnostics" in monthly_report
-            and "by_category_and_gate" in monthly_report
-        )
-
     return False