feat: Phase W W5 Phase W benchmark gate schema (WBG-) -- top-level gate asserting NCH+CMC+SCH+BCR all present; VALID_WBG_VERDICTS (3: hardened/partially_hardened/not_hardened), HARDENED_REQUIRED_PRESENT=4; closes Phase W; 61 tests (#1007)

Co-authored-by: OpenCode <opencode@example.com>

Author: cschanhniem

docs/research/NEXT_100_PR_MAP.md

@@ -280,4 +280,4 @@ Make it machine-verifiable that the pipeline produces novel candidates that beat
 | W2 | Add charge-matched challenge schema (CMC-) (complete). — src/openamp_foundry/evidence/charge_matched_challenge.py: VALID_CMC_VERDICTS (4: gap_meaningful/gap_marginal/gap_absent/challenge_not_run), VALID_CHARGE_BASELINE_METHODS (4), MEANINGFUL_GAP_THRESHOLD=0.05, MARGINAL_GAP_LOWER=0.02; auroc_gap auto-computed; verdict auto-derived; dry_lab_only=True enforced; 60 tests in tests/evidence/test_charge_matched_challenge.py. | Formally documents the charge-matched challenge: compares pipeline AUROC vs a charge-only baseline on the same candidate set; verdict controlled vocabulary (gap_meaningful/gap_marginal/gap_absent/not_run); blocks performance claims when the charge-only baseline explains the gap. | C |
 | W3 | Add similarity challenge harness schema (SCH-) (complete). — src/openamp_foundry/evidence/similarity_challenge_harness.py: VALID_SCH_VERDICTS (4: selection_adds_value/marginal_improvement/proximity_driven/challenge_not_run), VALID_SIMILARITY_METRICS (4), SELECTION_VALUE_GAP_THRESHOLD=0.10, MARGINAL_IMPROVEMENT_LOWER=0.03; SimilarityGroupStats helper; similarity_gap auto-computed; dry_lab_only=True enforced; 60 tests in tests/evidence/test_similarity_challenge_harness.py. | Documents whether pipeline-selected candidates are systematically more similar to known AMPs than random selection from the sequence space; flags selection bias from similarity clustering; prevents "novel panel" claim when selection is proximity-driven. | C |
 | W4 | Add benchmark challenge registry schema (BCR-) (complete). — src/openamp_foundry/evidence/benchmark_challenge_registry.py: REQUIRED_CHALLENGE_TYPES (3: NCH/CMC/SCH), VALID_CHALLENGE_VERDICTS (4: pass/marginal/fail/not_run), VALID_BCR_HARDNESS_GRADES (A-D); ChallengeEntry helper; verdict mapping (novel_batch→pass, mixed_novelty→marginal, gap_meaningful→pass, etc.); grade A=all pass, B=all pass+marginal, C=some pass/marginal, D=all fail/not_run; dry_lab_only=True; 64 tests in tests/evidence/test_benchmark_challenge_registry.py. | Machine-readable registry of which benchmark challenges (NCH/CMC/SCH) have been run and passed for a given pipeline version; aggregates challenge verdicts; overall hardness grade (A: all passed, B: most passed, C: some passed, D: none passed). | C |
-| W5 | Add Phase W benchmark gate (WBG-). | Top-level gate asserting NCH + CMC + SCH + BCR all present; overall verdict: hardened/partially_hardened/not_hardened; closes Phase W; no batch-level performance claim is credible without passing this gate. | C |
+| W5 | Add Phase W benchmark gate (WBG-) (complete). — src/openamp_foundry/evidence/phase_w_benchmark_gate.py: REQUIRED_W_COMPONENTS (4: NCH/CMC/SCH/BCR), VALID_WBG_VERDICTS (3: hardened/partially_hardened/not_hardened), HARDENED_REQUIRED_PRESENT=4, PARTIALLY_HARDENED_MIN_PRESENT=2; WComponentCheck helper; prefix validation for all artifact IDs; verdict: hardened=all 4 present, partially_hardened=2-3, not_hardened=0-1; dry_lab_only=True; 61 tests. Closes Phase W. | Top-level gate asserting NCH + CMC + SCH + BCR all present; overall verdict: hardened/partially_hardened/not_hardened; closes Phase W; no batch-level performance claim is credible without passing this gate. | C |

src/openamp_foundry/evidence/phase_w_benchmark_gate.py

@@ -0,0 +1,170 @@
+"""WBG- Phase W benchmark gate schema.
+
+Top-level gate asserting all four Phase W benchmark challenge artifacts
+(NCH, CMC, SCH, BCR) are present for a batch. Produces an overall
+verdict: hardened/partially_hardened/not_hardened. No batch-level
+performance claim is credible without a passing WBG-.
+"""
+
+from __future__ import annotations
+
+from dataclasses import dataclass
+
+REQUIRED_W_COMPONENTS: tuple[str, ...] = ("NCH", "CMC", "SCH", "BCR")
+
+VALID_WBG_VERDICTS: frozenset[str] = frozenset({
+    "hardened",
+    "partially_hardened",
+    "not_hardened",
+})
+
+_COMPONENT_ID_PREFIXES: dict[str, str] = {
+    "NCH": "NCH-",
+    "CMC": "CMC-",
+    "SCH": "SCH-",
+    "BCR": "BCR-",
+}
+
+HARDENED_REQUIRED_PRESENT: int = 4
+PARTIALLY_HARDENED_MIN_PRESENT: int = 2
+
+
+@dataclass
+class WComponentCheck:
+    component_type: str
+    artifact_id: str
+    present: bool
+
+
+@dataclass
+class PhaseWBenchmarkGate:
+    wbg_id: str
+    batch_id: str
+    pipeline_version: str
+    component_checks: list[WComponentCheck]
+    n_components_required: int
+    n_components_present: int
+    missing_component_types: list[str]
+    gate_verdict: str
+    dry_lab_only: bool
+    limitations: list[str]
+    created_at: str
+
+
+def validate_phase_w_benchmark_gate(gate: PhaseWBenchmarkGate) -> None:
+    if not gate.wbg_id.startswith("WBG-"):
+        raise ValueError(f"wbg_id must start with 'WBG-': {gate.wbg_id!r}")
+    if not gate.batch_id:
+        raise ValueError("batch_id must be non-empty")
+    if not gate.pipeline_version:
+        raise ValueError("pipeline_version must be non-empty")
+    seen: set[str] = set()
+    for check in gate.component_checks:
+        if check.component_type not in REQUIRED_W_COMPONENTS:
+            raise ValueError(
+                f"component_type {check.component_type!r} not in REQUIRED_W_COMPONENTS"
+            )
+        if check.component_type in seen:
+            raise ValueError(f"duplicate component_type: {check.component_type!r}")
+        seen.add(check.component_type)
+        expected_prefix = _COMPONENT_ID_PREFIXES[check.component_type]
+        if check.present and not check.artifact_id.startswith(expected_prefix):
+            raise ValueError(
+                f"artifact_id for {check.component_type} must start with "
+                f"'{expected_prefix}': {check.artifact_id!r}"
+            )
+    if gate.n_components_required != len(REQUIRED_W_COMPONENTS):
+        raise ValueError(
+            f"n_components_required must be {len(REQUIRED_W_COMPONENTS)}"
+        )
+    present_count = sum(1 for c in gate.component_checks if c.present)
+    if gate.n_components_present != present_count:
+        raise ValueError("n_components_present mismatch")
+    expected_missing = sorted(
+        c.component_type for c in gate.component_checks if not c.present
+    )
+    if gate.missing_component_types != expected_missing:
+        raise ValueError("missing_component_types mismatch")
+    if gate.gate_verdict not in VALID_WBG_VERDICTS:
+        raise ValueError(
+            f"gate_verdict {gate.gate_verdict!r} not in VALID_WBG_VERDICTS"
+        )
+    if not gate.dry_lab_only:
+        raise ValueError("dry_lab_only must be True")
+    if not gate.limitations:
+        raise ValueError("limitations must be non-empty")
+    if not gate.created_at:
+        raise ValueError("created_at must be non-empty")
+
+
+def _compute_verdict(n_present: int) -> str:
+    if n_present == HARDENED_REQUIRED_PRESENT:
+        return "hardened"
+    if n_present >= PARTIALLY_HARDENED_MIN_PRESENT:
+        return "partially_hardened"
+    return "not_hardened"
+
+
+def build_phase_w_benchmark_gate(
+    *,
+    wbg_id: str,
+    batch_id: str,
+    pipeline_version: str,
+    nch_artifact_id: str = "",
+    cmc_artifact_id: str = "",
+    sch_artifact_id: str = "",
+    bcr_artifact_id: str = "",
+    limitations: list[str],
+    created_at: str,
+) -> PhaseWBenchmarkGate:
+    artifact_map = {
+        "NCH": nch_artifact_id,
+        "CMC": cmc_artifact_id,
+        "SCH": sch_artifact_id,
+        "BCR": bcr_artifact_id,
+    }
+    checks = [
+        WComponentCheck(
+            component_type=ctype,
+            artifact_id=artifact_map[ctype],
+            present=bool(artifact_map[ctype]),
+        )
+        for ctype in REQUIRED_W_COMPONENTS
+    ]
+    n_present = sum(1 for c in checks if c.present)
+    missing = sorted(c.component_type for c in checks if not c.present)
+    verdict = _compute_verdict(n_present)
+    gate = PhaseWBenchmarkGate(
+        wbg_id=wbg_id,
+        batch_id=batch_id,
+        pipeline_version=pipeline_version,
+        component_checks=checks,
+        n_components_required=len(REQUIRED_W_COMPONENTS),
+        n_components_present=n_present,
+        missing_component_types=missing,
+        gate_verdict=verdict,
+        dry_lab_only=True,
+        limitations=limitations,
+        created_at=created_at,
+    )
+    validate_phase_w_benchmark_gate(gate)
+    return gate
+
+
+def format_phase_w_benchmark_gate(gate: PhaseWBenchmarkGate) -> str:
+    lines = [
+        f"Phase W Benchmark Gate — {gate.wbg_id}",
+        f"Batch: {gate.batch_id}  |  Pipeline: {gate.pipeline_version}",
+        f"Verdict: {gate.gate_verdict}  |  Present: {gate.n_components_present}/{gate.n_components_required}",
+        "Components:",
+    ]
+    for check in gate.component_checks:
+        status = "PRESENT" if check.present else "MISSING"
+        aid = check.artifact_id if check.artifact_id else "(none)"
+        lines.append(f"  {check.component_type}: {status}  {aid}")
+    if gate.missing_component_types:
+        lines.append(f"Missing: {', '.join(gate.missing_component_types)}")
+    lines.append(f"Created: {gate.created_at}")
+    lines.append(f"Limitations: {'; '.join(gate.limitations)}")
+    lines.append(f"dry_lab_only: {gate.dry_lab_only}")
+    return "\n".join(lines)