its-magic 0.1.2-9 → 0.1.2

package/scripts/intake_evidence_lib.py

@@ -0,0 +1,802 @@
+"""
+Deterministic intake evidence validation
+(US-0078 / US-0083 / DEC-0060 / DEC-0067 / R-0055 / BUG-0007 / R-0066).
+
+Consumes a logical intake_evidence bundle (dict). PO workflows MUST run this
+gate before mutating backlog/acceptance; failures are fail-closed.
+"""
+
+from __future__ import annotations
+
+import hashlib
+import json
+import re
+from dataclasses import dataclass, field
+from typing import Any
+
+IE_REF_RE = re.compile(r"^ie:([^:]+):(\d+):([0-9a-f]{16})$")
+PLAN_AREA_ID_RE = re.compile(r"^[a-z0-9][a-z0-9_-]{1,63}$")
+
+PACK_REQUIRED_KEYS: dict[str, tuple[str, ...]] = {
+    "first-intake-pack": (
+        "users_problem",
+        "runtime_target_environment",
+        "language_framework_runtime",
+        "architecture_preference",
+        "ui_design_expectations",
+        "security_compliance",
+        "non_functional_priorities",
+        "scope_timeline",
+    ),
+    "small-intake-pack": (
+        "outcome_success_criteria",
+        "impacted_components",
+        "constraints_compatibility_risks",
+        "required_tests_acceptance_checks",
+        "done_definition",
+    ),
+}
+
+ASSUMPTIONS_TOPIC_KEY = "assumptions_bundle"
+
+SAFE_ASSUMPTION_LITERALS = frozenset(
+    {
+        "",
+        "(none)",
+        "none",
+        "no",
+        "false",
+        "n/a",
+        "na",
+        "n",
+    }
+)
+
+FALSE_CONFIRMATION_LITERALS = frozenset({"yes", "true", "confirmed"})
+ALLOWED_SATISFIED_BY = frozenset({"answer_ref", "assumption_confirmation_ref", "delegation_ref"})
+DELEGATION_CONFIDENCE_VALUES = frozenset({"low", "medium", "high"})
+
+
+def canonical_json_sha256_16(obj: dict[str, Any]) -> str:
+    """Sorted-key compact JSON UTF-8 → first 16 hex chars of SHA-256 (DEC-0060)."""
+    blob = json.dumps(obj, sort_keys=True, separators=(",", ":"), ensure_ascii=False).encode("utf-8")
+    return hashlib.sha256(blob).hexdigest()[:16]
+
+
+def build_ie_ref(
+    intake_run_id: str,
+    turn_index: int,
+    topic_key: str,
+    satisfied_by: str,
+    quoted_user_text: str,
+) -> str:
+    """Construct ie: ref per DEC-0060 §4."""
+    payload = {
+        "intake_run_id": intake_run_id,
+        "turn_index": int(turn_index),
+        "topic_key": topic_key,
+        "satisfied_by": satisfied_by,
+        "quoted_user_text": (quoted_user_text or "").strip(),
+    }
+    digest = canonical_json_sha256_16(payload)
+    return f"ie:{intake_run_id}:{int(turn_index)}:{digest}"
+
+
+def parse_ie_ref(ref: str) -> tuple[str, int, str] | None:
+    if not ref or not isinstance(ref, str):
+        return None
+    m = IE_REF_RE.match(ref.strip())
+    if not m:
+        return None
+    return m.group(1), int(m.group(2)), m.group(3)
+
+
+def verify_ie_ref(
+    ref: str,
+    *,
+    intake_run_id: str,
+    turn_index: int,
+    topic_key: str,
+    satisfied_by: str,
+    quoted_user_text: str,
+) -> bool:
+    parsed = parse_ie_ref(ref)
+    if not parsed:
+        return False
+    prun, pturn, phash = parsed
+    if prun != intake_run_id or pturn != int(turn_index):
+        return False
+    payload = {
+        "intake_run_id": intake_run_id,
+        "turn_index": int(turn_index),
+        "topic_key": topic_key,
+        "satisfied_by": satisfied_by,
+        "quoted_user_text": (quoted_user_text or "").strip(),
+    }
+    return canonical_json_sha256_16(payload) == phash
+
+
+def _norm_assumption(s: str) -> str:
+    return (s or "").strip()
+
+
+def assumption_literal_requires_confirmation_ref(value: str) -> bool:
+    """R-0055 rules 4–5 + DEC-0060: non-placeholder confirmations need evidence."""
+    v = _norm_assumption(value)
+    if not v:
+        return False
+    low = v.lower()
+    if low in SAFE_ASSUMPTION_LITERALS:
+        return False
+    if low in FALSE_CONFIRMATION_LITERALS:
+        return True
+    # Any other non-empty narrative still implies confirmation was recorded → needs ref
+    return True
+
+
+@dataclass
+class ValidationResult:
+    ok: bool
+    primary_codes: list[str] = field(default_factory=list)
+    missing_topics: list[str] = field(default_factory=list)
+    diagnostics: list[str] = field(default_factory=list)
+
+    def add_code(self, code: str) -> None:
+        if code not in self.primary_codes:
+            self.primary_codes.append(code)
+
+
+def _topic_rows(bundle: dict[str, Any]) -> list[dict[str, Any]]:
+    raw = bundle.get("topic_coverage")
+    if raw is None:
+        return []
+    if not isinstance(raw, list):
+        return []
+    out: list[dict[str, Any]] = []
+    for row in raw:
+        if isinstance(row, dict):
+            out.append(row)
+    return out
+
+
+def _asked_set(bundle: dict[str, Any]) -> set[str]:
+    at = bundle.get("asked_topics")
+    if at is None:
+        return set()
+    if isinstance(at, str):
+        parts = [p.strip() for p in at.replace("\n", ",").split(",")]
+        return {p for p in parts if p}
+    if isinstance(at, list):
+        return {str(x).strip() for x in at if str(x).strip()}
+    return set()
+
+
+def _row_run_id(bundle: dict[str, Any], row: dict[str, Any]) -> str | None:
+    v = row.get("intake_run_id")
+    if v is not None and str(v).strip():
+        return str(v).strip()
+    v2 = bundle.get("intake_run_id")
+    if v2 is not None and str(v2).strip():
+        return str(v2).strip()
+    return None
+
+
+def _row_turn(row: dict[str, Any]) -> int | None:
+    t = row.get("turn_index")
+    if t is None:
+        return None
+    try:
+        return int(t)
+    except (TypeError, ValueError):
+        return None
+
+
+def _row_uses_equivalent_evidence(row: dict[str, Any]) -> bool:
+    """
+    US-0083 AC-1: allow required-topic accounting without forcing repetitive asks
+    when equivalent evidence is already captured and explicitly referenced.
+    """
+    marker = str(row.get("evidence_source") or "").strip()
+    if marker != "equivalent_evidence_ref":
+        return False
+    return bool(str(row.get("equivalent_evidence_ref") or "").strip())
+
+
+def _norm_answer_ref_quoted_user_text(value: Any) -> str:
+    """Normalize quoted_user_text for BUG-0007 duplicate detection (DEC-0060 strip parity)."""
+    if value is None:
+        return ""
+    return str(value).strip()
+
+
+def _row_exempt_from_answer_ref_topic_distinctness(row: dict[str, Any]) -> bool:
+    """BUG-0007: alternate satisfaction paths do not participate in answer_ref blob reuse checks."""
+    return _row_uses_equivalent_evidence(row)
+
+
+def _validate_answer_ref_topic_distinctness(
+    bundle: dict[str, Any],
+    required: list[str],
+    by_key: dict[str, dict[str, Any]],
+    res: ValidationResult,
+) -> None:
+    """
+    BUG-0007 / R-0066: distinct required topic_key rows must not reuse the same
+    quoted_user_text under satisfied_by=answer_ref (normalized), except exempt rows.
+    """
+    norm_to_topics: dict[str, list[str]] = {}
+    for k in required:
+        row = by_key.get(k)
+        if not row:
+            continue
+        sat = (row.get("satisfied_by") or "").strip()
+        if sat != "answer_ref":
+            continue
+        if _row_exempt_from_answer_ref_topic_distinctness(row):
+            continue
+        irid = _row_run_id(bundle, row)
+        tit = _row_turn(row)
+        if irid is None or tit is None:
+            continue
+        qraw = row.get("quoted_user_text")
+        qtxt = "" if qraw is None else str(qraw)
+        ref = (row.get("ref") or "").strip()
+        if not ref or not verify_ie_ref(
+            ref,
+            intake_run_id=irid,
+            turn_index=int(tit),
+            topic_key=k,
+            satisfied_by=sat,
+            quoted_user_text=qtxt,
+        ):
+            continue
+        norm = _norm_answer_ref_quoted_user_text(qraw)
+        norm_to_topics.setdefault(norm, []).append(k)
+
+    dup_groups: list[str] = []
+    for norm, topics in norm_to_topics.items():
+        uniq = sorted(set(topics))
+        if len(uniq) < 2:
+            continue
+        dup_groups.append("text=" + repr(norm[:120]) + " topics=" + ",".join(uniq))
+
+    if dup_groups:
+        res.ok = False
+        res.add_code("INTAKE_ANSWER_REF_NOT_TOPIC_DISTINCT")
+        res.diagnostics.append(
+            "Remediation: distinct required topics must not reuse the same quoted_user_text "
+            "under satisfied_by=answer_ref (BUG-0007 / R-0066). Duplicates: "
+            + "; ".join(dup_groups)
+            + ". Use per-topic answers, or an allowed alternate path "
+            "(evidence_source=equivalent_evidence_ref + equivalent_evidence_ref, "
+            "delegation_ref per DEC-0067 / US-0083, or assumption_confirmation_ref on the row)."
+        )
+
+
+def _candidate_story_ids(bundle: dict[str, Any]) -> set[str]:
+    out: set[str] = set()
+
+    raw_ids = bundle.get("candidate_story_ids")
+    if isinstance(raw_ids, list):
+        for sid in raw_ids:
+            sv = str(sid).strip()
+            if sv:
+                out.add(sv)
+
+    raw_story_ids = bundle.get("story_ids")
+    if isinstance(raw_story_ids, list):
+        for sid in raw_story_ids:
+            sv = str(sid).strip()
+            if sv:
+                out.add(sv)
+
+    raw_story_map = bundle.get("story_map")
+    if isinstance(raw_story_map, list):
+        for row in raw_story_map:
+            if not isinstance(row, dict):
+                continue
+            sv = str(row.get("story_id") or "").strip()
+            if sv:
+                out.add(sv)
+
+    return out
+
+
+def _validate_plan_coverage_contract(bundle: dict[str, Any], res: ValidationResult) -> None:
+    inventory_raw = bundle.get("plan_area_inventory")
+    coverage_raw = bundle.get("plan_area_coverage")
+    coverage_complete_raw = bundle.get("coverage_complete")
+    candidate_story_ids = _candidate_story_ids(bundle)
+
+    inventory_rows = inventory_raw if isinstance(inventory_raw, list) else []
+    coverage_rows = coverage_raw if isinstance(coverage_raw, list) else []
+
+    inventory_ids: list[str] = []
+    coverage_by_id: dict[str, dict[str, Any]] = {}
+
+    coverage_missing = False
+    id_invalid = False
+    contract_invalid = False
+    deferred_ref_missing = False
+
+    if not inventory_rows:
+        coverage_missing = True
+        res.diagnostics.append(
+            "Remediation: plan_area_inventory must be a non-empty list for first/new/broad intake."
+        )
+    if not coverage_rows:
+        coverage_missing = True
+        res.diagnostics.append(
+            "Remediation: plan_area_coverage must be a non-empty list for first/new/broad intake."
+        )
+
+    seen_inventory_ids: set[str] = set()
+    for row in inventory_rows:
+        if not isinstance(row, dict):
+            id_invalid = True
+            continue
+        plan_area_id = str(row.get("plan_area_id") or "").strip()
+        if not PLAN_AREA_ID_RE.match(plan_area_id):
+            id_invalid = True
+            continue
+        if plan_area_id in seen_inventory_ids:
+            id_invalid = True
+            continue
+        seen_inventory_ids.add(plan_area_id)
+        inventory_ids.append(plan_area_id)
+
+    for row in coverage_rows:
+        if not isinstance(row, dict):
+            id_invalid = True
+            continue
+        plan_area_id = str(row.get("plan_area_id") or "").strip()
+        if not PLAN_AREA_ID_RE.match(plan_area_id):
+            id_invalid = True
+            continue
+        if plan_area_id in coverage_by_id:
+            id_invalid = True
+            continue
+        coverage_by_id[plan_area_id] = row
+
+    inventory_id_set = set(inventory_ids)
+    coverage_id_set = set(coverage_by_id.keys())
+
+    if inventory_id_set != coverage_id_set:
+        coverage_missing = True
+        missing_ids = sorted(inventory_id_set - coverage_id_set)
+        extra_ids = sorted(coverage_id_set - inventory_id_set)
+        if missing_ids:
+            res.diagnostics.append(
+                "Remediation: add plan_area_coverage rows for uncovered plan_area_id values: "
+                + ", ".join(missing_ids)
+            )
+        if extra_ids:
+            contract_invalid = True
+            res.diagnostics.append(
+                "Remediation: remove unknown plan_area_coverage plan_area_id values not in plan_area_inventory: "
+                + ", ".join(extra_ids)
+            )
+
+    for plan_area_id in sorted(inventory_id_set & coverage_id_set):
+        row = coverage_by_id[plan_area_id]
+        story_ids_raw = row.get("story_ids")
+        deferred_ref = str(row.get("deferred_ref") or "").strip()
+
+        story_ids: list[str] = []
+        if isinstance(story_ids_raw, list):
+            for sid in story_ids_raw:
+                sv = str(sid).strip()
+                if sv:
+                    story_ids.append(sv)
+        has_story_ids = bool(story_ids)
+        has_deferred_ref = bool(deferred_ref)
+
+        if has_story_ids == has_deferred_ref:
+            contract_invalid = True
+            res.diagnostics.append(
+                "Remediation: each plan_area_coverage row must set exactly one mapping path "
+                "(story_ids xor deferred_ref) for plan_area_id "
+                + repr(plan_area_id)
+                + "."
+            )
+            continue
+
+        if has_story_ids:
+            if candidate_story_ids:
+                unknown_story_ids = sorted({sid for sid in story_ids if sid not in candidate_story_ids})
+                if unknown_story_ids:
+                    contract_invalid = True
+                    res.diagnostics.append(
+                        "Remediation: plan_area_id "
+                        + repr(plan_area_id)
+                        + " references unknown story_ids not present in candidate story set: "
+                        + ", ".join(unknown_story_ids)
+                    )
+        else:
+            deferred_reason = str(row.get("deferred_reason") or "").strip()
+            if not deferred_reason:
+                deferred_ref_missing = True
+                res.diagnostics.append(
+                    "Remediation: deferred mapping for plan_area_id "
+                    + repr(plan_area_id)
+                    + " requires both deferred_ref and deferred_reason."
+                )
+
+    derived_coverage_complete = not (coverage_missing or id_invalid or contract_invalid or deferred_ref_missing)
+    if coverage_complete_raw is not True:
+        contract_invalid = True
+        res.diagnostics.append(
+            "Remediation: set coverage_complete=true only after plan_area_inventory and "
+            "plan_area_coverage pass deterministic validation."
+        )
+    if bool(coverage_complete_raw) != derived_coverage_complete:
+        contract_invalid = True
+        res.diagnostics.append(
+            "Remediation: coverage_complete must match derived contract result from "
+            "plan_area_inventory/plan_area_coverage validation."
+        )
+
+    if coverage_missing:
+        res.ok = False
+        res.add_code("INTAKE_PLAN_COVERAGE_MISSING")
+    if id_invalid:
+        res.ok = False
+        res.add_code("INTAKE_PLAN_AREA_ID_INVALID")
+        res.diagnostics.append(
+            "Remediation: plan_area_id must be unique and match ^[a-z0-9][a-z0-9_-]{1,63}$ in inventory and coverage."
+        )
+    if deferred_ref_missing:
+        res.ok = False
+        res.add_code("INTAKE_PLAN_DEFERRED_REF_MISSING")
+    if contract_invalid:
+        res.ok = False
+        res.add_code("INTAKE_PLAN_COVERAGE_CONTRACT_INVALID")
+
+
+def validate_intake_evidence(
+    bundle: dict[str, Any],
+    *,
+    intake_guided_mode: int | None = None,
+) -> ValidationResult:
+    """
+    Validate logical intake_evidence. intake_guided_mode is accepted for API parity
+    (AC-5/AC-6): validation rules do not branch on mode — same pipeline for {0,1}.
+    """
+    _ = intake_guided_mode  # explicit no-op for parity contract
+    res = ValidationResult(ok=True)
+
+    pack = (bundle.get("selected_pack") or "").strip()
+    if pack not in PACK_REQUIRED_KEYS:
+        res.ok = False
+        res.add_code("INTAKE_REQUIRED_PACK_INCOMPLETE")
+        res.diagnostics.append(
+            f"Remediation: set selected_pack to one of {sorted(PACK_REQUIRED_KEYS.keys())!r}; "
+            f"unknown or empty pack is not allowed before persistence."
+        )
+        res.add_code("INTAKE_PERSISTENCE_BLOCKED")
+        return res
+
+    required = list(PACK_REQUIRED_KEYS[pack])
+    rows = _topic_rows(bundle)
+    by_key: dict[str, dict[str, Any]] = {}
+    for row in rows:
+        k = (row.get("topic_key") or "").strip()
+        if k:
+            by_key[k] = row
+
+    missing_cov: list[str] = []
+    for k in required:
+        row = by_key.get(k)
+        if not row:
+            missing_cov.append(k)
+            continue
+
+        ref = (row.get("ref") or "").strip()
+        sat = (row.get("satisfied_by") or "").strip()
+        qtxt = row.get("quoted_user_text")
+        if qtxt is None:
+            qtxt = ""
+        qtxt = str(qtxt)
+
+        if sat not in ALLOWED_SATISFIED_BY:
+            res.ok = False
+            res.diagnostics.append(
+                f"Remediation: topic {k!r} must set satisfied_by to "
+                f"'answer_ref', 'assumption_confirmation_ref', or 'delegation_ref' (got {sat!r})."
+            )
+            missing_cov.append(k)
+            continue
+
+        if sat == "delegation_ref":
+            scope = str(row.get("delegation_scope") or "").strip()
+            rationale = str(row.get("delegation_rationale") or "").strip()
+            confidence = str(row.get("delegation_confidence") or "").strip().lower()
+            if not scope or not rationale or not confidence:
+                res.ok = False
+                res.add_code("INTAKE_DELEGATION_EVIDENCE_MISSING")
+                missing_fields = []
+                if not scope:
+                    missing_fields.append("delegation_scope")
+                if not rationale:
+                    missing_fields.append("delegation_rationale")
+                if not confidence:
+                    missing_fields.append("delegation_confidence")
+                res.diagnostics.append(
+                    f"Remediation: delegated topic {k!r} requires non-empty fields: "
+                    + ", ".join(missing_fields)
+                    + "."
+                )
+                continue
+            if confidence not in DELEGATION_CONFIDENCE_VALUES:
+                res.ok = False
+                res.add_code("INTAKE_DELEGATION_EVIDENCE_INVALID")
+                res.diagnostics.append(
+                    f"Remediation: delegated topic {k!r} delegation_confidence must be one of "
+                    f"{sorted(DELEGATION_CONFIDENCE_VALUES)!r} (got {confidence!r})."
+                )
+                continue
+
+        irid = _row_run_id(bundle, row)
+        tit = _row_turn(row)
+        if irid is None or tit is None:
+            res.ok = False
+            res.diagnostics.append(
+                f"Remediation: topic {k!r} needs intake_run_id and turn_index (row or bundle) "
+                f"for ie: ref verification."
+            )
+            missing_cov.append(k)
+            continue
+
+        if not ref or not verify_ie_ref(
+            ref,
+            intake_run_id=irid,
+            turn_index=tit,
+            topic_key=k,
+            satisfied_by=sat,
+            quoted_user_text=qtxt,
+        ):
+            res.ok = False
+            if sat == "delegation_ref":
+                res.add_code("INTAKE_DELEGATION_EVIDENCE_INVALID")
+                res.diagnostics.append(
+                    f"Remediation: delegated topic {k!r} ref is malformed or hash mismatch — rebuild ie: ref "
+                    f"with DEC-0060 canonical JSON (sorted keys) and quoted_user_text."
+                )
+            else:
+                res.diagnostics.append(
+                    f"Remediation: topic {k!r} ref is malformed or hash mismatch — rebuild ie: ref "
+                    f"with DEC-0060 canonical JSON (sorted keys) and quoted_user_text."
+                )
+                missing_cov.append(k)
+
+    if missing_cov:
+        res.ok = False
+        res.missing_topics = sorted(set(missing_cov))
+        res.add_code("INTAKE_REQUIRED_TOPIC_MISSING")
+        if len(res.missing_topics) > 1:
+            res.add_code("INTAKE_REQUIRED_PACK_INCOMPLETE")
+        res.diagnostics.append(
+            "Remediation: supply complete topic_coverage with valid ie: refs for: "
+            + ", ".join(res.missing_topics)
+        )
+
+    asked = _asked_set(bundle)
+    for k in required:
+        if k not in by_key:
+            continue
+        if k not in asked:
+            row = by_key[k]
+            if _row_uses_equivalent_evidence(row):
+                continue
+            res.ok = False
+            res.add_code("INTAKE_REQUIRED_TOPIC_MISSING")
+            res.diagnostics.append(
+                f"Remediation: add {k!r} to asked_topics or mark evidence_source='equivalent_evidence_ref' "
+                f"with equivalent_evidence_ref when reusing previously captured equivalent evidence."
+            )
+            if k not in res.missing_topics:
+                res.missing_topics.append(k)
+    res.missing_topics = sorted(set(res.missing_topics))
+
+    _validate_answer_ref_topic_distinctness(bundle, required, by_key, res)
+
+    # US-0081 / DEC-0064: first/new/broad intake requires complete-plan coverage contract.
+    if pack == "first-intake-pack":
+        _validate_plan_coverage_contract(bundle, res)
+
+    ac = bundle.get("assumptions_confirmed")
+    ac_str = ac if isinstance(ac, str) else ("(none)" if ac is None else str(ac))
+
+    low = _norm_assumption(ac_str).lower()
+    if low in FALSE_CONFIRMATION_LITERALS and not (bundle.get("assumption_confirmation_ref") or "").strip():
+        res.ok = False
+        res.add_code("INTAKE_ASSUMPTION_CONFIRMATION_REQUIRED")
+        res.diagnostics.append(
+            "Remediation: literal yes/true/confirmed is rejected without "
+            "assumption_confirmation_ref (R-0055 rule 5)."
+        )
+
+    if assumption_literal_requires_confirmation_ref(ac_str):
+        aref = (bundle.get("assumption_confirmation_ref") or "").strip()
+        aquote = str(bundle.get("assumption_confirmation_quoted") or "")
+        irid = bundle.get("assumption_confirmation_intake_run_id")
+        tit = bundle.get("assumption_confirmation_turn_index")
+        if not aref or irid is None or tit is None:
+            res.ok = False
+            res.add_code("INTAKE_ASSUMPTION_CONFIRMATION_REQUIRED")
+            res.diagnostics.append(
+                "Remediation: affirmative assumptions_confirmed requires assumption_confirmation_ref "
+                "plus assumption_confirmation_intake_run_id, assumption_confirmation_turn_index, "
+                "and assumption_confirmation_quoted for ie: binding."
+            )
+        elif not verify_ie_ref(
+            aref,
+            intake_run_id=str(irid),
+            turn_index=int(tit),
+            topic_key=ASSUMPTIONS_TOPIC_KEY,
+            satisfied_by="assumption_confirmation_ref",
+            quoted_user_text=aquote,
+        ):
+            res.ok = False
+            res.add_code("INTAKE_ASSUMPTION_CONFIRMATION_REQUIRED")
+            res.diagnostics.append(
+                "Remediation: assumption_confirmation_ref is invalid or does not match "
+                "quoted affirmative user text / run metadata."
+            )
+
+    if not res.ok and "INTAKE_PERSISTENCE_BLOCKED" not in res.primary_codes:
+        res.add_code("INTAKE_PERSISTENCE_BLOCKED")
+
+    return res
+
+
+def format_blocked_message(result: ValidationResult) -> str:
+    lines = [
+        "INTAKE_PERSISTENCE_BLOCKED",
+        "primary_codes=" + ",".join(result.primary_codes),
+    ]
+    if result.missing_topics:
+        lines.append("missing_topics=" + ",".join(result.missing_topics))
+    lines.extend(result.diagnostics)
+    return "\n".join(lines)
+
+
+def self_test() -> None:
+    """Minimal sanity checks for --self-test / installer wiring."""
+    rid = "selftest-run"
+    ref = build_ie_ref(rid, 0, "outcome_success_criteria", "answer_ref", "ok")
+    assert parse_ie_ref(ref) is not None
+    assert verify_ie_ref(
+        ref,
+        intake_run_id=rid,
+        turn_index=0,
+        topic_key="outcome_success_criteria",
+        satisfied_by="answer_ref",
+        quoted_user_text="ok",
+    )
+    assert not verify_ie_ref(
+        ref,
+        intake_run_id=rid,
+        turn_index=0,
+        topic_key="outcome_success_criteria",
+        satisfied_by="answer_ref",
+        quoted_user_text="tampered",
+    )
+    # Mode argument must not change outcome (AC-6 / AC-5 parity)
+    small = PACK_REQUIRED_KEYS["small-intake-pack"]
+    rows = []
+    for i, key in enumerate(small):
+        rows.append(
+            {
+                "topic_key": key,
+                "satisfied_by": "answer_ref",
+                "quoted_user_text": f"a{i}",
+                "intake_run_id": rid,
+                "turn_index": i,
+                "ref": build_ie_ref(rid, i, key, "answer_ref", f"a{i}"),
+            }
+        )
+    bundle = {
+        "selected_pack": "small-intake-pack",
+        "intake_run_id": rid,
+        "asked_topics": list(small),
+        "missing_topics": [],
+        "assumptions_confirmed": "(none)",
+        "topic_coverage": rows,
+    }
+    r0 = validate_intake_evidence(bundle, intake_guided_mode=0)
+    r1 = validate_intake_evidence(bundle, intake_guided_mode=1)
+    assert r0.ok and r1.ok
+    assert r0.primary_codes == r1.primary_codes
+
+    # US-0083 delegated-topic pass path
+    delegated_key = "done_definition"
+    rows2 = []
+    for i, key in enumerate(small):
+        sat = "delegation_ref" if key == delegated_key else "answer_ref"
+        txt = f"d{i}"
+        row = {
+            "topic_key": key,
+            "satisfied_by": sat,
+            "quoted_user_text": txt,
+            "intake_run_id": rid,
+            "turn_index": 200 + i,
+            "ref": build_ie_ref(rid, 200 + i, key, sat, txt),
+        }
+        if sat == "delegation_ref":
+            row["delegation_scope"] = "Implementation defaults for done criteria wording"
+            row["delegation_rationale"] = "User asked to proceed without additional specificity."
+            row["delegation_confidence"] = "medium"
+        rows2.append(row)
+    delegated_bundle = {
+        "selected_pack": "small-intake-pack",
+        "intake_run_id": rid,
+        "asked_topics": list(small),
+        "missing_topics": [],
+        "assumptions_confirmed": "(none)",
+        "topic_coverage": rows2,
+    }
+    d0 = validate_intake_evidence(delegated_bundle, intake_guided_mode=0)
+    d1 = validate_intake_evidence(delegated_bundle, intake_guided_mode=1)
+    assert d0.ok and d1.ok
+    assert d0.primary_codes == d1.primary_codes
+
+    # BUG-0007: same quoted_user_text across multiple answer_ref required topics fails
+    dup_txt = "synthetic blob echoed for every topic"
+    dup_rows = []
+    for i, key in enumerate(small):
+        dup_rows.append(
+            {
+                "topic_key": key,
+                "satisfied_by": "answer_ref",
+                "quoted_user_text": dup_txt,
+                "intake_run_id": rid,
+                "turn_index": 300 + i,
+                "ref": build_ie_ref(rid, 300 + i, key, "answer_ref", dup_txt),
+            }
+        )
+    dup_bundle = {
+        "selected_pack": "small-intake-pack",
+        "intake_run_id": rid,
+        "asked_topics": list(small),
+        "missing_topics": [],
+        "assumptions_confirmed": "(none)",
+        "topic_coverage": dup_rows,
+    }
+    dup_res = validate_intake_evidence(dup_bundle)
+    assert not dup_res.ok
+    assert "INTAKE_ANSWER_REF_NOT_TOPIC_DISTINCT" in dup_res.primary_codes
+
+    # First-intake full-plan coverage contract (US-0081 / DEC-0064)
+    first = PACK_REQUIRED_KEYS["first-intake-pack"]
+    first_rows = []
+    for i, key in enumerate(first):
+        first_rows.append(
+            {
+                "topic_key": key,
+                "satisfied_by": "answer_ref",
+                "quoted_user_text": f"b{i}",
+                "intake_run_id": rid,
+                "turn_index": i,
+                "ref": build_ie_ref(rid, i, key, "answer_ref", f"b{i}"),
+            }
+        )
+    full_bundle = {
+        "selected_pack": "first-intake-pack",
+        "intake_run_id": rid,
+        "asked_topics": list(first),
+        "missing_topics": [],
+        "assumptions_confirmed": "(none)",
+        "topic_coverage": first_rows,
+        "candidate_story_ids": ["US-9001", "US-9002"],
+        "plan_area_inventory": [
+            {"plan_area_id": "auth", "title": "Auth"},
+            {"plan_area_id": "billing", "title": "Billing"},
+        ],
+        "plan_area_coverage": [
+            {"plan_area_id": "auth", "story_ids": ["US-9001"]},
+            {"plan_area_id": "billing", "story_ids": ["US-9002"]},
+        ],
+        "coverage_complete": True,
+    }
+    assert validate_intake_evidence(full_bundle, intake_guided_mode=0).ok