feat(governance): 新增 release authorization hold

apps/api/src/services/ai_agent_result_capture_release_authorization_hold.py

@@ -0,0 +1,433 @@
+"""
+AI Agent result capture release authorization hold snapshot.
+
+Loads the latest committed P2-134 release authorization hold.
+This module validates committed evidence only; it never authorizes owner
+release, approves maintenance windows, confirms rollback owners, passes
+release authorization, releases live apply, applies writers, writes receipts,
+writes result captures, writes learning records, updates PlayBook trust, writes
+reviewer / Gateway queues, sends Telegram messages, reads secrets, or performs
+destructive operations.
+"""
+
+from __future__ import annotations
+
+import json
+import re
+from pathlib import Path
+from typing import Any
+
+from src.services.snapshot_paths import default_evaluations_dir
+
+_DEFAULT_EVALUATIONS_DIR = default_evaluations_dir(Path(__file__))
+_SNAPSHOT_PATTERN = "ai_agent_result_capture_release_authorization_hold_*.json"
+_SCHEMA_VERSION = "ai_agent_result_capture_release_authorization_hold_v1"
+_RUNTIME_AUTHORITY = "result_capture_release_authorization_hold_only_no_live_write"
+
+
+def load_latest_ai_agent_result_capture_release_authorization_hold(
+    evaluations_dir: Path | None = None,
+) -> dict[str, Any]:
+    """Load the newest committed release authorization hold."""
+    directory = evaluations_dir or _DEFAULT_EVALUATIONS_DIR
+    candidates = sorted(directory.glob(_SNAPSHOT_PATTERN))
+    if not candidates:
+        raise FileNotFoundError(f"no AI Agent result capture release authorization hold snapshots found in {directory}")
+
+    latest = candidates[-1]
+    with latest.open(encoding="utf-8") as handle:
+        payload = json.load(handle)
+
+    if not isinstance(payload, dict):
+        raise ValueError(f"{latest}: expected JSON object")
+
+    label = str(latest)
+    _require_schema(payload, label)
+    _require_prior(payload, label)
+    _require_truth(payload, label)
+    _require_authorization_holds(payload, label)
+    _require_rollback_holds(payload, label)
+    _require_release_window_holds(payload, label)
+    _require_live_apply_holds(payload, label)
+    _require_blocked_transitions(payload, label)
+    _require_actions(payload, label)
+    _require_display_redaction(payload, label)
+    _require_no_forbidden_display_terms(payload, label)
+    _require_rollup_consistency(payload, label)
+    return payload
+
+
+def _require_schema(payload: dict[str, Any], label: str) -> None:
+    if payload.get("schema_version") != _SCHEMA_VERSION:
+        raise ValueError(f"{label}: expected schema_version={_SCHEMA_VERSION}")
+    status = payload.get("program_status") or {}
+    expected = {
+        "current_priority": "P2",
+        "current_task_id": "P2-134",
+        "next_task_id": "P2-135",
+        "read_only_mode": True,
+        "runtime_authority": _RUNTIME_AUTHORITY,
+        "overall_completion_percent": 100,
+    }
+    mismatches = _mismatches(status, expected)
+    if mismatches:
+        raise ValueError(f"{label}: program_status mismatch: {mismatches}")
+    if not status.get("status_note"):
+        raise ValueError(f"{label}: program_status.status_note is required")
+
+
+def _require_prior(payload: dict[str, Any], label: str) -> None:
+    prior = payload.get("prior_final_release_candidate_readback") or {}
+    expected = {
+        "schema_version": "ai_agent_result_capture_final_release_candidate_readback_v1",
+        "final_release_candidate_readback_count": 5,
+        "rollback_candidate_readback_count": 5,
+        "candidate_acceptance_hold_count": 5,
+        "live_apply_candidate_hold_count": 5,
+        "blocked_final_candidate_transition_count": 6,
+        "operator_action_count": 5,
+        "approval_required_total": 8,
+        "blocked_total": 9,
+        "owner_release_approved_count": 0,
+        "maintenance_window_approved_count": 0,
+        "rollback_owner_confirmed_count": 0,
+        "post_release_verifier_ready_count": 0,
+        "final_release_candidate_approved_count": 0,
+        "final_release_candidate_pass_count": 0,
+        "rollback_release_pass_count": 0,
+        "live_apply_release_pass_count": 0,
+        "writer_apply_count": 0,
+        "execution_apply_count": 0,
+        "receipt_write_count": 0,
+        "result_capture_write_count": 0,
+        "learning_write_count": 0,
+        "playbook_trust_write_count": 0,
+        "reviewer_queue_write_count": 0,
+        "gateway_queue_write_count": 0,
+        "telegram_send_count": 0,
+        "bot_api_call_count": 0,
+        "report_receipt_write_count": 0,
+    }
+    mismatches = _mismatches(prior, expected)
+    if mismatches:
+        raise ValueError(f"{label}: prior_final_release_candidate_readback mismatch: {mismatches}")
+    if not prior.get("readiness_note"):
+        raise ValueError(f"{label}: prior_final_release_candidate_readback.readiness_note is required")
+
+
+def _require_truth(payload: dict[str, Any], label: str) -> None:
+    truth = payload.get("release_authorization_truth") or {}
+    required_true = {
+        "p2_133_final_candidate_loaded",
+        "release_authorization_hold_ready",
+        "rollback_authorization_hold_ready",
+        "release_window_hold_active",
+        "live_apply_authorization_hold_active",
+        "owner_authorization_review_still_required",
+        "maintenance_window_review_still_required",
+        "rollback_owner_review_required",
+        "redaction_review_required",
+        "release_authorization_hold_only",
+    }
+    missing = sorted(field for field in required_true if truth.get(field) is not True)
+    if missing:
+        raise ValueError(f"{label}: release authorization hold flags must remain true: {missing}")
+
+    required_false = {
+        "owner_release_authorized",
+        "owner_release_approved",
+        "maintenance_window_approved",
+        "rollback_owner_confirmed",
+        "post_release_verifier_ready",
+        "final_release_candidate_approved",
+        "final_release_candidate_passed",
+        "release_authorization_granted",
+        "release_authorization_passed",
+        "rollback_release_passed",
+        "live_apply_release_passed",
+        "writer_apply_enabled",
+        "execution_apply_enabled",
+        "receipt_write_enabled",
+        "reviewer_queue_write_enabled",
+        "gateway_queue_write_enabled",
+        "telegram_send_enabled",
+        "bot_api_call_enabled",
+        "report_receipt_write_enabled",
+        "result_capture_write_enabled",
+        "learning_write_enabled",
+        "playbook_trust_write_enabled",
+        "production_write_enabled",
+        "secret_read_enabled",
+        "destructive_operation_enabled",
+    }
+    unsafe = sorted(field for field in required_false if truth.get(field) is not False)
+    if unsafe:
+        raise ValueError(f"{label}: release authorization live/send/write flags must remain false: {unsafe}")
+
+    zero_counts = {
+        "owner_release_authorized_count",
+        "owner_release_approved_count",
+        "maintenance_window_approved_count",
+        "rollback_owner_confirmed_count",
+        "post_release_verifier_ready_count",
+        "final_release_candidate_approved_count",
+        "final_release_candidate_pass_count",
+        "release_authorization_granted_count",
+        "release_authorization_pass_count",
+        "rollback_release_pass_count",
+        "live_apply_release_pass_count",
+        "writer_apply_count",
+        "execution_apply_count",
+        "receipt_write_count",
+        "reviewer_queue_write_count",
+        "gateway_queue_write_count",
+        "telegram_send_count",
+        "bot_api_call_count",
+        "report_receipt_write_count",
+        "result_capture_write_count",
+        "learning_write_count",
+        "playbook_trust_write_count",
+        "production_write_count",
+        "secret_read_count",
+        "destructive_operation_count",
+    }
+    non_zero = sorted(field for field in zero_counts if truth.get(field) != 0)
+    if non_zero:
+        raise ValueError(f"{label}: release authorization live counters must remain zero: {non_zero}")
+    if not truth.get("truth_note"):
+        raise ValueError(f"{label}: release_authorization_truth.truth_note is required")
+
+
+def _require_authorization_holds(payload: dict[str, Any], label: str) -> None:
+    items = payload.get("release_authorization_holds") or []
+    _require_ids(
+        items,
+        "hold_id",
+        {
+            "release_authorization_result_capture",
+            "release_authorization_learning",
+            "release_authorization_playbook_trust",
+            "release_authorization_reviewer_queue",
+            "release_authorization_gateway_queue",
+        },
+        label,
+        "release authorization holds",
+    )
+    for item in items:
+        item_id = item.get("hold_id")
+        if item.get("hold_mode") != "release_authorization_hold":
+            raise ValueError(f"{label}: release authorization hold {item_id} mode is invalid")
+        if item.get("owner_release_authorized") is not False:
+            raise ValueError(f"{label}: release authorization hold {item_id} must stay unauthorized")
+        if item.get("release_authorization_granted") is not False or item.get("release_authorization_passed") is not False:
+            raise ValueError(f"{label}: release authorization hold {item_id} must stay ungranted and unpassed")
+        _require_valid_status(item, label, f"release authorization hold {item_id}")
+        if not item.get("authorization_summary") or not _is_redacted_sha256(item.get("evidence_hash")):
+            raise ValueError(f"{label}: release authorization hold {item_id} must include summary and redacted evidence_hash")
+
+
+def _require_rollback_holds(payload: dict[str, Any], label: str) -> None:
+    items = payload.get("rollback_authorization_holds") or []
+    _require_count(items, 5, label, "rollback authorization holds")
+    for item in items:
+        item_id = item.get("hold_id")
+        if item.get("hold_mode") != "rollback_authorization_hold":
+            raise ValueError(f"{label}: rollback authorization hold {item_id} mode is invalid")
+        if item.get("rollback_owner_required") is not True:
+            raise ValueError(f"{label}: rollback authorization hold {item_id} must require rollback owner")
+        if item.get("rollback_owner_confirmed") is not False or item.get("rollback_release_enabled") is not False:
+            raise ValueError(f"{label}: rollback authorization hold {item_id} must stay unconfirmed and disabled")
+        _require_valid_status(item, label, f"rollback authorization hold {item_id}")
+        if not item.get("rollback_summary"):
+            raise ValueError(f"{label}: rollback authorization hold {item_id}.rollback_summary is required")
+
+
+def _require_release_window_holds(payload: dict[str, Any], label: str) -> None:
+    items = payload.get("release_window_holds") or []
+    _require_count(items, 5, label, "release window holds")
+    for item in items:
+        item_id = item.get("hold_id")
+        if item.get("hold_mode") != "release_window_hold":
+            raise ValueError(f"{label}: release window hold {item_id} mode is invalid")
+        if item.get("owner_release_authorized") is not False or item.get("maintenance_window_approved") is not False:
+            raise ValueError(f"{label}: release window hold {item_id} must keep owner release and maintenance unapproved")
+        if item.get("final_release_candidate_passed") is not False:
+            raise ValueError(f"{label}: release window hold {item_id} must keep final candidate unpassed")
+        _require_valid_status(item, label, f"release window hold {item_id}")
+        if not item.get("hold_reason"):
+            raise ValueError(f"{label}: release window hold {item_id}.hold_reason is required")
+
+
+def _require_live_apply_holds(payload: dict[str, Any], label: str) -> None:
+    items = payload.get("live_apply_authorization_holds") or []
+    _require_count(items, 5, label, "live apply authorization holds")
+    for item in items:
+        item_id = item.get("hold_id")
+        if item.get("hold_mode") != "live_apply_authorization_hold":
+            raise ValueError(f"{label}: live apply authorization hold {item_id} mode is invalid")
+        if item.get("release_authorization_granted") is not False or item.get("live_apply_release_enabled") is not False:
+            raise ValueError(f"{label}: live apply authorization hold {item_id} must stay ungranted and disabled")
+        _require_valid_status(item, label, f"live apply authorization hold {item_id}")
+        if not item.get("release_condition"):
+            raise ValueError(f"{label}: live apply authorization hold {item_id}.release_condition is required")
+
+
+def _require_blocked_transitions(payload: dict[str, Any], label: str) -> None:
+    items = payload.get("blocked_authorization_transitions") or []
+    _require_count(items, 6, label, "blocked authorization transitions")
+    critical_count = 0
+    for item in items:
+        item_id = item.get("blocker_id")
+        if item.get("severity") == "critical":
+            critical_count += 1
+        if item.get("status") not in {"approval_required", "blocked_by_policy"}:
+            raise ValueError(f"{label}: blocked authorization transition {item_id} status is invalid")
+        if not item.get("blocked_action") or not item.get("blocked_until") or not _is_redacted_sha256(item.get("evidence_hash")):
+            raise ValueError(f"{label}: blocked authorization transition {item_id} must include blocked action, until, and evidence hash")
+    if critical_count != 5:
+        raise ValueError(f"{label}: expected 5 critical authorization blockers, got {critical_count}")
+
+
+def _require_actions(payload: dict[str, Any], label: str) -> None:
+    items = payload.get("operator_actions") or []
+    _require_count(items, 5, label, "operator actions")
+    for item in items:
+        item_id = item.get("action_id")
+        if item.get("runtime_write_allowed") is not False:
+            raise ValueError(f"{label}: operator action {item_id} must not allow runtime writes")
+        if item.get("status") not in {"ready_for_operator_review", "approval_required"}:
+            raise ValueError(f"{label}: operator action {item_id} status is invalid")
+        if not item.get("operator_instruction"):
+            raise ValueError(f"{label}: operator action {item_id}.operator_instruction is required")
+
+
+def _require_display_redaction(payload: dict[str, Any], label: str) -> None:
+    contract = payload.get("display_redaction_contract") or {}
+    expected = {
+        "redaction_required": True,
+        "raw_prompt_display_allowed": False,
+        "private_reasoning_display_allowed": False,
+        "secret_value_display_allowed": False,
+        "raw_runtime_payload_display_allowed": False,
+        "internal_collaboration_content_display_allowed": False,
+    }
+    mismatches = _mismatches(contract, expected)
+    if mismatches:
+        raise ValueError(f"{label}: display_redaction_contract mismatch: {mismatches}")
+    if not contract.get("frontend_display_policy"):
+        raise ValueError(f"{label}: display_redaction_contract.frontend_display_policy is required")
+
+
+def _require_no_forbidden_display_terms(payload: dict[str, Any], label: str) -> None:
+    forbidden_terms = {
+        "work_window_transcript",
+        "session_id",
+        "browser_context",
+        "authorization_header",
+        "raw Telegram payload",
+        "private reasoning",
+        "raw prompt",
+        "chain-of-thought",
+        "批准！繼續",
+        "My request for Codex",
+        "In app browser",
+    }
+    display_blob = json.dumps(
+        {
+            "program_status": payload.get("program_status"),
+            "release_authorization_holds": payload.get("release_authorization_holds"),
+            "rollback_authorization_holds": payload.get("rollback_authorization_holds"),
+            "release_window_holds": payload.get("release_window_holds"),
+            "live_apply_authorization_holds": payload.get("live_apply_authorization_holds"),
+            "operator_actions": payload.get("operator_actions"),
+            "display_redaction_contract": payload.get("display_redaction_contract"),
+        },
+        ensure_ascii=False,
+    )
+    leaked = sorted(term for term in forbidden_terms if term in display_blob)
+    if leaked:
+        raise ValueError(f"{label}: forbidden display terms leaked: {leaked}")
+
+
+def _require_rollup_consistency(payload: dict[str, Any], label: str) -> None:
+    rollups = payload.get("rollups") or {}
+    expected = {
+        "release_authorization_hold_count": 5,
+        "rollback_authorization_hold_count": 5,
+        "release_window_hold_count": 5,
+        "live_apply_authorization_hold_count": 5,
+        "blocked_authorization_transition_count": 6,
+        "operator_action_count": 5,
+        "approval_required_authorization_count": 2,
+        "blocked_authorization_count": 1,
+        "approval_required_rollback_count": 2,
+        "blocked_rollback_count": 1,
+        "approval_required_release_window_count": 2,
+        "blocked_release_window_count": 1,
+        "approval_required_live_apply_count": 2,
+        "blocked_live_apply_count": 1,
+        "critical_blocker_count": 5,
+        "owner_release_authorized_count": 0,
+        "owner_release_approved_count": 0,
+        "maintenance_window_approved_count": 0,
+        "rollback_owner_confirmed_count": 0,
+        "post_release_verifier_ready_count": 0,
+        "final_release_candidate_approved_count": 0,
+        "final_release_candidate_pass_count": 0,
+        "release_authorization_granted_count": 0,
+        "release_authorization_pass_count": 0,
+        "rollback_release_pass_count": 0,
+        "live_apply_release_pass_count": 0,
+        "writer_apply_count": 0,
+        "execution_apply_count": 0,
+        "receipt_write_count": 0,
+        "reviewer_queue_write_count": 0,
+        "gateway_queue_write_count": 0,
+        "telegram_send_count": 0,
+        "bot_api_call_count": 0,
+        "report_receipt_write_count": 0,
+        "result_capture_write_count": 0,
+        "learning_write_count": 0,
+        "playbook_trust_write_count": 0,
+        "production_write_count": 0,
+        "secret_read_count": 0,
+        "destructive_operation_count": 0,
+    }
+    mismatches = _mismatches(rollups, expected)
+    if mismatches:
+        raise ValueError(f"{label}: rollups mismatch: {mismatches}")
+
+
+def _require_valid_status(item: dict[str, Any], label: str, item_label: str) -> None:
+    if item.get("status") not in {"ready_for_owner_review", "approval_required", "blocked_by_policy"}:
+        raise ValueError(f"{label}: {item_label} status is invalid")
+
+
+def _require_ids(
+    items: list[dict[str, Any]],
+    id_field: str,
+    expected_ids: set[str],
+    label: str,
+    item_label: str,
+) -> None:
+    actual_ids = {str(item.get(id_field)) for item in items}
+    if actual_ids != expected_ids:
+        raise ValueError(f"{label}: {item_label} ids mismatch: expected={sorted(expected_ids)} actual={sorted(actual_ids)}")
+
+
+def _require_count(items: list[dict[str, Any]], expected_count: int, label: str, item_label: str) -> None:
+    if len(items) != expected_count:
+        raise ValueError(f"{label}: expected {expected_count} {item_label}, got {len(items)}")
+
+
+def _mismatches(data: dict[str, Any], expected: dict[str, Any]) -> dict[str, dict[str, Any]]:
+    mismatches: dict[str, dict[str, Any]] = {}
+    for key, expected_value in expected.items():
+        actual_value = data.get(key)
+        if actual_value != expected_value:
+            mismatches[key] = {"expected": expected_value, "actual": actual_value}
+    return mismatches
+
+
+def _is_redacted_sha256(value: object) -> bool:
+    return isinstance(value, str) and re.fullmatch(r"sha256:[0-9a-f]{64}", value) is not None