feat(governance): 新增 report live delivery 批准包
This commit is contained in:
@@ -109,6 +109,9 @@ from src.services.ai_agent_runtime_readback_approval_package import (
|
||||
from src.services.ai_agent_runtime_readback_implementation_review import (
|
||||
load_latest_ai_agent_runtime_readback_implementation_review,
|
||||
)
|
||||
from src.services.ai_agent_report_live_delivery_approval_package import (
|
||||
load_latest_ai_agent_report_live_delivery_approval_package,
|
||||
)
|
||||
from src.services.ai_agent_report_automation_review import (
|
||||
load_latest_ai_agent_report_automation_review,
|
||||
)
|
||||
@@ -1359,6 +1362,36 @@ async def get_agent_runtime_readback_implementation_review() -> dict[str, Any]:
|
||||
) from exc
|
||||
|
||||
|
||||
@router.get(
|
||||
"/agent-report-live-delivery-approval-package",
|
||||
response_model=dict[str, Any],
|
||||
summary="取得 AI Agent report live delivery 批准包",
|
||||
description=(
|
||||
"讀取最新已提交的 P2-111 report live delivery approval package;"
|
||||
"此端點只回傳日報、週報、月報、失敗限定摘要與讀報回執的實發批准包、"
|
||||
"route lock、payload redaction、no-send receipt 與 operator action,"
|
||||
"不排程、不寫 Gateway queue、不送 Telegram、不呼叫 Bot API、不寫 report receipt、"
|
||||
"不啟動 AI analysis、不做中低風險自動優化、不寫 production target、不讀 secret。"
|
||||
),
|
||||
)
|
||||
async def get_agent_report_live_delivery_approval_package() -> dict[str, Any]:
|
||||
"""Return the latest read-only report live delivery approval package."""
|
||||
try:
|
||||
payload = await asyncio.to_thread(load_latest_ai_agent_report_live_delivery_approval_package)
|
||||
return redact_public_lan_topology(payload)
|
||||
except FileNotFoundError as exc:
|
||||
raise HTTPException(
|
||||
status_code=status.HTTP_404_NOT_FOUND,
|
||||
detail=str(exc),
|
||||
) from exc
|
||||
except (json.JSONDecodeError, ValueError) as exc:
|
||||
logger.error("ai_agent_report_live_delivery_approval_package_invalid", error=str(exc))
|
||||
raise HTTPException(
|
||||
status_code=status.HTTP_500_INTERNAL_SERVER_ERROR,
|
||||
detail="AI Agent report live delivery approval package 無效",
|
||||
) from exc
|
||||
|
||||
|
||||
@router.get(
|
||||
"/agent-owner-approved-fixture-dry-run",
|
||||
response_model=dict[str, Any],
|
||||
|
||||
@@ -0,0 +1,418 @@
|
||||
"""
|
||||
AI Agent report live delivery approval package snapshot.
|
||||
|
||||
Loads the latest committed P2-111 report delivery approval package. This module
|
||||
validates committed evidence only; it never schedules report delivery, writes
|
||||
Gateway queues, sends Telegram messages, calls Bot API, writes read receipts,
|
||||
starts AI analysis workers, writes production optimization results, reads
|
||||
secrets, or runs destructive operations.
|
||||
"""
|
||||
|
||||
from __future__ import annotations
|
||||
|
||||
import json
|
||||
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_report_live_delivery_approval_package_*.json"
|
||||
_SCHEMA_VERSION = "ai_agent_report_live_delivery_approval_package_v1"
|
||||
_RUNTIME_AUTHORITY = "report_live_delivery_approval_package_only_no_live_send_or_write"
|
||||
|
||||
|
||||
def load_latest_ai_agent_report_live_delivery_approval_package(
|
||||
evaluations_dir: Path | None = None,
|
||||
) -> dict[str, Any]:
|
||||
"""Load the newest committed report live delivery approval package."""
|
||||
directory = evaluations_dir or _DEFAULT_EVALUATIONS_DIR
|
||||
candidates = sorted(directory.glob(_SNAPSHOT_PATTERN))
|
||||
if not candidates:
|
||||
raise FileNotFoundError(f"no AI Agent report live delivery approval package 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")
|
||||
_require_schema(payload, str(latest))
|
||||
_require_prior_report_status_board(payload, str(latest))
|
||||
_require_prior_runtime_review(payload, str(latest))
|
||||
_require_delivery_truth(payload, str(latest))
|
||||
_require_delivery_packets(payload, str(latest))
|
||||
_require_route_lock_gates(payload, str(latest))
|
||||
_require_payload_redaction_checks(payload, str(latest))
|
||||
_require_dry_run_receipts(payload, str(latest))
|
||||
_require_operator_actions(payload, str(latest))
|
||||
_require_display_redaction(payload, str(latest))
|
||||
_require_no_forbidden_display_terms(payload, str(latest))
|
||||
_require_rollup_consistency(payload, str(latest))
|
||||
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 {}
|
||||
if status.get("read_only_mode") is not True:
|
||||
raise ValueError(f"{label}: program_status.read_only_mode must be true")
|
||||
if status.get("runtime_authority") != _RUNTIME_AUTHORITY:
|
||||
raise ValueError(f"{label}: runtime_authority must remain {_RUNTIME_AUTHORITY}")
|
||||
if status.get("current_task_id") != "P2-111":
|
||||
raise ValueError(f"{label}: current_task_id must be P2-111")
|
||||
if status.get("next_task_id") != "P2-112":
|
||||
raise ValueError(f"{label}: next_task_id must be P2-112")
|
||||
if status.get("overall_completion_percent") != 100:
|
||||
raise ValueError(f"{label}: P2-111 approval package must be 100 percent complete")
|
||||
|
||||
|
||||
def _require_prior_report_status_board(payload: dict[str, Any], label: str) -> None:
|
||||
prior = payload.get("prior_report_status_board") or {}
|
||||
expected = {
|
||||
"source_schema_version": "ai_agent_report_status_board_v1",
|
||||
"report_card_count": 3,
|
||||
"agent_status_report_count": 3,
|
||||
"visible_chart_count": 3,
|
||||
"operator_answer_count": 4,
|
||||
"work_units_total": 91,
|
||||
"work_units_done": 79,
|
||||
"work_units_waiting_approval": 12,
|
||||
"live_delivery_count": 0,
|
||||
"live_telegram_send_count_24h": 0,
|
||||
"live_auto_optimization_count_24h": 0,
|
||||
}
|
||||
mismatches = _mismatches(prior, expected)
|
||||
if mismatches:
|
||||
raise ValueError(f"{label}: P2-108 prior report status counts mismatch: {mismatches}")
|
||||
if not prior.get("readiness_note"):
|
||||
raise ValueError(f"{label}: prior_report_status_board.readiness_note is required")
|
||||
|
||||
|
||||
def _require_prior_runtime_review(payload: dict[str, Any], label: str) -> None:
|
||||
prior = payload.get("prior_runtime_review") or {}
|
||||
expected = {
|
||||
"approval_package_schema_version": "ai_agent_runtime_readback_approval_package_v1",
|
||||
"implementation_review_schema_version": "ai_agent_runtime_readback_implementation_review_v1",
|
||||
"telegram_failure_receipt_gate_count": 4,
|
||||
"implementation_blocker_count": 5,
|
||||
"no_write_verifier_check_count": 5,
|
||||
"owner_approval_received_count": 0,
|
||||
"runtime_readback_execution_count": 0,
|
||||
"live_query_count": 0,
|
||||
"telegram_failure_receipt_send_count": 0,
|
||||
"bot_api_call_count": 0,
|
||||
"gateway_queue_write_count": 0,
|
||||
"production_write_count": 0,
|
||||
}
|
||||
mismatches = _mismatches(prior, expected)
|
||||
if mismatches:
|
||||
raise ValueError(f"{label}: P2-109/P2-110 prior runtime counts mismatch: {mismatches}")
|
||||
if not prior.get("readiness_note"):
|
||||
raise ValueError(f"{label}: prior_runtime_review.readiness_note is required")
|
||||
|
||||
|
||||
def _require_delivery_truth(payload: dict[str, Any], label: str) -> None:
|
||||
truth = payload.get("delivery_approval_truth") or {}
|
||||
required_true = {
|
||||
"p2_108_report_status_loaded",
|
||||
"p2_109_failure_receipt_gate_loaded",
|
||||
"p2_110_implementation_review_loaded",
|
||||
"delivery_approval_package_ready",
|
||||
"daily_delivery_package_ready",
|
||||
"weekly_delivery_package_ready",
|
||||
"monthly_delivery_package_ready",
|
||||
"sre_war_room_route_locked",
|
||||
"payload_redaction_ready",
|
||||
"dry_run_receipt_ready",
|
||||
"owner_review_required_before_delivery",
|
||||
}
|
||||
missing = sorted(field for field in required_true if truth.get(field) is not True)
|
||||
if missing:
|
||||
raise ValueError(f"{label}: delivery approval ready flags must remain true: {missing}")
|
||||
|
||||
required_false = {
|
||||
"scheduler_enabled",
|
||||
"gateway_queue_write_enabled",
|
||||
"telegram_send_enabled",
|
||||
"bot_api_call_enabled",
|
||||
"report_receipt_write_enabled",
|
||||
"ai_analysis_run_enabled",
|
||||
"medium_low_auto_optimization_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}: live send/write flags must remain false: {unsafe}")
|
||||
|
||||
zero_counts = {
|
||||
"owner_approval_received_count",
|
||||
"scheduled_delivery_count_24h",
|
||||
"gateway_queue_write_count_24h",
|
||||
"telegram_send_count_24h",
|
||||
"bot_api_call_count_24h",
|
||||
"report_receipt_write_count_24h",
|
||||
"ai_analysis_run_count_24h",
|
||||
"auto_optimization_count_24h",
|
||||
"production_write_count_24h",
|
||||
"secret_read_count_24h",
|
||||
"destructive_operation_count_24h",
|
||||
}
|
||||
non_zero = sorted(field for field in zero_counts if truth.get(field) != 0)
|
||||
if non_zero:
|
||||
raise ValueError(f"{label}: report delivery live counters must remain zero: {non_zero}")
|
||||
if not truth.get("truth_note"):
|
||||
raise ValueError(f"{label}: delivery_approval_truth.truth_note is required")
|
||||
|
||||
|
||||
def _require_delivery_packets(payload: dict[str, Any], label: str) -> None:
|
||||
packets = payload.get("delivery_approval_packets") or []
|
||||
packet_ids = {packet.get("packet_id") for packet in packets}
|
||||
required = {
|
||||
"daily_report_delivery_approval",
|
||||
"weekly_report_delivery_approval",
|
||||
"monthly_report_delivery_approval",
|
||||
"failure_only_digest_approval",
|
||||
"report_receipt_readback_approval",
|
||||
}
|
||||
if packet_ids != required:
|
||||
raise ValueError(f"{label}: delivery approval packets must match {sorted(required)}")
|
||||
|
||||
valid_statuses = {"approval_required", "ready_for_owner_review", "blocked_by_policy"}
|
||||
valid_risks = {"medium", "high", "critical"}
|
||||
for packet in packets:
|
||||
packet_id = packet.get("packet_id")
|
||||
if packet.get("status") not in valid_statuses:
|
||||
raise ValueError(f"{label}: packet {packet_id} status is invalid")
|
||||
if packet.get("risk_tier") not in valid_risks:
|
||||
raise ValueError(f"{label}: packet {packet_id} risk_tier is invalid")
|
||||
if packet.get("approval_required") is not True:
|
||||
raise ValueError(f"{label}: packet {packet_id} approval_required must remain true")
|
||||
if packet.get("no_send_mode") is not True:
|
||||
raise ValueError(f"{label}: packet {packet_id} no_send_mode must remain true")
|
||||
if not packet.get("required_approval_fields") or not packet.get("blocked_runtime_actions"):
|
||||
raise ValueError(f"{label}: packet {packet_id} must list approval fields and blocked actions")
|
||||
if not packet.get("operator_guidance"):
|
||||
raise ValueError(f"{label}: packet {packet_id} must include operator guidance")
|
||||
if not _is_redacted_sha256(packet.get("evidence_hash")):
|
||||
raise ValueError(f"{label}: packet {packet_id} must expose evidence_hash")
|
||||
|
||||
|
||||
def _require_route_lock_gates(payload: dict[str, Any], label: str) -> None:
|
||||
gates = payload.get("route_lock_gates") or []
|
||||
gate_ids = {gate.get("gate_id") for gate in gates}
|
||||
required = {
|
||||
"sre_war_room_route_lock",
|
||||
"legacy_bot_suppression",
|
||||
"dedupe_fingerprint_lock",
|
||||
"delivery_window_gate",
|
||||
}
|
||||
if gate_ids != required:
|
||||
raise ValueError(f"{label}: route lock gates must match {sorted(required)}")
|
||||
for gate in gates:
|
||||
gate_id = gate.get("gate_id")
|
||||
if gate.get("status") not in {"ready_for_owner_review", "approval_required", "blocked_by_policy"}:
|
||||
raise ValueError(f"{label}: route gate {gate_id} status is invalid")
|
||||
if not gate.get("required_evidence") or not gate.get("blocked_routes"):
|
||||
raise ValueError(f"{label}: route gate {gate_id} must list evidence and blocked routes")
|
||||
for field in ("telegram_send_enabled", "bot_api_call_enabled", "gateway_queue_write_enabled"):
|
||||
if gate.get(field) is not False:
|
||||
raise ValueError(f"{label}: route gate {gate_id} {field} must remain false")
|
||||
if not _is_redacted_sha256(gate.get("evidence_hash")):
|
||||
raise ValueError(f"{label}: route gate {gate_id} must expose evidence_hash")
|
||||
|
||||
|
||||
def _require_payload_redaction_checks(payload: dict[str, Any], label: str) -> None:
|
||||
checks = payload.get("payload_redaction_checks") or []
|
||||
check_ids = {check.get("check_id") for check in checks}
|
||||
required = {
|
||||
"no_raw_prompt",
|
||||
"no_private_reasoning",
|
||||
"no_secret_values",
|
||||
"no_raw_telegram_payload",
|
||||
"no_internal_collaboration_content",
|
||||
}
|
||||
if check_ids != required:
|
||||
raise ValueError(f"{label}: payload redaction checks must match {sorted(required)}")
|
||||
for check in checks:
|
||||
check_id = check.get("check_id")
|
||||
if check.get("status") not in {"ready", "blocked_by_policy"}:
|
||||
raise ValueError(f"{label}: redaction check {check_id} status is invalid")
|
||||
if check.get("display_allowed") is not False:
|
||||
raise ValueError(f"{label}: redaction check {check_id} display_allowed must remain false")
|
||||
if not check.get("required_rule") or not check.get("failure_if_missing"):
|
||||
raise ValueError(f"{label}: redaction check {check_id} must include rule and failure explanation")
|
||||
if not _is_redacted_sha256(check.get("evidence_hash")):
|
||||
raise ValueError(f"{label}: redaction check {check_id} must expose evidence_hash")
|
||||
|
||||
|
||||
def _require_dry_run_receipts(payload: dict[str, Any], label: str) -> None:
|
||||
receipts = payload.get("dry_run_delivery_receipts") or []
|
||||
receipt_ids = {receipt.get("receipt_id") for receipt in receipts}
|
||||
required = {
|
||||
"daily_digest_no_send_receipt",
|
||||
"weekly_digest_no_send_receipt",
|
||||
"monthly_digest_no_send_receipt",
|
||||
"failure_only_no_send_receipt",
|
||||
}
|
||||
if receipt_ids != required:
|
||||
raise ValueError(f"{label}: dry-run receipts must match {sorted(required)}")
|
||||
for receipt in receipts:
|
||||
receipt_id = receipt.get("receipt_id")
|
||||
if receipt.get("status") not in {"ready_for_owner_review", "blocked_by_policy"}:
|
||||
raise ValueError(f"{label}: dry-run receipt {receipt_id} status is invalid")
|
||||
if receipt.get("live_send_count") != 0:
|
||||
raise ValueError(f"{label}: dry-run receipt {receipt_id} live_send_count must remain zero")
|
||||
if receipt.get("receipt_write_allowed") is not False:
|
||||
raise ValueError(f"{label}: dry-run receipt {receipt_id} receipt_write_allowed must remain false")
|
||||
if not receipt.get("required_fields"):
|
||||
raise ValueError(f"{label}: dry-run receipt {receipt_id} must list required fields")
|
||||
if not _is_redacted_sha256(receipt.get("evidence_hash")):
|
||||
raise ValueError(f"{label}: dry-run receipt {receipt_id} must expose evidence_hash")
|
||||
|
||||
|
||||
def _require_operator_actions(payload: dict[str, Any], label: str) -> None:
|
||||
actions = payload.get("operator_actions") or []
|
||||
action_types = {action.get("action_type") for action in actions}
|
||||
required = {
|
||||
"review_delivery_packet",
|
||||
"validate_sre_route",
|
||||
"validate_payload_redaction",
|
||||
"validate_zero_send_counters",
|
||||
"reject_or_promote",
|
||||
}
|
||||
if action_types != required:
|
||||
raise ValueError(f"{label}: operator actions must match {sorted(required)}")
|
||||
for action in actions:
|
||||
if action.get("live_send_allowed") is not False:
|
||||
raise ValueError(f"{label}: operator action {action.get('action_id')} must not allow live send")
|
||||
if not action.get("operator_instruction"):
|
||||
raise ValueError(f"{label}: operator action {action.get('action_id')} must include instruction")
|
||||
|
||||
|
||||
def _require_display_redaction(payload: dict[str, Any], label: str) -> None:
|
||||
contract = payload.get("display_redaction_contract") or {}
|
||||
if contract.get("redaction_required") is not True:
|
||||
raise ValueError(f"{label}: display redaction must remain required")
|
||||
required_false = {
|
||||
"raw_prompt_display_allowed",
|
||||
"private_reasoning_display_allowed",
|
||||
"secret_value_display_allowed",
|
||||
"raw_telegram_payload_display_allowed",
|
||||
"internal_collaboration_content_display_allowed",
|
||||
}
|
||||
unsafe = sorted(field for field in required_false if contract.get(field) is not False)
|
||||
if unsafe:
|
||||
raise ValueError(f"{label}: display redaction fields must remain false: {unsafe}")
|
||||
if not contract.get("frontend_display_policy"):
|
||||
raise ValueError(f"{label}: frontend_display_policy is required")
|
||||
|
||||
|
||||
def _require_no_forbidden_display_terms(payload: dict[str, Any], label: str) -> None:
|
||||
forbidden_terms = {
|
||||
"工作視窗",
|
||||
"對話內容",
|
||||
"批准!繼續",
|
||||
"In app browser",
|
||||
"My request for Codex",
|
||||
"browser_context",
|
||||
"codex_user_message",
|
||||
"prompt_text",
|
||||
"raw prompt",
|
||||
"private reasoning",
|
||||
"chain of thought",
|
||||
"private_reasoning",
|
||||
"chain_of_thought",
|
||||
"authorization_header",
|
||||
"authorization header",
|
||||
"secret value",
|
||||
"raw payload",
|
||||
"raw Telegram payload",
|
||||
"work window transcript",
|
||||
"internal collaboration transcript",
|
||||
}
|
||||
technical_identifier_fields = {
|
||||
"action_id",
|
||||
"action_type",
|
||||
"check_id",
|
||||
"gate_id",
|
||||
"packet_id",
|
||||
"receipt_id",
|
||||
"required_rule",
|
||||
}
|
||||
hits: list[str] = []
|
||||
|
||||
def walk(value: Any, path: str) -> None:
|
||||
if isinstance(value, dict):
|
||||
for key, nested in value.items():
|
||||
walk(nested, f"{path}.{key}" if path else str(key))
|
||||
return
|
||||
if isinstance(value, list):
|
||||
for index, nested in enumerate(value):
|
||||
walk(nested, f"{path}[{index}]")
|
||||
return
|
||||
if isinstance(value, str):
|
||||
field_name = path.rsplit(".", 1)[-1]
|
||||
if field_name in technical_identifier_fields:
|
||||
return
|
||||
matched = sorted(term for term in forbidden_terms if term in value)
|
||||
if matched:
|
||||
hits.append(f"{path}: {', '.join(matched)}")
|
||||
|
||||
walk(payload, "")
|
||||
if hits:
|
||||
raise ValueError(f"{label}: forbidden display terms found: {hits}")
|
||||
|
||||
|
||||
def _require_rollup_consistency(payload: dict[str, Any], label: str) -> None:
|
||||
rollups = payload.get("rollups") or {}
|
||||
truth = payload.get("delivery_approval_truth") or {}
|
||||
packets = payload.get("delivery_approval_packets") or []
|
||||
gates = payload.get("route_lock_gates") or []
|
||||
checks = payload.get("payload_redaction_checks") or []
|
||||
receipts = payload.get("dry_run_delivery_receipts") or []
|
||||
actions = payload.get("operator_actions") or []
|
||||
expected = {
|
||||
"delivery_approval_packet_count": len(packets),
|
||||
"route_lock_gate_count": len(gates),
|
||||
"payload_redaction_check_count": len(checks),
|
||||
"dry_run_delivery_receipt_count": len(receipts),
|
||||
"operator_action_count": len(actions),
|
||||
"approval_required_packet_count": sum(1 for packet in packets if packet.get("status") == "approval_required"),
|
||||
"blocked_packet_count": sum(1 for packet in packets if packet.get("status") == "blocked_by_policy"),
|
||||
"blocked_route_gate_count": sum(1 for gate in gates if gate.get("status") == "blocked_by_policy"),
|
||||
"blocked_receipt_count": sum(1 for receipt in receipts if receipt.get("status") == "blocked_by_policy"),
|
||||
"owner_approval_received_count": truth.get("owner_approval_received_count"),
|
||||
"scheduled_delivery_count": truth.get("scheduled_delivery_count_24h"),
|
||||
"gateway_queue_write_count": truth.get("gateway_queue_write_count_24h"),
|
||||
"telegram_send_count": truth.get("telegram_send_count_24h"),
|
||||
"bot_api_call_count": truth.get("bot_api_call_count_24h"),
|
||||
"report_receipt_write_count": truth.get("report_receipt_write_count_24h"),
|
||||
"ai_analysis_run_count": truth.get("ai_analysis_run_count_24h"),
|
||||
"auto_optimization_count": truth.get("auto_optimization_count_24h"),
|
||||
"production_write_count": truth.get("production_write_count_24h"),
|
||||
"secret_read_count": truth.get("secret_read_count_24h"),
|
||||
"destructive_operation_count": truth.get("destructive_operation_count_24h"),
|
||||
}
|
||||
mismatches = _mismatches(rollups, expected)
|
||||
if mismatches:
|
||||
raise ValueError(f"{label}: rollup counts mismatch: {mismatches}")
|
||||
|
||||
|
||||
def _mismatches(payload: dict[str, Any], expected: dict[str, Any]) -> dict[str, dict[str, Any]]:
|
||||
return {
|
||||
key: {"expected": expected_value, "actual": payload.get(key)}
|
||||
for key, expected_value in expected.items()
|
||||
if payload.get(key) != expected_value
|
||||
}
|
||||
|
||||
|
||||
def _is_redacted_sha256(value: Any) -> bool:
|
||||
if not isinstance(value, str):
|
||||
return False
|
||||
if not value.startswith("sha256:") or len(value) != 71:
|
||||
return False
|
||||
return all(char in "0123456789abcdef" for char in value.removeprefix("sha256:"))
|
||||
Reference in New Issue
Block a user