Merge remote-tracking branch 'origin/main' into litellm_fix_concurrent_single_agent_publish

.github/workflows/test-unit.yml

@@ -137,6 +137,7 @@ jobs:
         include:
           - shard: core-utils
             test-path: |-
+              tests/unit/accounting
               tests/unit/batch_completion
               tests/unit/decisions
               tests/unit/extraction

ci_cd/spend_logs_guard.py

@@ -4,7 +4,9 @@ needs an approving review from the spend-log owner before it can merge.
 CODEOWNERS can only match file paths, and every schema change edits the same schema.prisma copies,
 so a path rule would put every migration on the owner. This gate reads the diff instead: when no
 added line in the watched files mentions the table it passes at once, otherwise it passes only once
-the owner's latest review on the pull request is an approval. The workflow re-runs on every review.
+the owner's latest review on the pull request is an approval that still covers every spend-log line
+in the head (approvals survive pushes, so a later commit may not add spend-log lines the owner never
+saw). The workflow re-runs on every review.
 """
 
 from __future__ import annotations
@@ -13,7 +15,7 @@ import argparse
 import re
 import subprocess
 import sys
-from collections.abc import Sequence
+from collections.abc import Callable, Sequence
 from pathlib import Path
 from typing import Final
 
@@ -37,6 +39,7 @@ class Review(BaseModel):
     model_config = ConfigDict(frozen=True, extra="ignore")
     state: str
     user: Reviewer
+    commit_id: str | None = None
 
 
 _REVIEWS: Final = TypeAdapter(list[Review])
@@ -75,6 +78,24 @@ def approved_by(reviews: Sequence[Review], login: str) -> bool:
     return bool(states) and states[-1] == "APPROVED"
 
 
+def latest_approval(reviews: Sequence[Review], login: str) -> Review | None:
+    """`login`'s most recent APPROVED review on the pull request, the one GitHub keeps in force after pushes."""
+    own: Final = tuple(review for review in reviews if review.user.login == login and review.state == "APPROVED")
+    return own[-1] if own else None
+
+
+def approval_covers(
+    approval: Review, head: str, hits: Sequence[str], hits_at: Callable[[str], Sequence[str] | None]
+) -> bool:
+    """True when the approval is on the head, or the head adds no spend-log line the approved commit did not."""
+    if approval.commit_id is None:
+        return False
+    if approval.commit_id == head:
+        return True
+    reviewed: Final = hits_at(approval.commit_id)
+    return reviewed is not None and set(hits) <= set(reviewed)
+
+
 def added_diff(base: str, head: str) -> str:
     result: Final = subprocess.run(
         ("git", "diff", "--unified=0", "--no-color", f"{base}..{head}", "--", *WATCHED, MIGRATIONS_PREFIX),
@@ -85,6 +106,15 @@ def added_diff(base: str, head: str) -> str:
     return result.stdout
 
 
+def fetched(commit: str) -> bool:
+    result: Final = subprocess.run(("git", "fetch", "--no-tags", "--depth=1", "origin", commit), check=False)
+    return result.returncode == 0
+
+
+def reviewed_hits(base: str, commit: str) -> tuple[str, ...] | None:
+    return spend_log_hits(added_diff(base, commit)) if fetched(commit) else None
+
+
 class Arguments(BaseModel):
     model_config = ConfigDict(frozen=True)
     base: str
@@ -104,14 +134,21 @@ def main(argv: Sequence[str]) -> int:
     if not hits:
         sys.stdout.write("No added line in schema.prisma or a migration names LiteLLM_SpendLogs; nothing to review.\n")
         return 0
-    if approved_by(_REVIEWS.validate_json(args.reviews.read_bytes()), args.approver):
+    reviews: Final = _REVIEWS.validate_json(args.reviews.read_bytes())
+    approval: Final = latest_approval(reviews, args.approver) if approved_by(reviews, args.approver) else None
+    if approval is not None and approval_covers(
+        approval, args.head, hits, lambda commit: reviewed_hits(args.base, commit)
+    ):
         sys.stdout.write(f"LiteLLM_SpendLogs change approved by {args.approver}: {', '.join(hits)}\n")
         return 0
     listed: Final = "".join(f"  {hit}\n" for hit in hits)
-    sys.stdout.write(
+    reason: Final = (
         f"This pull request changes LiteLLM_SpendLogs and needs an approving review from @{args.approver}:\n"
-        f"{listed}The check re-runs when a review is submitted.\n"
+        if approval is None
+        else f"LiteLLM_SpendLogs lines changed after @{args.approver} approved {approval.commit_id}, "
+        f"a new approving review is needed:\n"
     )
+    sys.stdout.write(f"{reason}{listed}The check re-runs when a review is submitted.\n")
     return 1
 
 

enterprise/litellm_enterprise/enterprise_callbacks/secret_detection.py

@@ -32,6 +32,7 @@ from litellm.integrations.custom_guardrail import (
     CustomGuardrail,
     log_guardrail_information,
 )
+from litellm.llms.base_llm.guardrail_translation.utils import effective_skip_tool_message_for_guardrail
 from litellm.proxy._types import UserAPIKeyAuth
 from litellm.proxy.guardrails._content_utils import iter_tool_json_fields, map_tool_json_text, walk_user_text
 from litellm.types.guardrails import GuardrailEventHooks
@@ -753,7 +754,12 @@ class _ENTERPRISE_SecretDetection(CustomGuardrail):
         _masked_entity_count.set({})
 
         # Covers multimodal list content + Responses-API input.
-        inspected = walk_user_text(data, self.redact_text, self.redact_tool_payload) + self._redact_prompt(data)
+        inspected = walk_user_text(
+            data,
+            self.redact_text,
+            self.redact_tool_payload,
+            skip_tool_message=effective_skip_tool_message_for_guardrail(self),
+        ) + self._redact_prompt(data)
 
         if inspected == 0:
             # Image-only, empty-text, and unsupported payloads inspected

litellm-proxy-extras/litellm_proxy_extras/migrations/20261027000000_add_accounting_ledger_tables/migration.sql

@@ -0,0 +1,47 @@
+SET LOCAL lock_timeout = '10s';
+
+CREATE TYPE "AccountingEventKind" AS ENUM ('inbound_usage', 'outbound_usage', 'cost');
+
+CREATE TYPE "AccountingDeliveryState" AS ENUM ('pending', 'applied', 'parked');
+
+CREATE TYPE "AccountingDeliveryTarget" AS ENUM ('LiteLLM_VerificationToken', 'LiteLLM_UserTable', 'LiteLLM_TeamTable', 'LiteLLM_TeamMembership', 'LiteLLM_OrganizationTable', 'LiteLLM_OrganizationMembership', 'LiteLLM_EndUserTable', 'LiteLLM_AgentsTable', 'LiteLLM_TagTable', 'LiteLLM_ProjectTable', 'LiteLLM_ModelAccessGroupBudgetTable', 'LiteLLM_BudgetTable', 'LiteLLM_BudgetWindowSpend', 'LiteLLM_DailyUserSpend', 'LiteLLM_DailyTeamSpend', 'LiteLLM_DailyOrganizationSpend', 'LiteLLM_DailyEndUserSpend', 'LiteLLM_DailyAgentSpend', 'LiteLLM_DailyTagSpend', 'LiteLLM_DailyModelUsage', 'LiteLLM_DailyToolSpend', 'LiteLLM_AutoRouterDailySpend');
+
+-- CreateTable
+CREATE TABLE IF NOT EXISTS "LiteLLM_AccountingEvent" (
+    "event_id" TEXT NOT NULL,
+    "kind" "AccountingEventKind" NOT NULL,
+    "source_event_id" TEXT,
+    "schema_version" INTEGER NOT NULL,
+    "payload_hash" TEXT NOT NULL,
+    "occurred_at" TIMESTAMP(3) NOT NULL,
+    "recorded_at" TIMESTAMP(3) NOT NULL DEFAULT CURRENT_TIMESTAMP,
+    "payload" JSONB NOT NULL,
+
+    CONSTRAINT "LiteLLM_AccountingEvent_schema_version_check" CHECK ("schema_version" > 0),
+    CONSTRAINT "LiteLLM_AccountingEvent_pkey" PRIMARY KEY ("event_id")
+);
+
+-- CreateTable
+CREATE TABLE IF NOT EXISTS "LiteLLM_AccountingDelivery" (
+    "event_id" TEXT NOT NULL,
+    "target" "AccountingDeliveryTarget" NOT NULL,
+    "version" INTEGER NOT NULL,
+    "state" "AccountingDeliveryState" NOT NULL DEFAULT 'pending',
+    "attempts" INTEGER NOT NULL DEFAULT 0,
+    "next_attempt_at" TIMESTAMP(3) NOT NULL DEFAULT CURRENT_TIMESTAMP,
+    "applied_at" TIMESTAMP(3),
+    "last_error" TEXT,
+
+    CONSTRAINT "LiteLLM_AccountingDelivery_version_check" CHECK ("version" > 0),
+    CONSTRAINT "LiteLLM_AccountingDelivery_attempts_check" CHECK ("attempts" >= 0),
+    CONSTRAINT "LiteLLM_AccountingDelivery_pkey" PRIMARY KEY ("event_id","target","version")
+);
+
+-- CreateIndex
+CREATE INDEX IF NOT EXISTS "LiteLLM_AccountingEvent_recorded_at_idx" ON "LiteLLM_AccountingEvent"("recorded_at");
+
+-- CreateIndex
+CREATE INDEX IF NOT EXISTS "LiteLLM_AccountingEvent_kind_source_event_id_idx" ON "LiteLLM_AccountingEvent"("kind", "source_event_id");
+
+-- CreateIndex
+CREATE INDEX IF NOT EXISTS "LiteLLM_AccountingDelivery_target_state_next_attempt_at_idx" ON "LiteLLM_AccountingDelivery"("target", "state", "next_attempt_at");

litellm-proxy-extras/litellm_proxy_extras/schema.prisma

@@ -2061,3 +2061,73 @@ model LiteLLM_TelemetryReport {
   @@index([window_end])
   @@index([created_at])
 }
+
+enum AccountingEventKind {
+  inbound_usage
+  outbound_usage
+  cost
+}
+
+enum AccountingDeliveryState {
+  pending
+  applied
+  parked
+}
+
+enum AccountingDeliveryTarget {
+  LiteLLM_VerificationToken
+  LiteLLM_UserTable
+  LiteLLM_TeamTable
+  LiteLLM_TeamMembership
+  LiteLLM_OrganizationTable
+  LiteLLM_OrganizationMembership
+  LiteLLM_EndUserTable
+  LiteLLM_AgentsTable
+  LiteLLM_TagTable
+  LiteLLM_ProjectTable
+  LiteLLM_ModelAccessGroupBudgetTable
+  LiteLLM_BudgetTable
+  LiteLLM_BudgetWindowSpend
+  LiteLLM_DailyUserSpend
+  LiteLLM_DailyTeamSpend
+  LiteLLM_DailyOrganizationSpend
+  LiteLLM_DailyEndUserSpend
+  LiteLLM_DailyAgentSpend
+  LiteLLM_DailyTagSpend
+  LiteLLM_DailyModelUsage
+  LiteLLM_DailyToolSpend
+  LiteLLM_AutoRouterDailySpend
+}
+
+// Durable accounting ledger. One row per usage or cost event; payload holds the
+// private event envelope versioned by schema_version. Usage rows retain request-scoped
+// pricing context; cost rows include resolved rates and the finalized pricing snapshot.
+model LiteLLM_AccountingEvent {
+  event_id        String              @id
+  kind            AccountingEventKind
+  source_event_id String?
+  schema_version  Int
+  payload_hash    String
+  occurred_at     DateTime
+  recorded_at     DateTime            @default(now())
+  payload         Json
+
+  @@index([recorded_at])
+  @@index([kind, source_event_id])
+}
+
+// One row per (event, target table, version): pending work for the projector and,
+// once applied in the same transaction as the target write, the receipt that stops a retry.
+model LiteLLM_AccountingDelivery {
+  event_id        String
+  target          AccountingDeliveryTarget
+  version         Int
+  state           AccountingDeliveryState  @default(pending)
+  attempts        Int                      @default(0)
+  next_attempt_at DateTime                 @default(now())
+  applied_at      DateTime?
+  last_error      String?
+
+  @@id([event_id, target, version])
+  @@index([target, state, next_attempt_at])
+}

litellm-rust/crates/core-utils/src/core_helpers.rs

@@ -13,6 +13,7 @@ const FINISH_REASONS: &[(&str, &str)] = &[
     ("max_tokens", "length"),
     ("refusal", "content_filter"),
     ("compaction", "length"),
+    ("model_context_window_exceeded", "length"),
     ("guardrail_intervened", "content_filter"),
     ("content_filtered", "content_filter"),
     ("content_filter", "content_filter"),
@@ -74,6 +75,7 @@ mod tests {
     #[case::end_turn("end_turn", "stop")]
     #[case::stop_sequence("stop_sequence", "stop")]
     #[case::max_tokens("max_tokens", "length")]
+    #[case::model_context_window_exceeded("model_context_window_exceeded", "length")]
     #[case::refusal("refusal", "content_filter")]
     #[case::guardrail_intervened("guardrail_intervened", "content_filter")]
     #[case::content_filtered("content_filtered", "content_filter")]

litellm/accounting/__init__.py

@@ -0,0 +1,62 @@
+"""Public accounting contracts. Importing these values does not enable accounting."""
+
+from litellm.accounting.measurements import measurements_for
+from litellm.accounting.protocols import BudgetEnforcer, CostAggregator, CostCompiler, UsageMeter
+from litellm.accounting.types import (
+    AdmissionDecision,
+    AggregateQuery,
+    AggregateRow,
+    CostEvent,
+    DeliveryState,
+    DeliveryTarget,
+    Dimension,
+    DimensionFilter,
+    DimensionName,
+    EventKind,
+    Granularity,
+    InboundUsageEvent,
+    Measurement,
+    Metric,
+    MetricName,
+    Multiplier,
+    NamedCost,
+    OutboundUsageEvent,
+    Outcome,
+    PricedQuantity,
+    PricingSnapshot,
+    Rate,
+    Unpriced,
+    validate_aggregate_rows,
+)
+
+__all__ = [
+    "AdmissionDecision",
+    "AggregateQuery",
+    "AggregateRow",
+    "BudgetEnforcer",
+    "CostAggregator",
+    "CostCompiler",
+    "CostEvent",
+    "DeliveryState",
+    "DeliveryTarget",
+    "Dimension",
+    "DimensionFilter",
+    "DimensionName",
+    "EventKind",
+    "Granularity",
+    "InboundUsageEvent",
+    "Measurement",
+    "Metric",
+    "MetricName",
+    "Multiplier",
+    "NamedCost",
+    "OutboundUsageEvent",
+    "Outcome",
+    "PricedQuantity",
+    "PricingSnapshot",
+    "Rate",
+    "Unpriced",
+    "UsageMeter",
+    "measurements_for",
+    "validate_aggregate_rows",
+]

litellm/accounting/measurements.py

@@ -0,0 +1,13 @@
+from typing import Final
+
+from litellm.accounting.types import CostEvent, InboundUsageEvent, Measurement, OutboundUsageEvent
+
+
+def measurements_for(event: InboundUsageEvent | OutboundUsageEvent | CostEvent) -> tuple[Measurement, ...]:
+    """Return the complete contribution set for one event, ready for one apply call."""
+    metrics: Final = event.charges if isinstance(event, CostEvent) else event.metrics
+    occurred_at: Final = event.occurred_at if isinstance(event, CostEvent) else event.started_at
+    return tuple(
+        Measurement(source_event_id=event.event_id, occurred_at=occurred_at, dimensions=event.dimensions, metric=metric)
+        for metric in metrics or ()
+    )

litellm/accounting/protocols.py

@@ -0,0 +1,85 @@
+"""Contracts only. Implementations own persistence, pricing, projections and live budgets.
+
+The event row's ``payload`` is a private envelope versioned by the row's ``schema_version``.
+Usage envelopes retain the public event and request-scoped pricing context: quantities with units,
+model/provider, requested and billed tier/region when known, and per-request/deployment overrides.
+These inputs must survive request teardown. Cost-map rates are resolved at compilation, not capture.
+Cost envelopes retain the public CostEvent, including its required finalized pricing snapshot.
+T5/T6 own envelope serialization; this public API does not introduce a second persisted schema.
+"""
+
+from typing import Protocol
+
+from litellm.accounting.types import (
+    AdmissionDecision,
+    AggregateQuery,
+    AggregateRow,
+    CostEvent,
+    Dimension,
+    InboundUsageEvent,
+    Measurement,
+    Metric,
+    OutboundUsageEvent,
+    Unpriced,
+)
+
+
+class UsageMeter(Protocol):
+    async def record(self, event: InboundUsageEvent | OutboundUsageEvent) -> None:
+        """Acknowledge commit, no-op identical IDs and reject conflicting content.
+
+        Persist the event and its request-scoped pricing context before acknowledging.
+        HTTP acknowledgement is a separate policy. Unknown usage remains None.
+        """
+        ...
+
+
+class CostCompiler(Protocol):
+    async def compile(self, event: OutboundUsageEvent) -> CostEvent | Unpriced:
+        """Reuse the committed charge or finalize one charge and its delivery intent.
+
+        Resolve captured request-scoped context by event identity and cost-map rates
+        at compilation. Persist the resolved inputs in the required snapshot. Replay
+        reuses the committed result and never consumes free allowance again.
+        Missing pricing is Unpriced, never an inferred zero.
+        """
+        ...
+
+
+class CostAggregator(Protocol):
+    async def apply(self, measurements: tuple[Measurement, ...]) -> None:
+        """Apply one source event's complete contribution set per target/version.
+
+        Each target commits all contributions and its receipt atomically. A receipt
+        for the first metric must not suppress the rest. Usage need not wait for cost.
+        """
+        ...
+
+    async def query(self, query: AggregateQuery) -> tuple[AggregateRow, ...]:
+        """Return disjoint [start, end) rows per group using existing daily projections.
+
+        Filters combine with AND, values within a filter with IN. DAY emits daily
+        buckets; None emits whole-interval totals. Reject unsupported metric/scope
+        combinations. Different groups may overlap, including non-additive tags.
+        """
+        ...
+
+
+class BudgetEnforcer(Protocol):
+    async def reserve(
+        self, reservation_id: str, dimensions: tuple[Dimension, ...], estimated: tuple[Metric, ...]
+    ) -> AdmissionDecision:
+        """Apply existing admission policies and estimates; duplicate IDs add no hold."""
+        ...
+
+    async def settle(self, reservation_id: str | None, cost_event: CostEvent) -> None:
+        """Settle the committed charge with atomic deduplication and hold mutation.
+
+        Preserve scope/window attribution and retry unfinished settlement after
+        restart. SQL projection receipts do not provide live-counter atomicity.
+        """
+        ...
+
+    async def release(self, reservation_id: str) -> None:
+        """Idempotently release unused exposure at the request boundary, preserving charges."""
+        ...

litellm/accounting/types.py

@@ -0,0 +1,428 @@
+"""Accounting values use UTC instants, decimal strings on the wire and USD charges."""
+
+from collections.abc import Mapping
+from datetime import datetime, timezone
+from decimal import Decimal
+from enum import Enum
+from typing import Annotated, Final, Literal, TypeAlias
+
+from pydantic import AfterValidator, AwareDatetime, BaseModel, ConfigDict, Field, model_validator
+from typing_extensions import Self
+
+from litellm.types.utils import CallTypes
+
+_CONFIG: Final = ConfigDict(frozen=True, extra="forbid", validate_default=True)
+Identifier: TypeAlias = Annotated[str, Field(strict=True, min_length=1, pattern=r"\S")]
+Quantity: TypeAlias = Annotated[Decimal, Field(strict=True, ge=0, allow_inf_nan=False)]
+
+
+def _utc(value: datetime) -> datetime:
+    return value.astimezone(timezone.utc)
+
+
+Instant: TypeAlias = Annotated[AwareDatetime, AfterValidator(_utc)]
+
+
+class MetricName(str, Enum):
+    """Reasoning tokens are included in completion tokens; above-1hr writes are included in cache creation tokens."""
+
+    REQUESTS = "requests"
+    SUCCESSFUL_REQUESTS = "successful_requests"
+    FAILED_REQUESTS = "failed_requests"
+    PROMPT_TOKENS = "prompt_tokens"
+    COMPLETION_TOKENS = "completion_tokens"
+    REASONING_TOKENS = "reasoning_tokens"
+    TOTAL_TOKENS = "total_tokens"
+    CACHE_READ_INPUT_TOKENS = "cache_read_input_tokens"
+    CACHE_CREATION_INPUT_TOKENS = "cache_creation_input_tokens"
+    CACHE_CREATION_INPUT_TOKENS_ABOVE_1HR = "cache_creation_input_tokens_above_1hr"
+    SPEND = "spend"
+
+
+class DimensionName(str, Enum):
+    KEY = "key"
+    USER = "user"
+    TEAM = "team"
+    ORGANIZATION = "organization"
+    END_USER = "end_user"
+    MODEL = "model"
+    MODEL_GROUP = "model_group"
+    DEPLOYMENT = "deployment"
+    PROVIDER = "provider"
+    TOOL = "tool"
+    AGENT = "agent"
+    PROJECT = "project"
+    MODEL_ACCESS_GROUP = "model_access_group"
+    API_BASE = "api_base"
+    SERVICE_TIER = "service_tier"
+    TAG = "tag"
+
+
+class Outcome(str, Enum):
+    SUCCEEDED = "succeeded"
+    FAILED = "failed"
+    CANCELLED = "cancelled"
+
+
+class Granularity(str, Enum):
+    DAY = "day"
+
+
+class EventKind(str, Enum):
+    INBOUND_USAGE = "inbound_usage"
+    OUTBOUND_USAGE = "outbound_usage"
+    COST = "cost"
+
+
+class DeliveryState(str, Enum):
+    PENDING = "pending"
+    APPLIED = "applied"
+    PARKED = "parked"
+
+
+class DeliveryTarget(str, Enum):
+    """One member per existing table the projector writes; each is applied and receipted independently."""
+
+    KEY_SPEND = "LiteLLM_VerificationToken"
+    USER_SPEND = "LiteLLM_UserTable"
+    TEAM_SPEND = "LiteLLM_TeamTable"
+    TEAM_MEMBER_SPEND = "LiteLLM_TeamMembership"
+    ORGANIZATION_SPEND = "LiteLLM_OrganizationTable"
+    ORGANIZATION_MEMBER_SPEND = "LiteLLM_OrganizationMembership"
+    END_USER_SPEND = "LiteLLM_EndUserTable"
+    AGENT_SPEND = "LiteLLM_AgentsTable"
+    TAG_SPEND = "LiteLLM_TagTable"
+    PROJECT_SPEND = "LiteLLM_ProjectTable"
+    MODEL_ACCESS_GROUP_SPEND = "LiteLLM_ModelAccessGroupBudgetTable"
+    BUDGET_SPEND = "LiteLLM_BudgetTable"
+    BUDGET_WINDOW_SPEND = "LiteLLM_BudgetWindowSpend"
+    DAILY_USER_SPEND = "LiteLLM_DailyUserSpend"
+    DAILY_TEAM_SPEND = "LiteLLM_DailyTeamSpend"
+    DAILY_ORGANIZATION_SPEND = "LiteLLM_DailyOrganizationSpend"
+    DAILY_END_USER_SPEND = "LiteLLM_DailyEndUserSpend"
+    DAILY_AGENT_SPEND = "LiteLLM_DailyAgentSpend"
+    DAILY_TAG_SPEND = "LiteLLM_DailyTagSpend"
+    DAILY_MODEL_USAGE = "LiteLLM_DailyModelUsage"
+    DAILY_TOOL_SPEND = "LiteLLM_DailyToolSpend"
+    AUTOROUTER_DAILY_SPEND = "LiteLLM_AutoRouterDailySpend"
+
+
+_SUBSET_PAIRS: Final = (
+    (MetricName.REASONING_TOKENS, MetricName.COMPLETION_TOKENS),
+    (MetricName.CACHE_CREATION_INPUT_TOKENS_ABOVE_1HR, MetricName.CACHE_CREATION_INPUT_TOKENS),
+)
+
+
+def _check_subsets(values: Mapping[str, Decimal]) -> None:
+    for part, whole in _SUBSET_PAIRS:
+        if part.value in values and whole.value in values and values[part.value] > values[whole.value]:
+            raise ValueError(f"{part.value} cannot exceed {whole.value}")
+
+
+class Metric(BaseModel):
+    model_config = _CONFIG
+
+    name: MetricName
+    value: Quantity
+    unit: Literal["count", "USD"] = "count"
+
+    @model_validator(mode="after")
+    def validate_unit(self) -> Self:
+        if (self.name is MetricName.SPEND) != (self.unit == "USD"):
+            raise ValueError("SPEND requires USD; usage metrics require count")
+        if self.unit == "count" and self.value != self.value.to_integral_value():
+            raise ValueError("Count metrics require integral values")
+        return self
+
+
+class Dimension(BaseModel):
+    model_config = _CONFIG
+
+    name: DimensionName
+    value: Identifier
+
+
+_REPEATABLE_DIMENSIONS: Final = frozenset({DimensionName.TAG, DimensionName.MODEL_ACCESS_GROUP})
+
+
+def _dimensions(values: tuple[Dimension, ...]) -> tuple[Dimension, ...]:
+    if len(frozenset(values)) != len(values):
+        raise ValueError("Duplicate dimension values")
+    single_names: Final = tuple(value.name for value in values if value.name not in _REPEATABLE_DIMENSIONS)
+    if len(frozenset(single_names)) != len(single_names):
+        raise ValueError("Only tag and model access group dimensions may have multiple values")
+    return values
+
+
+Dimensions: TypeAlias = Annotated[tuple[Dimension, ...], AfterValidator(_dimensions)]
+_CALLER_METRICS: Final = frozenset((MetricName.REQUESTS, MetricName.SUCCESSFUL_REQUESTS, MetricName.FAILED_REQUESTS))
+
+
+def _metrics(values: tuple[Metric, ...]) -> tuple[Metric, ...]:
+    if len(frozenset(value.name for value in values)) != len(values):
+        raise ValueError("Duplicate metric names")
+    return values
+
+
+Metrics: TypeAlias = Annotated[tuple[Metric, ...], AfterValidator(_metrics)]
+
+
+def _interval(start: datetime, end: datetime, *, allow_empty: bool = False) -> None:
+    if end < start or (end == start and not allow_empty):
+        raise ValueError("End must follow start")
+
+
+class InboundUsageEvent(BaseModel):
+    model_config = _CONFIG
+
+    event_id: Identifier
+    request_id: Identifier
+    call_type: CallTypes
+    started_at: Instant
+    ended_at: Instant
+    outcome: Outcome
+    dimensions: Dimensions
+    metrics: Metrics
+
+    @model_validator(mode="after")
+    def validate_caller_activity(self) -> Self:
+        _interval(self.started_at, self.ended_at, allow_empty=True)
+        if any(metric.name not in _CALLER_METRICS for metric in self.metrics):
+            raise ValueError("Inbound events own only caller request metrics")
+        if not any(metric.name is MetricName.REQUESTS and metric.value == 1 for metric in self.metrics):
+            raise ValueError("Inbound events count exactly one caller request")
+        for metric in self.metrics:
+            if metric.name is MetricName.SUCCESSFUL_REQUESTS and metric.value != int(self.outcome is Outcome.SUCCEEDED):
+                raise ValueError("Successful request count must match outcome")
+            if metric.name is MetricName.FAILED_REQUESTS and metric.value != int(self.outcome is not Outcome.SUCCEEDED):
+                raise ValueError("Failed request count must match outcome")
+        return self
+
+
+class OutboundUsageEvent(BaseModel):
+    model_config = _CONFIG
+
+    event_id: Identifier
+    attempt_id: Identifier
+    request_id: Identifier | None
+    retry_of_attempt_id: Identifier | None
+    call_type: CallTypes
+    started_at: Instant
+    ended_at: Instant
+    outcome: Outcome
+    dimensions: Dimensions
+    metrics: Metrics | None
+
+    @model_validator(mode="after")
+    def validate_attempt_usage(self) -> Self:
+        _interval(self.started_at, self.ended_at, allow_empty=True)
+        if self.retry_of_attempt_id == self.attempt_id:
+            raise ValueError("An attempt cannot retry itself")
+        if self.metrics is not None and any(
+            metric.name in _CALLER_METRICS or metric.name is MetricName.SPEND for metric in self.metrics
+        ):
+            raise ValueError("Outbound events own only attempt usage metrics")
+        if self.metrics is not None:
+            _check_subsets({metric.name.value: metric.value for metric in self.metrics})
+        return self
+
+
+PricingUnit: TypeAlias = Literal["token", "second", "character", "image", "request"]
+
+
+class Rate(BaseModel):
+    model_config = _CONFIG
+
+    measure: Identifier
+    unit: PricingUnit
+    usd_per_unit: Quantity
+
+
+class PricedQuantity(BaseModel):
+    """A pricing input in its billing unit. Only seconds may be fractional."""
+
+    model_config = _CONFIG
+
+    measure: Identifier
+    unit: PricingUnit
+    value: Quantity
+
+    @model_validator(mode="after")
+    def validate_discrete_quantity(self) -> Self:
+        if self.unit != "second" and self.value != self.value.to_integral_value():
+            raise ValueError("Discrete priced quantities require integral values")
+        return self
+
+
+class Multiplier(BaseModel):
+    """A regional or tier factor applied on top of the priced key, e.g. data_residency=eu or vertex_location=us-east5."""
+
+    model_config = _CONFIG
+
+    name: Identifier
+    value: Identifier
+    factor: Annotated[Decimal, Field(strict=True, gt=0, allow_inf_nan=False)]
+
+
+class NamedCost(BaseModel):
+    """A free-form extra charge by name, e.g. azure_model_router_flat_cost."""
+
+    model_config = _CONFIG
+
+    name: Identifier
+    amount: Quantity
+
+
+class PricingSnapshot(BaseModel):
+    """Retained inputs and result of authoritative compilation, reproducible after cost-map rates change.
+
+    Request-scoped inputs and overrides come from the usage row's versioned payload. Cost-map rates are
+    resolved at compilation, not frozen at capture. Retries reuse the committed charge and snapshot.
+    Component names match ``CostBreakdown``: reasoning_cost is part of output_cost, not an additional charge.
+    Component sums and float tolerances belong to reconciliation, not this structural contract.
+    """
+
+    model_config = _CONFIG
+
+    schema_version: Literal[1] = 1
+    calculator_version: Identifier
+    model: Identifier
+    provider: Identifier | None
+    pricing_source: Literal["cost_map", "deployment", "per_request", "upstream_header"]
+    priced_model_key: Identifier
+    multipliers: tuple[Multiplier, ...] = ()
+    rates: tuple[Rate, ...] = ()
+    usage: tuple[PricedQuantity, ...] = ()
+    input_cost: Quantity | None = None
+    output_cost: Quantity | None = None
+    cache_read_cost: Quantity | None = None
+    cache_creation_cost: Quantity | None = None
+    reasoning_cost: Quantity | None = None
+    tool_usage_cost: Quantity | None = None
+    guardrail_cost: Quantity | None = None
+    additional_costs: tuple[NamedCost, ...] = ()
+    original_cost: Quantity | None = None
+    discount_percent: Quantity | None = None
+    discount_amount: Quantity | None = None
+    margin_percent: Quantity | None = None
+    margin_fixed_amount: Quantity | None = None
+    margin_total_amount: Quantity | None = None
+    upstream_reported_cost: Quantity | None = None
+    total_cost: Quantity
+
+    @model_validator(mode="after")
+    def validate_snapshot(self) -> Self:
+        rate_keys: Final = frozenset((rate.measure, rate.unit) for rate in self.rates)
+        quantity_keys: Final = frozenset((quantity.measure, quantity.unit) for quantity in self.usage)
+        if len(rate_keys) != len(self.rates):
+            raise ValueError("Duplicate rate measures and units")
+        if len(quantity_keys) != len(self.usage):
+            raise ValueError("Duplicate priced quantity measures and units")
+        if not rate_keys.issubset(quantity_keys):
+            raise ValueError("Every rate requires one priced quantity with the same measure and unit")
+        if len(frozenset(cost.name for cost in self.additional_costs)) != len(self.additional_costs):
+            raise ValueError("Duplicate additional cost names")
+        if len(frozenset(multiplier.name for multiplier in self.multipliers)) != len(self.multipliers):
+            raise ValueError("Duplicate multiplier names")
+        _check_subsets({quantity.measure: quantity.value for quantity in self.usage if quantity.unit == "token"})
+        return self
+
+
+class CostEvent(BaseModel):
+    model_config = _CONFIG
+
+    event_id: Identifier
+    usage_event_id: Identifier
+    occurred_at: Instant
+    dimensions: Dimensions
+    charges: Metrics
+    snapshot: PricingSnapshot
+
+    @model_validator(mode="after")
+    def validate_charge(self) -> Self:
+        if len(self.charges) != 1 or self.charges[0].name is not MetricName.SPEND:
+            raise ValueError("A cost event requires one authoritative USD spend metric")
+        if self.event_id == self.usage_event_id:
+            raise ValueError("Cost and usage events require distinct identities")
+        if self.snapshot.total_cost != self.charges[0].value:
+            raise ValueError("Snapshot total must equal the authoritative spend")
+        return self
+
+
+class Unpriced(BaseModel):
+    model_config = _CONFIG
+
+    usage_event_id: Identifier
+    reason: Identifier
+
+
+class Measurement(BaseModel):
+    model_config = _CONFIG
+
+    source_event_id: Identifier
+    occurred_at: Instant
+    dimensions: Dimensions
+    metric: Metric
+
+
+class DimensionFilter(BaseModel):
+    model_config = _CONFIG
+
+    name: DimensionName
+    values: Annotated[tuple[Identifier, ...], Field(min_length=1)]
+
+
+class AggregateQuery(BaseModel):
+    model_config = _CONFIG
+
+    metric: MetricName
+    filters: tuple[DimensionFilter, ...]
+    group_by: tuple[DimensionName, ...]
+    start: Instant
+    end: Instant
+    granularity: Granularity | None
+
+    @model_validator(mode="after")
+    def validate_query(self) -> Self:
+        _interval(self.start, self.end)
+        if len(frozenset(item.name for item in self.filters)) != len(self.filters):
+            raise ValueError("Use one IN filter per dimension; filters combine with AND")
+        if len(frozenset(self.group_by)) != len(self.group_by):
+            raise ValueError("Duplicate group-by dimensions")
+        if any(
+            (instant.hour, instant.minute, instant.second, instant.microsecond) != (0, 0, 0, 0)
+            for instant in (self.start, self.end)
+        ):
+            raise ValueError("Initial reporting queries require UTC day boundaries")
+        return self
+
+
+class AggregateRow(BaseModel):
+    model_config = _CONFIG
+
+    start: Instant
+    end: Instant
+    dimensions: Dimensions
+    value: Quantity
+
+    @model_validator(mode="after")
+    def validate_bounds(self) -> Self:
+        _interval(self.start, self.end)
+        return self
+
+
+def validate_aggregate_rows(rows: tuple[AggregateRow, ...]) -> tuple[AggregateRow, ...]:
+    """Validate [start, end) buckets per group; overlapping tag groups are not additive."""
+    groups: Final = frozenset(frozenset(row.dimensions) for row in rows)
+    for group in groups:
+        ordered = sorted((row for row in rows if frozenset(row.dimensions) == group), key=lambda row: row.start)
+        if any(left.end > right.start for left, right in zip(ordered, ordered[1:])):
+            raise ValueError("Aggregate buckets overlap within a dimension group")
+    return rows
+
+
+class AdmissionDecision(BaseModel):
+    model_config = _CONFIG
+
+    allowed: bool
+    reason: Identifier | None = None

litellm/integrations/otel/model/spans.py

@@ -338,6 +338,8 @@ _PRISMA_MODELS: Final[frozenset[str]] = frozenset(
         "LiteLLM_ToolTable",
         "LiteLLM_TelemetryReport",
         "LiteLLM_AccessGroupTable",
+        "LiteLLM_AccountingDelivery",
+        "LiteLLM_AccountingEvent",
         "LiteLLM_ClaudeCodePluginTable",
         "LiteLLM_MemoryTable",
         "LiteLLM_AdaptiveRouterState",

litellm/litellm_core_utils/core_helpers.py

@@ -266,6 +266,7 @@ _FINISH_REASON_MAP: Final[dict[str, OpenAIChatCompletionFinishReason]] = {
     "tool_use": "tool_calls",
     "refusal": "content_filter",
     "compaction": "length",
+    "model_context_window_exceeded": "length",
     # Cohere
     "COMPLETE": "stop",
     "ERROR_TOXIC": "content_filter",

litellm/proxy/guardrails/_content_utils.py

@@ -305,6 +305,11 @@ def iter_message_text(data: Mapping[str, object]) -> Iterator[str]:
         yield from _iter_text_parts_in_content(message.get("content"))
 
 
+def _is_tool_message(message: object) -> bool:
+    fields: Final = _text_object(message)
+    return fields is not None and str(fields.get("role") or "").lower() == "tool"
+
+
 def _call_item(item: object) -> tuple[str, object] | None:
     """Return ``(field, value)`` for a Responses tool-call input item, else None."""
     fields: Final = _text_object(item)
@@ -316,11 +321,14 @@ def walk_user_text(
     data: dict[str, object],
     visit: Callable[[str], str],
     visit_tool: Callable[[JsonValue], JsonValue] | None = None,
+    *,
+    skip_tool_message: bool = False,
 ) -> int:
     """Rewrite every text fragment in place via ``visit``.
 
     Mutates ``data["messages"]`` and ``data["input"]`` (including tool payloads), ``data["system"]`` and
-    ``data["instructions"]``. Returns the number
+    ``data["instructions"]``. ``skip_tool_message`` leaves ``role: tool`` messages and ``tool_result``
+    blocks as sent. Returns the number
     of fragments visited so callers can short-circuit when nothing was
     inspected.
     """
@@ -353,6 +361,8 @@ def walk_user_text(
 
     def _rewrite_part(part: Mapping[str, object], native: bool) -> Mapping[str, object]:
         nonlocal visited
+        if skip_tool_message and part.get("type") == "tool_result":
+            return part
         text: Final = _part_text(part)
         rewritten: Final = {**part, **({"text": visit(text)} if text is not None else {})}
         if text is not None:
@@ -384,18 +394,17 @@ def walk_user_text(
         return content
 
     messages: Final = data.get("messages")
-    if isinstance(messages, list):
-        for message in messages:
-            if not isinstance(message, dict):
-                continue
-            if "content" in message:
-                message["content"] = _rewrite_content(message["content"])
-            calls = message.get("tool_calls")
-            if isinstance(calls, list):
-                for call in calls:
-                    function = call.get("function") if isinstance(call, dict) else None
-                    if isinstance(function, dict) and "arguments" in function:
-                        function["arguments"] = _rewrite_arguments(function["arguments"])
+    for message in _object_list(messages) or ():
+        if (skip_tool_message and _is_tool_message(message)) or not isinstance(message, dict):
+            continue
+        if "content" in message:
+            message["content"] = _rewrite_content(message["content"])
+        calls = message.get("tool_calls")
+        if isinstance(calls, list):
+            for call in calls:
+                function = call.get("function") if isinstance(call, dict) else None
+                if isinstance(function, dict) and "arguments" in function:
+                    function["arguments"] = _rewrite_arguments(function["arguments"])
 
     if "system" in data:
         data["system"] = _rewrite_content(data["system"])  # rebind-ok: preserve the native hook's in-place contract

litellm/proxy/schema.prisma

@@ -2061,3 +2061,73 @@ model LiteLLM_TelemetryReport {
   @@index([window_end])
   @@index([created_at])
 }
+
+enum AccountingEventKind {
+  inbound_usage
+  outbound_usage
+  cost
+}
+
+enum AccountingDeliveryState {
+  pending
+  applied
+  parked
+}
+
+enum AccountingDeliveryTarget {
+  LiteLLM_VerificationToken
+  LiteLLM_UserTable
+  LiteLLM_TeamTable
+  LiteLLM_TeamMembership
+  LiteLLM_OrganizationTable
+  LiteLLM_OrganizationMembership
+  LiteLLM_EndUserTable
+  LiteLLM_AgentsTable
+  LiteLLM_TagTable
+  LiteLLM_ProjectTable
+  LiteLLM_ModelAccessGroupBudgetTable
+  LiteLLM_BudgetTable
+  LiteLLM_BudgetWindowSpend
+  LiteLLM_DailyUserSpend
+  LiteLLM_DailyTeamSpend
+  LiteLLM_DailyOrganizationSpend
+  LiteLLM_DailyEndUserSpend
+  LiteLLM_DailyAgentSpend
+  LiteLLM_DailyTagSpend
+  LiteLLM_DailyModelUsage
+  LiteLLM_DailyToolSpend
+  LiteLLM_AutoRouterDailySpend
+}
+
+// Durable accounting ledger. One row per usage or cost event; payload holds the
+// private event envelope versioned by schema_version. Usage rows retain request-scoped
+// pricing context; cost rows include resolved rates and the finalized pricing snapshot.
+model LiteLLM_AccountingEvent {
+  event_id        String              @id
+  kind            AccountingEventKind
+  source_event_id String?
+  schema_version  Int
+  payload_hash    String
+  occurred_at     DateTime
+  recorded_at     DateTime            @default(now())
+  payload         Json
+
+  @@index([recorded_at])
+  @@index([kind, source_event_id])
+}
+
+// One row per (event, target table, version): pending work for the projector and,
+// once applied in the same transaction as the target write, the receipt that stops a retry.
+model LiteLLM_AccountingDelivery {
+  event_id        String
+  target          AccountingDeliveryTarget
+  version         Int
+  state           AccountingDeliveryState  @default(pending)
+  attempts        Int                      @default(0)
+  next_attempt_at DateTime                 @default(now())
+  applied_at      DateTime?
+  last_error      String?
+
+  @@id([event_id, target, version])
+  @@index([target, state, next_attempt_at])
+}

schema.prisma

@@ -2061,3 +2061,73 @@ model LiteLLM_TelemetryReport {
   @@index([window_end])
   @@index([created_at])
 }
+
+enum AccountingEventKind {
+  inbound_usage
+  outbound_usage
+  cost
+}
+
+enum AccountingDeliveryState {
+  pending
+  applied
+  parked
+}
+
+enum AccountingDeliveryTarget {
+  LiteLLM_VerificationToken
+  LiteLLM_UserTable
+  LiteLLM_TeamTable
+  LiteLLM_TeamMembership
+  LiteLLM_OrganizationTable
+  LiteLLM_OrganizationMembership
+  LiteLLM_EndUserTable
+  LiteLLM_AgentsTable
+  LiteLLM_TagTable
+  LiteLLM_ProjectTable
+  LiteLLM_ModelAccessGroupBudgetTable
+  LiteLLM_BudgetTable
+  LiteLLM_BudgetWindowSpend
+  LiteLLM_DailyUserSpend
+  LiteLLM_DailyTeamSpend
+  LiteLLM_DailyOrganizationSpend
+  LiteLLM_DailyEndUserSpend
+  LiteLLM_DailyAgentSpend
+  LiteLLM_DailyTagSpend
+  LiteLLM_DailyModelUsage
+  LiteLLM_DailyToolSpend
+  LiteLLM_AutoRouterDailySpend
+}
+
+// Durable accounting ledger. One row per usage or cost event; payload holds the
+// private event envelope versioned by schema_version. Usage rows retain request-scoped
+// pricing context; cost rows include resolved rates and the finalized pricing snapshot.
+model LiteLLM_AccountingEvent {
+  event_id        String              @id
+  kind            AccountingEventKind
+  source_event_id String?
+  schema_version  Int
+  payload_hash    String
+  occurred_at     DateTime
+  recorded_at     DateTime            @default(now())
+  payload         Json
+
+  @@index([recorded_at])
+  @@index([kind, source_event_id])
+}
+
+// One row per (event, target table, version): pending work for the projector and,
+// once applied in the same transaction as the target write, the receipt that stops a retry.
+model LiteLLM_AccountingDelivery {
+  event_id        String
+  target          AccountingDeliveryTarget
+  version         Int
+  state           AccountingDeliveryState  @default(pending)
+  attempts        Int                      @default(0)
+  next_attempt_at DateTime                 @default(now())
+  applied_at      DateTime?
+  last_error      String?
+
+  @@id([event_id, target, version])
+  @@index([target, state, next_attempt_at])
+}

tests/integration/database/test_accounting_ledger.py

@@ -0,0 +1,128 @@
+import os
+import subprocess
+import sys
+from collections.abc import Iterator
+from pathlib import Path
+from typing import Final
+
+import psycopg
+import pytest
+from psycopg import sql
+
+from litellm.accounting import DeliveryState, DeliveryTarget, EventKind
+from tests.integration._support.database import scratch_database
+
+_SCHEMA: Final = Path(__file__).resolve().parents[3] / "litellm-proxy-extras/litellm_proxy_extras/schema.prisma"
+_EVENT_INSERT: Final = (
+    'INSERT INTO "LiteLLM_AccountingEvent" '
+    "(event_id, kind, schema_version, payload_hash, occurred_at, payload) "
+    "VALUES ('event', %s, %s, 'hash', '2026-01-01', '{}') RETURNING kind::text, schema_version"
+)
+_DELIVERY_INSERT: Final = (
+    'INSERT INTO "LiteLLM_AccountingDelivery" (event_id, target, version, state, attempts) '
+    "VALUES ('event', %s, %s, %s, %s) RETURNING target::text, version, state::text, attempts"
+)
+
+
+@pytest.fixture(scope="module")
+def ledger_database() -> Iterator[str]:
+    with scratch_database("accounting") as database_url:
+        result: Final = subprocess.run(
+            [sys.executable, "-I", "-m", "prisma", "migrate", "deploy", "--schema", str(_SCHEMA)],
+            env={**os.environ, "DATABASE_URL": database_url},
+            capture_output=True,
+            text=True,
+            timeout=300,
+        )
+        assert result.returncode == 0, result.stdout + result.stderr
+        yield database_url
+
+
+@pytest.fixture
+def ledger(ledger_database: str) -> Iterator[psycopg.Connection[tuple[object, ...]]]:
+    with psycopg.connect(ledger_database) as connection:
+        try:
+            yield connection
+        finally:
+            connection.rollback()
+
+
+@pytest.mark.parametrize("kind", tuple(EventKind))
+def test_ledger_accepts_every_public_event_kind(
+    ledger: psycopg.Connection[tuple[object, ...]], kind: EventKind
+) -> None:
+    row: Final = ledger.execute(_EVENT_INSERT, (kind.value, 1)).fetchone()
+    assert row == (kind.value, 1)
+
+
+@pytest.mark.parametrize("target", tuple(DeliveryTarget))
+@pytest.mark.parametrize("state", tuple(DeliveryState))
+def test_delivery_accepts_every_public_target_and_state(
+    ledger: psycopg.Connection[tuple[object, ...]], target: DeliveryTarget, state: DeliveryState
+) -> None:
+    row: Final = ledger.execute(_DELIVERY_INSERT, (target.value, 1, state.value, 0)).fetchone()
+    assert row == (target.value, 1, state.value, 0)
+
+
+@pytest.mark.parametrize(
+    ("enum_name", "values"),
+    [
+        ("AccountingEventKind", tuple(value.value for value in EventKind)),
+        ("AccountingDeliveryState", tuple(value.value for value in DeliveryState)),
+        ("AccountingDeliveryTarget", tuple(value.value for value in DeliveryTarget)),
+    ],
+)
+def test_migrated_enums_match_the_public_contract(
+    ledger: psycopg.Connection[tuple[object, ...]], enum_name: str, values: tuple[str, ...]
+) -> None:
+    rows: Final = ledger.execute(
+        "SELECT e.enumlabel FROM pg_enum e JOIN pg_type t ON t.oid = e.enumtypid WHERE t.typname = %s",
+        (enum_name,),
+    ).fetchall()
+    assert frozenset(row[0] for row in rows) == frozenset(values)
+
+
+@pytest.mark.parametrize("version", [0, -1])
+def test_event_rejects_nonpositive_schema_versions(
+    ledger: psycopg.Connection[tuple[object, ...]], version: int
+) -> None:
+    with pytest.raises(psycopg.errors.CheckViolation, match="AccountingEvent_schema_version_check"):
+        ledger.execute(_EVENT_INSERT, (EventKind.OUTBOUND_USAGE.value, version))
+
+
+@pytest.mark.parametrize(
+    ("version", "attempts", "constraint"),
+    [
+        (0, 0, "AccountingDelivery_version_check"),
+        (-1, 0, "AccountingDelivery_version_check"),
+        (1, -1, "AccountingDelivery_attempts_check"),
+    ],
+)
+def test_delivery_rejects_invalid_versions_and_attempts(
+    ledger: psycopg.Connection[tuple[object, ...]], version: int, attempts: int, constraint: str
+) -> None:
+    with pytest.raises(psycopg.errors.CheckViolation, match=constraint):
+        ledger.execute(_DELIVERY_INSERT, (DeliveryTarget.KEY_SPEND.value, version, "pending", attempts))
+
+
+@pytest.mark.parametrize("column", ["state", "target"])
+def test_delivery_rejects_unknown_enum_values(ledger: psycopg.Connection[tuple[object, ...]], column: str) -> None:
+    ledger.execute(_DELIVERY_INSERT, (DeliveryTarget.KEY_SPEND.value, 1, "pending", 0))
+    with pytest.raises(psycopg.errors.InvalidTextRepresentation, match="invalid input value for enum"):
+        ledger.execute(
+            sql.SQL('UPDATE "LiteLLM_AccountingDelivery" SET {} = %s').format(sql.Identifier(column)), ("invalid",)
+        )
+
+
+def test_event_rejects_unknown_kind(ledger: psycopg.Connection[tuple[object, ...]]) -> None:
+    with pytest.raises(psycopg.errors.InvalidTextRepresentation, match="invalid input value for enum"):
+        ledger.execute(_EVENT_INSERT, ("invalid", 1))
+
+
+def test_new_delivery_defaults_to_pending_with_no_attempts(ledger: psycopg.Connection[tuple[object, ...]]) -> None:
+    row: Final = ledger.execute(
+        'INSERT INTO "LiteLLM_AccountingDelivery" (event_id, target, version) '
+        "VALUES ('event', %s, 1) RETURNING state::text, attempts",
+        (DeliveryTarget.KEY_SPEND.value,),
+    ).fetchone()
+    assert row == ("pending", 0)

tests/integration/providers/test_anthropic_context_window_stop_wire.py

@@ -0,0 +1,122 @@
+import json
+import uuid
+from collections.abc import Mapping
+from typing import Final
+
+import pytest
+from integration._support.client import Gateway
+from integration._support.wire import Reply, Request, wire_server
+from pydantic import JsonValue
+
+MODEL: Final = "claude-haiku-4-5"
+
+
+def _sse(event: Mapping[str, JsonValue]) -> bytes:
+    return f"event: {event['type']}\ndata: {json.dumps(event)}\n\n".encode()
+
+
+def _message(identity: str, stop_reason: str | None) -> dict[str, JsonValue]:
+    return {
+        "id": "msg_" + identity,
+        "type": "message",
+        "role": "assistant",
+        "model": MODEL,
+        "content": [{"type": "text", "text": "alpha bravo"}],
+        "stop_reason": stop_reason,
+        "stop_sequence": None,
+        "usage": {"input_tokens": 199878, "output_tokens": 12799},
+    }
+
+
+def _stream(identity: str, stop_reason: str) -> tuple[bytes, ...]:
+    return (
+        _sse({"type": "message_start", "message": {**_message(identity, None), "content": []}}),
+        _sse({"type": "content_block_start", "index": 0, "content_block": {"type": "text", "text": ""}}),
+        _sse({"type": "content_block_delta", "index": 0, "delta": {"type": "text_delta", "text": "alpha bravo"}}),
+        _sse({"type": "content_block_stop", "index": 0}),
+        _sse(
+            {
+                "type": "message_delta",
+                "delta": {"stop_reason": stop_reason, "stop_sequence": None},
+                "usage": {"output_tokens": 12799},
+            }
+        ),
+        _sse({"type": "message_stop"}),
+    )
+
+
+def _anthropic(identity: str, stop_reason: str):
+    def respond(request: Request) -> Reply:
+        assert request.method == "POST" and request.target == "/v1/messages", request.target
+        if json.loads(request.body).get("stream"):
+            return Reply(content_type="text/event-stream", chunks=_stream(identity, stop_reason))
+        return Reply(body=json.dumps(_message(identity, stop_reason)).encode())
+
+    return respond
+
+
+def _events(gateway: Gateway, path: str, body: Mapping[str, JsonValue]) -> tuple[dict[str, JsonValue], ...]:
+    with gateway.client.stream(
+        "POST", path, json={**body, "stream": True}, headers={"Authorization": f"Bearer {gateway.key}"}
+    ) as response:
+        assert response.status_code == 200, response.read().decode()
+        return tuple(
+            json.loads(line.removeprefix("data: "))
+            for line in response.iter_lines()
+            if line.startswith("data: ") and line != "data: [DONE]"
+        )
+
+
+def _chat_finish_reason(gateway: Gateway, model: str, *, stream: bool) -> JsonValue:
+    body: Final = {"model": model, "max_tokens": 64000, "messages": [{"role": "user", "content": "copy the list"}]}
+    if stream:
+        reasons: Final = tuple(
+            choice["finish_reason"]
+            for chunk in _events(gateway, "/v1/chat/completions", body)
+            for choice in chunk["choices"]  # comprehension-ok: chunk -> choices flatten
+            if choice.get("finish_reason")
+        )
+        assert len(reasons) == 1, reasons
+        return reasons[0]
+    response: Final = gateway.request("POST", "/v1/chat/completions", body)
+    assert response.status_code == 200, response.text
+    return response.json()["choices"][0]["finish_reason"]
+
+
+def _responses_status(gateway: Gateway, model: str, *, stream: bool) -> JsonValue:
+    body: Final = {"model": model, "max_output_tokens": 64000, "input": "copy the list"}
+    if stream:
+        terminal: Final = tuple(
+            event["response"]
+            for event in _events(gateway, "/v1/responses", body)
+            if event["type"] in ("response.completed", "response.incomplete", "response.failed")
+        )
+        assert len(terminal) == 1, terminal
+        return terminal[0]["status"]
+    response: Final = gateway.request("POST", "/v1/responses", body)
+    assert response.status_code == 200, response.text
+    return response.json()["status"]
+
+
+@pytest.mark.parametrize("stream", [False, True], ids=["json", "stream"])
+@pytest.mark.parametrize(
+    ("stop_reason", "finish_reason"),
+    [("model_context_window_exceeded", "length"), ("max_tokens", "length"), ("end_turn", "stop")],
+)
+def test_chat_maps_anthropic_stop_reason(gateway: Gateway, stop_reason: str, finish_reason: str, stream: bool) -> None:
+    identity: Final = uuid.uuid4().hex
+    with wire_server(_anthropic(identity, stop_reason)) as wire, gateway.scenario() as scenario:
+        model: Final = scenario.model(model=f"anthropic/{MODEL}", api_base=wire.url, api_key="scripted-anthropic-key")
+        assert _chat_finish_reason(gateway, model, stream=stream) == finish_reason
+
+
+@pytest.mark.parametrize("stream", [False, True], ids=["json", "stream"])
+@pytest.mark.parametrize(
+    ("stop_reason", "status"),
+    [("model_context_window_exceeded", "incomplete"), ("max_tokens", "incomplete"), ("end_turn", "completed")],
+)
+def test_responses_maps_anthropic_stop_reason(gateway: Gateway, stop_reason: str, status: str, stream: bool) -> None:
+    identity: Final = uuid.uuid4().hex
+    with wire_server(_anthropic(identity, stop_reason)) as wire, gateway.scenario() as scenario:
+        model: Final = scenario.model(model=f"anthropic/{MODEL}", api_base=wire.url, api_key="scripted-anthropic-key")
+        assert _responses_status(gateway, model, stream=stream) == status

tests/integration/security/test_hide_secrets_tool_traffic.py

@@ -413,3 +413,71 @@ def test_resent_tool_history_is_scanned_once(single_worker: Rig, shape: str) ->
     second: Final = timed(lambda: single_worker.proxy.request("POST", path, following))
     assert first.value.status_code == 200 and second.value.status_code == 200, (first.value.text, second.value.text)
     assert_within_factor_of_control(second.seconds, control=first.seconds, factor=0.5)
+
+
+@pytest.mark.parametrize("placement", ("guardrail", "global"))
+def test_skip_tool_message_in_guardrail_forwards_tool_results(tmp_path: Path, placement: str) -> None:
+    def configure(configuration: dict[str, JsonValue]) -> None:
+        guardrails: Final = configuration["guardrails"]
+        settings: Final = configuration["litellm_settings"]
+        assert isinstance(guardrails, list) and isinstance(guardrails[0], dict) and isinstance(settings, dict)
+        params: Final = guardrails[0]["litellm_params"]
+        assert isinstance(params, dict)
+        (params if placement == "guardrail" else settings)["skip_tool_message_in_guardrail"] = True
+
+    with tool_rig(tmp_path, configure=configure) as rig:
+        rig.wire.drain()
+        replies: Final = (
+            rig.proxy.request(
+                "POST",
+                "/v1/messages",
+                {
+                    "model": "claude",
+                    "max_tokens": 32,
+                    "cache": {"no-cache": True},
+                    "messages": [
+                        {"role": "user", "content": TEXT},
+                        {
+                            "role": "assistant",
+                            "content": [
+                                {"type": "tool_use", "id": "tool_1", "name": "read_file", "input": {"env": TEXT}}
+                            ],
+                        },
+                        {
+                            "role": "user",
+                            "content": [{"type": "tool_result", "tool_use_id": "tool_1", "content": TEXT}],
+                        },
+                    ],
+                },
+            ),
+            rig.proxy.request(
+                "POST",
+                "/v1/chat/completions",
+                {
+                    "model": "gpt",
+                    "cache": {"no-cache": True},
+                    "messages": [
+                        {"role": "user", "content": TEXT},
+                        {
+                            "role": "assistant",
+                            "content": None,
+                            "tool_calls": [
+                                {
+                                    "id": "call_1",
+                                    "type": "function",
+                                    "function": {"name": "read_file", "arguments": json.dumps({"env": TEXT})},
+                                }
+                            ],
+                        },
+                        {"role": "tool", "tool_call_id": "call_1", "content": TEXT},
+                    ],
+                },
+            ),
+        )
+        assert all(reply.status_code == 200 for reply in replies), [reply.text for reply in replies]
+        sent: Final = rig.wire.drain()
+        assert len(sent) == 2
+        for request in sent:
+            messages = json.loads(request.body)["messages"]
+            assert SECRET not in json.dumps(messages[:2])
+            assert SECRET in json.dumps(messages[2])

tests/unit/accounting/test_measurements.py

@@ -0,0 +1,235 @@
+from datetime import datetime, timezone
+from decimal import Decimal
+from typing import Final
+
+import pytest
+
+from litellm.accounting import (
+    CostEvent,
+    Dimension,
+    DimensionName,
+    InboundUsageEvent,
+    Measurement,
+    Metric,
+    MetricName,
+    Outcome,
+    OutboundUsageEvent,
+    PricingSnapshot,
+    measurements_for,
+)
+from litellm.types.utils import CallTypes
+
+_START: Final = datetime(2026, 1, 1, 23, 59, 59, tzinfo=timezone.utc)
+_END: Final = datetime(2026, 1, 2, tzinfo=timezone.utc)
+_DIMENSIONS: Final = (
+    Dimension(name=DimensionName.TEAM, value="team"),
+    Dimension(name=DimensionName.TAG, value="alpha"),
+    Dimension(name=DimensionName.TAG, value="beta"),
+)
+
+
+def _snapshot(amount: Decimal) -> PricingSnapshot:
+    return PricingSnapshot(
+        calculator_version="test",
+        model="test-model",
+        provider=None,
+        pricing_source="upstream_header",
+        priced_model_key="test-model",
+        upstream_reported_cost=amount,
+        total_cost=amount,
+    )
+
+
+@pytest.mark.parametrize(
+    ("outcome", "outcome_metric"),
+    [
+        (Outcome.SUCCEEDED, MetricName.SUCCESSFUL_REQUESTS),
+        (Outcome.FAILED, MetricName.FAILED_REQUESTS),
+        (Outcome.CANCELLED, MetricName.FAILED_REQUESTS),
+    ],
+)
+def test_caller_measurements_keep_both_request_and_outcome_counts(outcome, outcome_metric):
+    event = InboundUsageEvent(
+        event_id="caller-event",
+        request_id="caller",
+        call_type=CallTypes.acompletion,
+        started_at=_START,
+        ended_at=_END,
+        outcome=outcome,
+        dimensions=_DIMENSIONS,
+        metrics=(Metric(name=MetricName.REQUESTS, value=Decimal(1)), Metric(name=outcome_metric, value=Decimal(1))),
+    )
+    assert measurements_for(event) == (
+        Measurement(
+            source_event_id="caller-event",
+            occurred_at=_START,
+            dimensions=_DIMENSIONS,
+            metric=Metric(name=MetricName.REQUESTS, value=Decimal(1)),
+        ),
+        Measurement(
+            source_event_id="caller-event",
+            occurred_at=_START,
+            dimensions=_DIMENSIONS,
+            metric=Metric(name=outcome_metric, value=Decimal(1)),
+        ),
+    )
+
+
+def test_usage_measurements_keep_every_metric_in_the_start_time_bucket():
+    event = OutboundUsageEvent(
+        event_id="usage-event",
+        attempt_id="attempt",
+        request_id="caller",
+        retry_of_attempt_id=None,
+        call_type=CallTypes.acompletion,
+        started_at=_START,
+        ended_at=_END,
+        outcome=Outcome.SUCCEEDED,
+        dimensions=_DIMENSIONS,
+        metrics=(
+            Metric(name=MetricName.PROMPT_TOKENS, value=Decimal(10)),
+            Metric(name=MetricName.COMPLETION_TOKENS, value=Decimal(2)),
+            Metric(name=MetricName.TOTAL_TOKENS, value=Decimal(12)),
+            Metric(name=MetricName.CACHE_READ_INPUT_TOKENS, value=Decimal(3)),
+            Metric(name=MetricName.CACHE_CREATION_INPUT_TOKENS, value=Decimal(1)),
+        ),
+    )
+    assert measurements_for(event) == (
+        Measurement(
+            source_event_id="usage-event",
+            occurred_at=_START,
+            dimensions=_DIMENSIONS,
+            metric=Metric(name=MetricName.PROMPT_TOKENS, value=Decimal(10)),
+        ),
+        Measurement(
+            source_event_id="usage-event",
+            occurred_at=_START,
+            dimensions=_DIMENSIONS,
+            metric=Metric(name=MetricName.COMPLETION_TOKENS, value=Decimal(2)),
+        ),
+        Measurement(
+            source_event_id="usage-event",
+            occurred_at=_START,
+            dimensions=_DIMENSIONS,
+            metric=Metric(name=MetricName.TOTAL_TOKENS, value=Decimal(12)),
+        ),
+        Measurement(
+            source_event_id="usage-event",
+            occurred_at=_START,
+            dimensions=_DIMENSIONS,
+            metric=Metric(name=MetricName.CACHE_READ_INPUT_TOKENS, value=Decimal(3)),
+        ),
+        Measurement(
+            source_event_id="usage-event",
+            occurred_at=_START,
+            dimensions=_DIMENSIONS,
+            metric=Metric(name=MetricName.CACHE_CREATION_INPUT_TOKENS, value=Decimal(1)),
+        ),
+    )
+
+
+@pytest.mark.parametrize("metrics", [None, ()])
+def test_unknown_or_empty_usage_does_not_fabricate_zero_measurements(metrics):
+    event = OutboundUsageEvent(
+        event_id="usage-event",
+        attempt_id="attempt",
+        request_id=None,
+        retry_of_attempt_id=None,
+        call_type=CallTypes.acompletion,
+        started_at=_START,
+        ended_at=_END,
+        outcome=Outcome.SUCCEEDED,
+        dimensions=(),
+        metrics=metrics,
+    )
+    assert measurements_for(event) == ()
+
+
+@pytest.mark.parametrize("amount", [Decimal(0), Decimal("0.1234567890123456789")])
+def test_cost_measurements_preserve_authoritative_amount_identity_and_occurrence(amount):
+    event = CostEvent(
+        event_id="cost-event",
+        usage_event_id="usage-event",
+        occurred_at=_END,
+        dimensions=_DIMENSIONS,
+        charges=(Metric(name=MetricName.SPEND, value=amount, unit="USD"),),
+        snapshot=_snapshot(amount),
+    )
+    assert measurements_for(event) == (
+        Measurement(
+            source_event_id="cost-event",
+            occurred_at=_END,
+            dimensions=_DIMENSIONS,
+            metric=Metric(name=MetricName.SPEND, value=amount, unit="USD"),
+        ),
+    )
+
+
+def test_billed_retry_and_success_contribute_one_caller_request_and_both_charges():
+    caller = InboundUsageEvent(
+        event_id="caller-event",
+        request_id="caller",
+        call_type=CallTypes.acompletion,
+        started_at=_START,
+        ended_at=_END,
+        outcome=Outcome.SUCCEEDED,
+        dimensions=_DIMENSIONS,
+        metrics=(Metric(name=MetricName.REQUESTS, value=Decimal(1)),),
+    )
+    retry = OutboundUsageEvent(
+        event_id="retry-usage",
+        attempt_id="retry",
+        request_id="caller",
+        retry_of_attempt_id=None,
+        call_type=CallTypes.acompletion,
+        started_at=_START,
+        ended_at=_END,
+        outcome=Outcome.FAILED,
+        dimensions=_DIMENSIONS,
+        metrics=(Metric(name=MetricName.TOTAL_TOKENS, value=Decimal(4)),),
+    )
+    success = OutboundUsageEvent(
+        event_id="success-usage",
+        attempt_id="success",
+        request_id="caller",
+        retry_of_attempt_id="retry",
+        call_type=CallTypes.acompletion,
+        started_at=_START,
+        ended_at=_END,
+        outcome=Outcome.SUCCEEDED,
+        dimensions=_DIMENSIONS,
+        metrics=(Metric(name=MetricName.TOTAL_TOKENS, value=Decimal(12)),),
+    )
+    retry_cost = CostEvent(
+        event_id="retry-cost",
+        usage_event_id="retry-usage",
+        occurred_at=_START,
+        dimensions=_DIMENSIONS,
+        charges=(Metric(name=MetricName.SPEND, value=Decimal("0.025"), unit="USD"),),
+        snapshot=_snapshot(Decimal("0.025")),
+    )
+    success_cost = CostEvent(
+        event_id="success-cost",
+        usage_event_id="success-usage",
+        occurred_at=_START,
+        dimensions=_DIMENSIONS,
+        charges=(Metric(name=MetricName.SPEND, value=Decimal("0.125"), unit="USD"),),
+        snapshot=_snapshot(Decimal("0.125")),
+    )
+    measurements = (
+        *measurements_for(caller),
+        *measurements_for(retry),
+        *measurements_for(success),
+        *measurements_for(retry_cost),
+        *measurements_for(success_cost),
+    )
+    assert sum(m.metric.value for m in measurements if m.metric.name is MetricName.REQUESTS) == Decimal(1)
+    assert sum(m.metric.value for m in measurements if m.metric.name is MetricName.TOTAL_TOKENS) == Decimal(16)
+    assert sum(m.metric.value for m in measurements if m.metric.name is MetricName.SPEND) == Decimal("0.150")
+    assert {m.source_event_id for m in measurements} == {
+        "caller-event",
+        "retry-usage",
+        "success-usage",
+        "retry-cost",
+        "success-cost",
+    }

tests/unit/accounting/test_types.py

@@ -0,0 +1,484 @@
+import re
+from collections.abc import Mapping
+from datetime import datetime, timedelta, timezone
+from decimal import Decimal
+from enum import Enum
+from pathlib import Path
+from typing import Final
+
+import pytest
+from pydantic import ValidationError
+
+from litellm.accounting import (
+    AdmissionDecision,
+    AggregateQuery,
+    AggregateRow,
+    CostEvent,
+    DeliveryState,
+    DeliveryTarget,
+    Dimension,
+    DimensionFilter,
+    DimensionName,
+    EventKind,
+    Granularity,
+    InboundUsageEvent,
+    Metric,
+    MetricName,
+    Multiplier,
+    Outcome,
+    OutboundUsageEvent,
+    PricedQuantity,
+    PricingSnapshot,
+    Rate,
+    validate_aggregate_rows,
+)
+from litellm.accounting.types import PricingUnit
+from litellm.types.utils import CallTypes
+
+_START: Final = datetime(2026, 1, 1, tzinfo=timezone.utc)
+_END: Final = datetime(2026, 1, 2, tzinfo=timezone.utc)
+_REQUEST: Final = Metric(name=MetricName.REQUESTS, value=Decimal(1))
+_TOKEN: Final = Metric(name=MetricName.PROMPT_TOKENS, value=Decimal(10))
+_SPEND: Final = Metric(name=MetricName.SPEND, value=Decimal("0.1234567890123456789"), unit="USD")
+_TEAM: Final = Dimension(name=DimensionName.TEAM, value="team")
+
+
+def _inbound(metrics=(_REQUEST,), outcome=Outcome.SUCCEEDED):
+    return InboundUsageEvent(
+        event_id="inbound:request",
+        request_id="request",
+        call_type=CallTypes.acompletion,
+        started_at=_START,
+        ended_at=_START,
+        outcome=outcome,
+        dimensions=(_TEAM,),
+        metrics=metrics,
+    )
+
+
+def _outbound(metrics=(_TOKEN,), **changes):
+    return OutboundUsageEvent.model_validate(
+        {
+            "event_id": "outbound:attempt",
+            "attempt_id": "attempt",
+            "request_id": "request",
+            "retry_of_attempt_id": None,
+            "call_type": CallTypes.acompletion,
+            "started_at": _START,
+            "ended_at": _END,
+            "outcome": Outcome.SUCCEEDED,
+            "dimensions": (_TEAM,),
+            "metrics": metrics,
+            **changes,
+        }
+    )
+
+
+def _query(**changes):
+    return AggregateQuery.model_validate(
+        {
+            "metric": MetricName.SPEND,
+            "filters": (DimensionFilter(name=DimensionName.TEAM, values=("one", "two")),),
+            "group_by": (DimensionName.MODEL,),
+            "start": _START,
+            "end": _END,
+            "granularity": Granularity.DAY,
+            **changes,
+        }
+    )
+
+
+def test_wire_values_preserve_decimal_precision_and_normalize_aware_times():
+    offset = timezone(timedelta(hours=-8))
+    event = CostEvent(
+        event_id="cost:attempt",
+        usage_event_id="outbound:attempt",
+        occurred_at=datetime(2025, 12, 31, 16, tzinfo=offset),
+        dimensions=(_TEAM,),
+        charges=(_SPEND,),
+        snapshot=_snapshot(),
+    )
+    wire = event.model_dump(mode="json")
+    assert wire["occurred_at"] == "2026-01-01T00:00:00Z"
+    assert wire["charges"] == [{"name": "spend", "value": "0.1234567890123456789", "unit": "USD"}]
+    assert event.occurred_at == _START
+    assert CostEvent.model_validate_json(event.model_dump_json()).charges[0].value == _SPEND.value
+
+
+@pytest.mark.parametrize(
+    ("name", "value", "unit", "message"),
+    [
+        (MetricName.SPEND, Decimal(1), "count", "SPEND requires USD"),
+        (MetricName.REQUESTS, Decimal(1), "USD", "SPEND requires USD"),
+        (MetricName.PROMPT_TOKENS, Decimal("1.5"), "count", "integral"),
+        (MetricName.SPEND, Decimal("NaN"), "USD", "finite"),
+        (MetricName.SPEND, Decimal("Infinity"), "USD", "finite"),
+        (MetricName.SPEND, Decimal(-1), "USD", "greater than or equal"),
+        (MetricName.SPEND, 0.1, "USD", "instance of Decimal"),
+    ],
+)
+def test_metrics_reject_wrong_units_nonfinite_values_and_binary_float_inputs(name, value, unit, message):
+    with pytest.raises(ValidationError, match=message):
+        Metric(name=name, value=value, unit=unit)
+
+
+@pytest.mark.parametrize("metrics", [(_TOKEN,), (_SPEND,), ()])
+def test_inbound_activity_cannot_own_usage_or_cost_or_omit_request_count(metrics):
+    with pytest.raises(ValidationError, match="Inbound events"):
+        _inbound(metrics)
+
+
+@pytest.mark.parametrize(
+    ("outcome", "metric"),
+    [
+        (Outcome.FAILED, Metric(name=MetricName.SUCCESSFUL_REQUESTS, value=Decimal(1))),
+        (Outcome.SUCCEEDED, Metric(name=MetricName.FAILED_REQUESTS, value=Decimal(1))),
+    ],
+)
+def test_inbound_outcome_counts_cannot_contradict_outcome(outcome, metric):
+    with pytest.raises(ValidationError, match="count must match outcome"):
+        _inbound((_REQUEST, metric), outcome)
+
+
+@pytest.mark.parametrize("metrics", [(_REQUEST,), (_SPEND,)])
+def test_outbound_attempts_cannot_own_caller_counts_or_charges(metrics):
+    with pytest.raises(ValidationError, match="Outbound events own only"):
+        _outbound(metrics)
+
+
+@pytest.mark.parametrize(
+    ("changes", "message"),
+    [
+        ({"started_at": datetime(2026, 1, 1)}, "timezone"),
+        ({"ended_at": _START - timedelta(seconds=1)}, "End must follow start"),
+        ({"retry_of_attempt_id": "attempt"}, "cannot retry itself"),
+        ({"dimensions": (_TEAM, _TEAM)}, "Duplicate dimension values"),
+        ({"dimensions": (_TEAM, Dimension(name=DimensionName.TEAM, value="other"))}, "Only tag and model access group"),
+        (
+            {
+                "metrics": (
+                    Metric(name=MetricName.COMPLETION_TOKENS, value=Decimal(5)),
+                    Metric(name=MetricName.REASONING_TOKENS, value=Decimal(6)),
+                )
+            },
+            "reasoning_tokens cannot exceed completion_tokens",
+        ),
+        (
+            {
+                "metrics": (
+                    Metric(name=MetricName.CACHE_CREATION_INPUT_TOKENS, value=Decimal(5)),
+                    Metric(name=MetricName.CACHE_CREATION_INPUT_TOKENS_ABOVE_1HR, value=Decimal(6)),
+                )
+            },
+            "cache_creation_input_tokens_above_1hr cannot exceed cache_creation_input_tokens",
+        ),
+        ({"event_id": " "}, "pattern"),
+        ({"metrics": (_TOKEN, _TOKEN)}, "Duplicate metric names"),
+    ],
+)
+def test_outbound_events_reject_invalid_identity_time_and_duplicate_attribution(changes, message):
+    with pytest.raises(ValidationError, match=message):
+        _outbound(**changes)
+
+
+def test_unknown_and_known_zero_usage_are_distinct_and_values_are_immutable():
+    unknown = _outbound(None)
+    zero = _outbound((Metric(name=MetricName.PROMPT_TOKENS, value=Decimal(0)),))
+    assert unknown.metrics is None
+    assert zero.metrics is not None and zero.metrics[0].value == 0
+    assert _inbound().started_at == _inbound().ended_at
+    with pytest.raises(ValidationError, match="frozen"):
+        zero.attempt_id = "different"
+    with pytest.raises(ValidationError, match="frozen"):
+        zero.metrics[0].value = Decimal(10)
+
+
+@pytest.mark.parametrize("charges", [(), (_TOKEN,), (_SPEND, _SPEND)])
+def test_cost_events_require_one_authoritative_spend(charges):
+    with pytest.raises(ValidationError, match=r"authoritative USD spend|Duplicate metric"):
+        CostEvent(
+            event_id="cost",
+            usage_event_id="usage",
+            occurred_at=_START,
+            dimensions=(),
+            charges=charges,
+            snapshot=_snapshot(),
+        )
+
+
+def test_cost_event_cannot_reuse_its_usage_identity():
+    with pytest.raises(ValidationError, match="distinct identities"):
+        CostEvent(
+            event_id="usage",
+            usage_event_id="usage",
+            occurred_at=_START,
+            dimensions=(),
+            charges=(_SPEND,),
+            snapshot=_snapshot(),
+        )
+
+
+@pytest.mark.parametrize("granularity", [Granularity.DAY, None])
+def test_daily_and_whole_interval_queries_preserve_filter_and_group_semantics(granularity):
+    query = _query(granularity=granularity)
+    assert query.granularity is granularity
+    assert query.filters[0].values == ("one", "two")
+    assert query.group_by == (DimensionName.MODEL,)
+    assert query.start == _START and query.end == _END
+
+
+@pytest.mark.parametrize(
+    ("changes", "message"),
+    [
+        ({"end": _START}, "End must follow start"),
+        ({"start": _START + timedelta(hours=1)}, "UTC day boundaries"),
+        ({"end": _END + timedelta(microseconds=1)}, "UTC day boundaries"),
+        ({"granularity": "hour"}, "day"),
+        ({"group_by": (DimensionName.TEAM, DimensionName.TEAM)}, "Duplicate group-by"),
+        ({"filters": (DimensionFilter(name=DimensionName.TEAM, values=("one",)),) * 2}, "one IN filter"),
+        ({"filters": ({"name": "team", "values": ()},)}, "at least 1"),
+    ],
+)
+def test_queries_reject_unsupported_or_ambiguous_ranges_and_filters(changes, message):
+    with pytest.raises(ValidationError, match=message):
+        _query(**changes)
+
+
+def test_rows_allow_adjacent_buckets_and_overlapping_distinct_groups():
+    first = AggregateRow(start=_START, end=_END, dimensions=(_TEAM,), value=Decimal(1))
+    second = AggregateRow(start=_END, end=_END + timedelta(days=1), dimensions=(_TEAM,), value=Decimal(2))
+    other = AggregateRow(start=_START, end=_END, dimensions=(), value=Decimal(3))
+    rows = (second, other, first)
+    assert validate_aggregate_rows(rows) is rows
+    assert validate_aggregate_rows(()) == ()
+
+
+def test_rows_reject_overlap_within_a_group_regardless_of_dimension_order():
+    model = Dimension(name=DimensionName.MODEL, value="model")
+    first = AggregateRow(start=_START, end=_END, dimensions=(_TEAM, model), value=Decimal(1))
+    overlapping = AggregateRow(start=_START, end=_END, dimensions=(model, _TEAM), value=Decimal(2))
+    with pytest.raises(ValueError, match="overlap within a dimension group"):
+        validate_aggregate_rows((first, overlapping))
+
+
+def test_rows_reject_empty_intervals():
+    with pytest.raises(ValidationError, match="End must follow start"):
+        AggregateRow(start=_START, end=_START, dimensions=(), value=Decimal(0))
+
+
+def test_admission_decision_preserves_a_denial_reason():
+    decision = AdmissionDecision(allowed=False, reason="Team budget exhausted")
+    assert not decision.allowed
+    assert decision.reason == "Team budget exhausted"
+
+
+def _snapshot(**changes: object) -> PricingSnapshot:
+    return PricingSnapshot.model_validate(
+        {
+            "calculator_version": "1.80.0",
+            "model": "gpt-4o",
+            "provider": "openai",
+            "pricing_source": "cost_map",
+            "priced_model_key": "openai/gpt-4o",
+            "multipliers": (Multiplier(name="data_residency", value="eu", factor=Decimal("1.1")),),
+            "rates": (Rate(measure="input", unit="token", usd_per_unit=Decimal("0.0000025")),),
+            "usage": (PricedQuantity(measure="input", unit="token", value=_TOKEN.value),),
+            "input_cost": Decimal("0.000025"),
+            "total_cost": _SPEND.value,
+            **changes,
+        }
+    )
+
+
+def test_cost_event_carries_a_snapshot_whose_total_matches_the_spend():
+    event = CostEvent(
+        event_id="cost:attempt",
+        usage_event_id="outbound:attempt",
+        occurred_at=_START,
+        dimensions=(_TEAM,),
+        charges=(_SPEND,),
+        snapshot=_snapshot(),
+    )
+    assert event.snapshot is not None
+    assert event.snapshot.total_cost == _SPEND.value
+    assert event.snapshot.multipliers[0].factor == Decimal("1.1")
+    with pytest.raises(ValidationError, match="Snapshot total must equal the authoritative spend"):
+        CostEvent(
+            event_id="cost:attempt",
+            usage_event_id="outbound:attempt",
+            occurred_at=_START,
+            dimensions=(_TEAM,),
+            charges=(_SPEND,),
+            snapshot=_snapshot(total_cost=Decimal("0.5")),
+        )
+
+
+@pytest.mark.parametrize(
+    ("changes", "message"),
+    (
+        (
+            {
+                "rates": (
+                    Rate(measure="input", unit="token", usd_per_unit=Decimal(1)),
+                    Rate(measure="input", unit="token", usd_per_unit=Decimal(2)),
+                )
+            },
+            "Duplicate rate measures and units",
+        ),
+        ({"usage": (_SPEND,)}, "PricedQuantity"),
+        ({"usage": (_REQUEST,)}, "PricedQuantity"),
+        ({"pricing_source": "guess"}, "pricing_source"),
+        ({"multipliers": ({"name": "x", "value": "y", "factor": Decimal(0)},)}, "factor"),
+        ({"schema_version": 2}, "schema_version"),
+        (
+            {"additional_costs": ({"name": "fee", "amount": Decimal(1)}, {"name": "fee", "amount": Decimal(2)})},
+            "Duplicate additional cost names",
+        ),
+    ),
+)
+def test_snapshots_reject_ambiguous_rates_wrong_usage_and_unknown_sources(changes: Mapping[str, object], message: str):
+    with pytest.raises(ValidationError, match=message):
+        _snapshot(**changes)
+
+
+@pytest.mark.parametrize(
+    ("unit", "value"),
+    [("token", "10"), ("second", "1.25"), ("character", "25"), ("image", "2"), ("request", "1")],
+)
+def test_snapshot_preserves_matching_rates_and_quantities_for_each_billing_unit(unit: PricingUnit, value: str) -> None:
+    quantity: Final = PricedQuantity(measure="input", unit=unit, value=Decimal(value))
+    rate: Final = Rate(measure="input", unit=unit, usd_per_unit=Decimal("0.003"))
+    amount: Final = quantity.value * rate.usd_per_unit
+    snapshot: Final = _snapshot(rates=(rate,), usage=(quantity,), total_cost=amount)
+    restored: Final = PricingSnapshot.model_validate_json(snapshot.model_dump_json())
+    assert restored == snapshot
+    assert restored.usage[0].value * restored.rates[0].usd_per_unit == amount
+
+
+@pytest.mark.parametrize(
+    ("usage", "message"),
+    [
+        ((), "Every rate requires"),
+        (({"measure": "output", "unit": "token", "value": Decimal(10)},), "Every rate requires"),
+        (({"measure": "input", "unit": "second", "value": Decimal(10)},), "Every rate requires"),
+        (
+            ({"measure": "input", "unit": "token", "value": Decimal(10)},) * 2,
+            "Duplicate priced quantity",
+        ),
+    ],
+)
+def test_snapshot_rejects_missing_mismatched_or_duplicate_priced_quantities(
+    usage: tuple[Mapping[str, object], ...], message: str
+) -> None:
+    with pytest.raises(ValidationError, match=message):
+        _snapshot(usage=usage)
+
+
+@pytest.mark.parametrize("unit", ["token", "character", "image", "request"])
+def test_priced_counts_remain_integral(unit: PricingUnit) -> None:
+    with pytest.raises(ValidationError, match="integral"):
+        PricedQuantity(measure="input", unit=unit, value=Decimal("1.25"))
+
+
+@pytest.mark.parametrize(
+    ("value", "message"),
+    [
+        (Decimal(-1), "greater than or equal"),
+        (Decimal("NaN"), "finite"),
+        (Decimal("Infinity"), "finite"),
+        (1.25, "instance of Decimal"),
+    ],
+)
+def test_pricing_values_reject_invalid_amounts(value: Decimal | float, message: str) -> None:
+    with pytest.raises(ValidationError, match=message):
+        PricedQuantity.model_validate({"measure": "input", "unit": "second", "value": value})
+    with pytest.raises(ValidationError, match=message):
+        Rate.model_validate({"measure": "input", "unit": "second", "usd_per_unit": value})
+
+
+@pytest.mark.parametrize("snapshot_fields", [{}, {"snapshot": None}])
+def test_cost_event_cannot_omit_its_derivation(snapshot_fields: Mapping[str, object]) -> None:
+    with pytest.raises(ValidationError, match="snapshot"):
+        CostEvent.model_validate(
+            {
+                "event_id": "cost",
+                "usage_event_id": "usage",
+                "occurred_at": _START,
+                "dimensions": (),
+                "charges": (_SPEND,),
+                **snapshot_fields,
+            }
+        )
+
+
+def test_upstream_charge_snapshot_preserves_reported_cost_without_inventing_rates() -> None:
+    snapshot: Final = _snapshot(
+        pricing_source="upstream_header", rates=(), usage=(), upstream_reported_cost=_SPEND.value
+    )
+    restored: Final = PricingSnapshot.model_validate_json(snapshot.model_dump_json())
+    assert restored.upstream_reported_cost == restored.total_cost == _SPEND.value
+    assert restored.rates == restored.usage == ()
+
+
+def test_subset_token_counts_round_trip_without_changing_totals() -> None:
+    metrics: Final = (
+        Metric(name=MetricName.COMPLETION_TOKENS, value=Decimal(10)),
+        Metric(name=MetricName.REASONING_TOKENS, value=Decimal(3)),
+        Metric(name=MetricName.CACHE_CREATION_INPUT_TOKENS, value=Decimal(8)),
+        Metric(name=MetricName.CACHE_CREATION_INPUT_TOKENS_ABOVE_1HR, value=Decimal(2)),
+    )
+    event: Final = _outbound(metrics=metrics)
+    assert OutboundUsageEvent.model_validate_json(event.model_dump_json()).metrics == metrics
+
+
+@pytest.mark.parametrize(
+    "schema_path",
+    [
+        "schema.prisma",
+        "litellm/proxy/schema.prisma",
+        "litellm-proxy-extras/litellm_proxy_extras/schema.prisma",
+    ],
+)
+@pytest.mark.parametrize(
+    ("enum_name", "python_enum"),
+    [
+        ("AccountingEventKind", EventKind),
+        ("AccountingDeliveryState", DeliveryState),
+        ("AccountingDeliveryTarget", DeliveryTarget),
+    ],
+)
+def test_prisma_accepts_exactly_the_public_accounting_enum_values(
+    schema_path: str, enum_name: str, python_enum: type[Enum]
+) -> None:
+    schema: Final = (Path(__file__).resolve().parents[3] / schema_path).read_text()
+    declaration: Final = re.search(r"enum " + enum_name + r"\s*\{([^}]+)\}", schema)
+    assert declaration is not None, f"Missing persisted enum {enum_name}"
+    assert frozenset(declaration[1].split()) == frozenset(member.value for member in python_enum)
+
+
+def test_one_event_carries_every_matched_model_access_group():
+    groups = (
+        Dimension(name=DimensionName.MODEL_ACCESS_GROUP, value="gpt-group"),
+        Dimension(name=DimensionName.MODEL_ACCESS_GROUP, value="claude-group"),
+    )
+    assert _outbound(dimensions=(_TEAM, *groups)).dimensions == (_TEAM, *groups)
+    with pytest.raises(ValidationError, match="Duplicate dimension values"):
+        _outbound(dimensions=(groups[0], groups[0]))
+
+
+def test_snapshots_reject_duplicate_multipliers_and_contradictory_token_counts():
+    multiplier = Multiplier(name="data_residency", value="eu", factor=Decimal("1.1"))
+    with pytest.raises(ValidationError, match="Duplicate multiplier names"):
+        _snapshot(multipliers=(multiplier, multiplier))
+    with pytest.raises(ValidationError, match="reasoning_tokens cannot exceed completion_tokens"):
+        _snapshot(
+            usage=(
+                PricedQuantity(measure="input", unit="token", value=_TOKEN.value),
+                PricedQuantity(measure="completion_tokens", unit="token", value=Decimal(5)),
+                PricedQuantity(measure="reasoning_tokens", unit="token", value=Decimal(6)),
+            )
+        )
+
+
+def test_agent_live_spend_is_a_delivery_target():
+    assert DeliveryTarget.AGENT_SPEND.value == "LiteLLM_AgentsTable"

tests/unit/litellm_core_utils/test_core_helpers.py

@@ -256,6 +256,7 @@ class TestMapFinishReasonAnthropic:
             ("max_tokens", "length"),
             ("tool_use", "tool_calls"),
             ("compaction", "length"),
+            ("model_context_window_exceeded", "length"),
             ("content_filtered", "content_filter"),
         ],
     )

tests/unit/proxy/guardrails/test_content_utils.py

@@ -2,6 +2,7 @@
 
 import json
 from functools import reduce
+from typing import Final
 
 import pytest
 from pydantic import JsonValue
@@ -924,6 +925,34 @@ def test_native_writer_visits_each_tool_and_system_slot(data):
     assert "payload" not in json.dumps(data)
 
 
+def test_skip_tool_message_leaves_tool_results_and_masks_the_rest() -> None:
+    data: Final[dict[str, object]] = {
+        "messages": [
+            {"role": "user", "content": "secret"},
+            {
+                "role": "assistant",
+                "content": [{"type": "tool_use", "id": "tool", "name": "read", "input": {"env": "secret"}}],
+                "tool_calls": [{"id": "call", "function": {"name": "read", "arguments": '{"env": "secret"}'}}],
+            },
+            {"role": "user", "content": [{"type": "tool_result", "tool_use_id": "tool", "content": "secret"}]},
+            {"role": "tool", "tool_call_id": "call", "content": "secret"},
+        ]
+    }
+    assert walk_user_text(data, lambda text: text.replace("secret", "[MASKED]"), skip_tool_message=True) == 3
+    assert data == {
+        "messages": [
+            {"role": "user", "content": "[MASKED]"},
+            {
+                "role": "assistant",
+                "content": [{"type": "tool_use", "id": "tool", "name": "read", "input": {"env": "[MASKED]"}}],
+                "tool_calls": [{"id": "call", "function": {"name": "read", "arguments": '{"env": "[MASKED]"}'}}],
+            },
+            {"role": "user", "content": [{"type": "tool_result", "tool_use_id": "tool", "content": "secret"}]},
+            {"role": "tool", "tool_call_id": "call", "content": "secret"},
+        ]
+    }
+
+
 @pytest.mark.parametrize(
     "part",
     (

tests/unit/test_spend_logs_guard.py

@@ -42,6 +42,10 @@ class TestSpendLogHits:
         assert guard.spend_log_hits(diff) == ("litellm/proxy/schema.prisma: a LiteLLM_SpendLogs",)
 
 
+def _review(login: str, state: str, commit_id: str | None = "head") -> "guard.Review":
+    return guard.Review(state=state, user=guard.Reviewer(login=login), commit_id=commit_id)
+
+
 class TestApprovedBy:
     def test_the_latest_review_of_the_approver_decides(self):
         reviews: Final = tuple(
@@ -61,6 +65,42 @@ class TestApprovedBy:
         assert guard.approved_by((), "kerry-berri") is False
 
 
+class TestLatestApproval:
+    def test_the_approvers_last_approved_review_is_returned_with_its_commit(self):
+        reviews: Final = (
+            _review("kerry-berri", "APPROVED", "old"),
+            _review("someone-else", "APPROVED", "head"),
+            _review("kerry-berri", "APPROVED", "new"),
+            _review("kerry-berri", "COMMENTED", "head"),
+        )
+        assert guard.latest_approval(reviews, "kerry-berri") is reviews[2]
+        assert guard.latest_approval(reviews[:2], "kerry-berri") is reviews[0]
+        assert guard.latest_approval(reviews[1:2], "kerry-berri") is None
+        assert guard.latest_approval((), "kerry-berri") is None
+
+
+class TestApprovalCovers:
+    _HITS: Final = ("schema.prisma: a LiteLLM_SpendLogs", "schema.prisma: b LiteLLM_SpendLogs")
+
+    def test_an_approval_on_the_head_covers_it_without_reading_the_approved_commit(self):
+        def unreadable(commit: str) -> None:
+            raise AssertionError(f"read {commit}")
+
+        assert guard.approval_covers(_review("kerry-berri", "APPROVED", "head"), "head", self._HITS, unreadable) is True
+
+    def test_an_older_approval_covers_the_head_only_when_it_saw_every_spend_log_line(self):
+        approval: Final = _review("kerry-berri", "APPROVED", "old")
+        assert guard.approval_covers(approval, "head", self._HITS, lambda commit: self._HITS) is True
+        assert guard.approval_covers(approval, "head", self._HITS[:1], lambda commit: self._HITS) is True
+        assert guard.approval_covers(approval, "head", self._HITS, lambda commit: self._HITS[:1]) is False
+
+    def test_an_approval_whose_commit_cannot_be_read_does_not_cover_the_head(self):
+        gone: Final = _review("kerry-berri", "APPROVED", "gone")
+        assert guard.approval_covers(gone, "head", self._HITS, lambda commit: None) is False
+        without_commit: Final = _review("kerry-berri", "APPROVED", None)
+        assert guard.approval_covers(without_commit, "head", self._HITS, lambda commit: self._HITS) is False
+
+
 class TestExitCode:
     def test_a_pull_request_without_spend_log_lines_passes_without_reviews(self, tmp_path, monkeypatch, capsys):
         monkeypatch.setattr(guard, "added_diff", lambda base, head: _diff("schema.prisma", "  foo String", 1))
@@ -75,5 +115,15 @@ class TestExitCode:
         reviews.write_text('[{"user": {"login": "someone-else"}, "state": "APPROVED"}]')
         assert guard.main(["--base", "a", "--head", "b", "--approver", "kerry-berri", "--reviews", str(reviews)]) == 1
         assert "schema.prisma: x LiteLLM_SpendLogs" in capsys.readouterr().out
-        reviews.write_text('[{"user": {"login": "kerry-berri"}, "state": "APPROVED"}]')
+        reviews.write_text('[{"user": {"login": "kerry-berri"}, "state": "APPROVED", "commit_id": "b"}]')
         assert guard.main(["--base", "a", "--head", "b", "--approver", "kerry-berri", "--reviews", str(reviews)]) == 0
+
+    def test_an_approval_before_a_new_spend_log_line_was_pushed_fails(self, tmp_path, monkeypatch, capsys):
+        monkeypatch.setattr(guard, "added_diff", lambda base, head: _diff("schema.prisma", "  x LiteLLM_SpendLogs", 7))
+        monkeypatch.setattr(guard, "reviewed_hits", lambda base, commit: () if commit == "old" else None, raising=False)
+        reviews: Final = tmp_path / "r.json"
+        reviews.write_text('[{"user": {"login": "kerry-berri"}, "state": "APPROVED", "commit_id": "old"}]')
+        assert guard.main(["--base", "a", "--head", "b", "--approver", "kerry-berri", "--reviews", str(reviews)]) == 1
+        assert "changed after @kerry-berri approved old" in capsys.readouterr().out
+        reviews.write_text('[{"user": {"login": "kerry-berri"}, "state": "APPROVED", "commit_id": "gone"}]')
+        assert guard.main(["--base", "a", "--head", "b", "--approver", "kerry-berri", "--reviews", str(reviews)]) == 1