@homeflare/distilled-litellm 0.2.0 → 0.3.0

package/src/services/email_management.ts

@@ -1,4 +1,4 @@
-// AUTO-GENERATED by scripts/generate.ts from .generated-specs (pinned litellm-openapi @ 1.100.0 — see README). Do not edit.
+// AUTO-GENERATED by scripts/generate.ts from .generated-specs (pinned litellm-openapi @ 1.103.0 — see README). Do not edit.
 import * as S from "@distilled.cloud/core/schema";
 import * as API from "@distilled.cloud/core/api";
 import * as C from "@distilled.cloud/core/category";

package/src/services/embeddings.ts

@@ -1,4 +1,4 @@
-// AUTO-GENERATED by scripts/generate.ts from .generated-specs (pinned litellm-openapi @ 1.100.0 — see README). Do not edit.
+// AUTO-GENERATED by scripts/generate.ts from .generated-specs (pinned litellm-openapi @ 1.103.0 — see README). Do not edit.
 import * as S from "@distilled.cloud/core/schema";
 import * as API from "@distilled.cloud/core/api";
 import * as C from "@distilled.cloud/core/category";

package/src/services/evals.ts

@@ -1,4 +1,4 @@
-// AUTO-GENERATED by scripts/generate.ts from .generated-specs (pinned litellm-openapi @ 1.100.0 — see README). Do not edit.
+// AUTO-GENERATED by scripts/generate.ts from .generated-specs (pinned litellm-openapi @ 1.103.0 — see README). Do not edit.
 import * as S from "@distilled.cloud/core/schema";
 import * as API from "@distilled.cloud/core/api";
 import * as C from "@distilled.cloud/core/category";

package/src/services/experimental.ts

@@ -1,4 +1,4 @@
-// AUTO-GENERATED by scripts/generate.ts from .generated-specs (pinned litellm-openapi @ 1.100.0 — see README). Do not edit.
+// AUTO-GENERATED by scripts/generate.ts from .generated-specs (pinned litellm-openapi @ 1.103.0 — see README). Do not edit.
 import * as S from "@distilled.cloud/core/schema";
 import * as API from "@distilled.cloud/core/api";
 import * as C from "@distilled.cloud/core/category";

package/src/services/fallback_management.ts

@@ -1,4 +1,4 @@
-// AUTO-GENERATED by scripts/generate.ts from .generated-specs (pinned litellm-openapi @ 1.100.0 — see README). Do not edit.
+// AUTO-GENERATED by scripts/generate.ts from .generated-specs (pinned litellm-openapi @ 1.103.0 — see README). Do not edit.
 import * as S from "@distilled.cloud/core/schema";
 import * as API from "@distilled.cloud/core/api";
 import * as C from "@distilled.cloud/core/category";

package/src/services/files.ts

@@ -1,4 +1,4 @@
-// AUTO-GENERATED by scripts/generate.ts from .generated-specs (pinned litellm-openapi @ 1.100.0 — see README). Do not edit.
+// AUTO-GENERATED by scripts/generate.ts from .generated-specs (pinned litellm-openapi @ 1.103.0 — see README). Do not edit.
 import * as S from "@distilled.cloud/core/schema";
 import * as API from "@distilled.cloud/core/api";
 import * as C from "@distilled.cloud/core/category";

package/src/services/fine_tuning.ts

@@ -1,4 +1,4 @@
-// AUTO-GENERATED by scripts/generate.ts from .generated-specs (pinned litellm-openapi @ 1.100.0 — see README). Do not edit.
+// AUTO-GENERATED by scripts/generate.ts from .generated-specs (pinned litellm-openapi @ 1.103.0 — see README). Do not edit.
 import * as S from "@distilled.cloud/core/schema";
 import * as API from "@distilled.cloud/core/api";
 import * as C from "@distilled.cloud/core/category";

package/src/services/gemini_agents.ts

@@ -1,4 +1,4 @@
-// AUTO-GENERATED by scripts/generate.ts from .generated-specs (pinned litellm-openapi @ 1.100.0 — see README). Do not edit.
+// AUTO-GENERATED by scripts/generate.ts from .generated-specs (pinned litellm-openapi @ 1.103.0 — see README). Do not edit.
 import * as S from "@distilled.cloud/core/schema";
 import * as API from "@distilled.cloud/core/api";
 import * as C from "@distilled.cloud/core/category";

package/src/services/google_genai_endpoints.ts

@@ -1,4 +1,4 @@
-// AUTO-GENERATED by scripts/generate.ts from .generated-specs (pinned litellm-openapi @ 1.100.0 — see README). Do not edit.
+// AUTO-GENERATED by scripts/generate.ts from .generated-specs (pinned litellm-openapi @ 1.103.0 — see README). Do not edit.
 import * as S from "@distilled.cloud/core/schema";
 import * as API from "@distilled.cloud/core/api";
 import * as C from "@distilled.cloud/core/category";

package/src/services/guardrails.ts

@@ -1,4 +1,4 @@
-// AUTO-GENERATED by scripts/generate.ts from .generated-specs (pinned litellm-openapi @ 1.100.0 — see README). Do not edit.
+// AUTO-GENERATED by scripts/generate.ts from .generated-specs (pinned litellm-openapi @ 1.103.0 — see README). Do not edit.
 import * as S from "@distilled.cloud/core/schema";
 import * as API from "@distilled.cloud/core/api";
 import * as C from "@distilled.cloud/core/category";
@@ -34,18 +34,24 @@ export type PiiEntityType =
   | "PHONE_NUMBER"
   | "MEDICAL_LICENSE"
   | "URL"
+  | "MAC_ADDRESS"
+  | "UUID"
   | "US_BANK_NUMBER"
   | "US_DRIVER_LICENSE"
   | "US_ITIN"
   | "US_PASSPORT"
   | "US_SSN"
+  | "US_MBI"
+  | "US_NPI"
   | "UK_NHS"
   | "UK_NINO"
   | "UK_PASSPORT"
   | "UK_POSTCODE"
   | "UK_VEHICLE_REGISTRATION"
+  | "UK_DRIVING_LICENCE"
   | "ES_NIF"
   | "ES_NIE"
+  | "ES_PASSPORT"
   | "IT_FISCAL_CODE"
   | "IT_DRIVER_LICENSE"
   | "IT_VAT_CODE"
@@ -63,7 +69,38 @@ export type PiiEntityType =
   | "IN_VEHICLE_REGISTRATION"
   | "IN_VOTER"
   | "IN_PASSPORT"
-  | "FI_PERSONAL_IDENTITY_CODE";
+  | "IN_GSTIN"
+  | "FI_PERSONAL_IDENTITY_CODE"
+  | "DE_TAX_ID"
+  | "DE_TAX_NUMBER"
+  | "DE_VAT_ID"
+  | "DE_PASSPORT"
+  | "DE_ID_CARD"
+  | "DE_FUEHRERSCHEIN"
+  | "DE_SOCIAL_SECURITY"
+  | "DE_HEALTH_INSURANCE"
+  | "DE_LANR"
+  | "DE_BSNR"
+  | "DE_KFZ"
+  | "DE_HANDELSREGISTER"
+  | "DE_PLZ"
+  | "KR_RRN"
+  | "KR_FRN"
+  | "KR_PASSPORT"
+  | "KR_DRIVER_LICENSE"
+  | "KR_BRN"
+  | "CA_SIN"
+  | "SE_PERSONNUMMER"
+  | "SE_ORGANISATIONSNUMMER"
+  | "TH_TNIN"
+  | "TR_NATIONAL_ID"
+  | "TR_LICENSE_PLATE"
+  | "NG_NIN"
+  | "NG_VEHICLE_REGISTRATION"
+  | "PH_TIN"
+  | "PH_UMID"
+  | "PH_PASSPORT"
+  | "ZA_ID_NUMBER";
 export const PiiEntityType = S.String;
 
 export type ApplyGuardrailApplyGuardrailPostRequestEntitiesList = Array<
@@ -612,7 +649,11 @@ export const LitellmParamsOnFlagged = S.String;
 export type LitellmParamsOnSensitiveData = "block" | "route";
 export const LitellmParamsOnSensitiveData = S.String;
 
-export type LitellmParamsOnViolation = "warn" | "end_session";
+export type LitellmParamsOnViolation =
+  | "warn"
+  | "end_session"
+  | "block"
+  | "alert";
 export const LitellmParamsOnViolation = S.String;
 
 export type CiscoAIDefenseRuleEntityTypesList = Array<string>;
@@ -856,6 +897,8 @@ export interface LitellmParams {
   additional_provider_specific_params?: LitellmParamsAdditionalProviderSpecificParamsMap | null;
   /** Custom advisory message template used when on_flagged='inject_system_message'. Must contain a {reason} placeholder. Defaults to a generic advisory message if unset. */
   advisory_system_message?: string | null;
+  /** Agent identity reported to Agent 365 with every tool evaluation. When unset, the caller's key alias is used. */
+  agent_id?: string | null;
   /** Akto account ID for multi-tenant deployments. Env: AKTO_ACCOUNT_ID. Default: '1000000'. */
   akto_account_id?: string | null;
   /** API key for Akto. Env: AKTO_API_KEY. */
@@ -930,10 +973,16 @@ export interface LitellmParams {
   categories?: LitellmParamsCategoriesList | null;
   /** Threshold configuration for Lakera guardrail categories */
   category_thresholds?: LakeraCategoryThresholds | null;
+  /** Inject the Headroom retrieval tool for hashes declared by the compression service. */
+  ccr_retrieval?: boolean;
   /** Inline safeguards for the resource-less InvokeGuardrailChecks API (contentFilter / promptAttack / sensitiveInformation). When set, the guardrail calls InvokeGuardrailChecks instead of ApplyGuardrail and no guardrailIdentifier is required. Mutually exclusive with guardrailIdentifier. */
   checks?: BedrockChecksConfigModel | null;
   /** ApplyGuardrail: batch size, in characters, used to re-send content after AWS has rejected a request as too large. Requests AWS accepts are always sent in a single call, so this has no effect until a rejection happens. Defaults to 25,000; a batch AWS still rejects is bisected automatically, so this value only trades round trips against batch size and cannot fail a request on its own. */
   chunk_budget_chars?: number;
+  /** Client id of the gateway's Entra app registration (a confidential client). Falls back to the AGENT365_CLIENT_ID environment variable. */
+  client_id?: string | null;
+  /** Client secret of the gateway's Entra app registration, used to perform the On-Behalf-Of exchange. Falls back to the AGENT365_CLIENT_SECRET environment variable. */
+  client_secret?: string | null;
   /** Only block or mask when detection confidence >= this value; below threshold, allow or log_only. */
   confidence_threshold?: number;
   /** Additional configuration for the guardrail */
@@ -942,6 +991,8 @@ export interface LitellmParams {
   content_filter_threshold?: number | null;
   /** Enable content moderation to check for harmful content (harassment, hate speech, etc.). */
   content_moderation_check?: boolean | null;
+  /** ApplyGuardrail: when True, post-call scans of a request with no grounding_source / query content parts send the system and developer messages as the grounding source and the latest user message as the query, so the guardrail's contextual grounding policy can score the response. Bedrock bills contextual grounding units for these scans and rejects queries, sources and responses over its contextual grounding length limits, so leave this off for guardrails without a contextual grounding policy. Default False: plain messages are never sent as grounding context. */
+  contextual_grounding_from_messages?: boolean;
   /** Path to Google Cloud credentials JSON file or JSON string */
   credentials?: string | null;
   /** Python-like code containing the apply_guardrail function for custom guardrail logic */
@@ -999,6 +1050,8 @@ export interface LitellmParams {
   include_evidence?: boolean | null;
   /** Include scanner category summaries in responses (sets `plr_scanners` header). */
   include_scanners?: boolean | null;
+  /** When True, the Aim and Cato Networks guardrails send /embeddings `input` to the vendor as user messages. Off by default because embedding input is documents being indexed, not a conversation. */
+  inspect_embeddings?: boolean | null;
   /** Boolean flag to determine if calling a detector server (True) or the FMS Orchestrator (False). Defaults to True. */
   is_detector_server?: boolean | null;
   /** Tag to use for keyword redaction */
@@ -1033,7 +1086,7 @@ export interface LitellmParams {
   on_flagged_action?: string | null;
   /** Action to take when sensitive data is detected. 'block' raises an exception (default behavior). 'route' reroutes the request to the model specified in sensitive_data_route_to_model. */
   on_sensitive_data?: LitellmParamsOnSensitiveData | (string & {}) | null;
-  /** For /v1/realtime sessions: 'warn' speaks the violation message and continues; 'end_session' speaks the message and closes the connection. */
+  /** For /v1/realtime sessions: 'warn' speaks the violation message and continues; 'end_session' speaks the message and closes the connection. For guardrail='mcp_security': 'block' rejects the request; 'alert' only logs a warning. */
   on_violation?: LitellmParamsOnViolation | (string & {}) | null;
   /** When True, the guardrail only scans messages that have not already been scanned earlier in the same session (identified by litellm_session_id / session_id). Message content is hashed per session and cached; only the diff (new or edited messages) is sent to the guardrail provider on follow-up calls. Falls back to a full scan when the request has no session id or the cache is unavailable. Intended for blocking/detection guardrails; not applied when mask_request_content is set. */
   only_scan_new_messages?: boolean | null;
@@ -1098,6 +1151,8 @@ export interface LitellmParams {
   prompt_injections?: boolean | null;
   /** The message the bot speaks aloud when a /v1/realtime guardrail fires. Falls back to violation_message_template if not set. */
   realtime_violation_message?: string | null;
+  /** Application id of the Agent 365 resource the OBO token is minted for. Defaults to the production resource ea9ffc3e-8a23-4a7d-836d-234d7c7565c1; the Test and PreProd environments use a different id. Falls back to the AGENT365_RESOURCE_APP_ID environment variable. */
+  resource_app_id?: string | null;
   /** Ordered allow/deny rules. Patterns use regex for tool names/types and optional regex constraints on tool arguments. */
   rules?: LitellmParamsRulesList | null;
   /** When True, this pre_call or post_call guardrail runs concurrently with other opted-in guardrails of the same hook, after the sequential guardrails have run. Use only for block-only guardrails that inspect and reject; do not enable it for guardrails that modify the request or response (e.g. PII masking or sensitive-data routing), since parallel runs share one snapshot and their mutations would race. */
@@ -1136,6 +1191,8 @@ export interface LitellmParams {
   sticky_session_routing?: boolean | null;
   /** The ID of your Model Armor template */
   template_id?: string | null;
+  /** Entra tenant id used for the On-Behalf-Of token exchange. Falls back to the AGENT365_TENANT_ID environment variable. */
+  tenant_id?: string | null;
   /** Per-request timeout for the guardrail provider API call (seconds). Accepts int, float, or numeric string; coerced to float on load. Each guardrail handler chooses its own default when unset. */
   timeout?: number | null;
   /** Enable tool selection quality check to evaluate quality of tool/function calls. */
@@ -1164,6 +1221,7 @@ export const LitellmParams = /*@__PURE__*/ S.suspend(() =>
       S.NullOr(LitellmParamsAdditionalProviderSpecificParamsMap),
     ),
     advisory_system_message: S.optional(S.NullOr(S.String)),
+    agent_id: S.optional(S.NullOr(S.String)),
     akto_account_id: S.optional(S.NullOr(S.String)),
     akto_api_key: S.optional(S.NullOr(S.String)),
     akto_base_url: S.optional(S.NullOr(S.String)),
@@ -1201,12 +1259,16 @@ export const LitellmParams = /*@__PURE__*/ S.suspend(() =>
     breakdown: S.optional(S.NullOr(S.Boolean)),
     categories: S.optional(S.NullOr(LitellmParamsCategoriesList)),
     category_thresholds: S.optional(S.NullOr(LakeraCategoryThresholds)),
+    ccr_retrieval: S.optional(S.Boolean),
     checks: S.optional(S.NullOr(BedrockChecksConfigModel)),
     chunk_budget_chars: S.optional(S.Number),
+    client_id: S.optional(S.NullOr(S.String)),
+    client_secret: S.optional(S.NullOr(S.String)),
     confidence_threshold: S.optional(S.Number),
     config: S.optional(S.NullOr(LitellmParamsConfigMap)),
     content_filter_threshold: S.optional(S.NullOr(S.Number)),
     content_moderation_check: S.optional(S.NullOr(S.Boolean)),
+    contextual_grounding_from_messages: S.optional(S.Boolean),
     credentials: S.optional(S.NullOr(S.String)),
     custom_code: S.optional(S.NullOr(S.String)),
     deepkeep_firewall_id: S.optional(S.NullOr(S.String)),
@@ -1238,6 +1300,7 @@ export const LitellmParams = /*@__PURE__*/ S.suspend(() =>
     hallucinations_check: S.optional(S.NullOr(S.Boolean)),
     include_evidence: S.optional(S.NullOr(S.Boolean)),
     include_scanners: S.optional(S.NullOr(S.Boolean)),
+    inspect_embeddings: S.optional(S.NullOr(S.Boolean)),
     is_detector_server: S.optional(S.NullOr(S.Boolean)),
     keyword_redaction_tag: S.optional(S.NullOr(S.String)),
     lasso_conversation_id: S.optional(S.NullOr(S.String)),
@@ -1296,6 +1359,7 @@ export const LitellmParams = /*@__PURE__*/ S.suspend(() =>
     prompt_attack_threshold: S.optional(S.NullOr(S.Number)),
     prompt_injections: S.optional(S.NullOr(S.Boolean)),
     realtime_violation_message: S.optional(S.NullOr(S.String)),
+    resource_app_id: S.optional(S.NullOr(S.String)),
     rules: S.optional(S.NullOr(LitellmParamsRulesList)),
     run_in_parallel: S.optional(S.NullOr(S.Boolean)),
     sanitize_error_detail: S.optional(S.NullOr(S.Boolean)),
@@ -1315,6 +1379,7 @@ export const LitellmParams = /*@__PURE__*/ S.suspend(() =>
     skip_unscannable_attachments: S.optional(S.NullOr(S.Boolean)),
     sticky_session_routing: S.optional(S.NullOr(S.Boolean)),
     template_id: S.optional(S.NullOr(S.String)),
+    tenant_id: S.optional(S.NullOr(S.String)),
     timeout: S.optional(S.NullOr(S.Number)),
     tool_selection_quality_check: S.optional(S.NullOr(S.Boolean)),
     tracker_api_base: S.optional(S.NullOr(S.String)),
@@ -1587,6 +1652,30 @@ export const GetGuardrailsUsageDetailGuardrailsUsageDetailGuardrailIdRequest =
       "GetGuardrailsUsageDetailGuardrailsUsageDetailGuardrailIdRequest",
   }) as any as S.Schema<GetGuardrailsUsageDetailGuardrailsUsageDetailGuardrailIdRequest>;
 
+export type UsageDetailResponseCostByKeyMap = {
+  [key: string]: number | null | undefined;
+};
+export const UsageDetailResponseCostByKeyMap = /*@__PURE__*/ S.Record(
+  S.String,
+  S.NullOr(S.Number),
+) as any as S.Schema<UsageDetailResponseCostByKeyMap>;
+
+export type UsageDetailResponseCostByTeamMap = {
+  [key: string]: number | null | undefined;
+};
+export const UsageDetailResponseCostByTeamMap = /*@__PURE__*/ S.Record(
+  S.String,
+  S.NullOr(S.Number),
+) as any as S.Schema<UsageDetailResponseCostByTeamMap>;
+
+export type UsageDetailResponseCostByUnitMap = {
+  [key: string]: number | null | undefined;
+};
+export const UsageDetailResponseCostByUnitMap = /*@__PURE__*/ S.Record(
+  S.String,
+  S.NullOr(S.Number),
+) as any as S.Schema<UsageDetailResponseCostByUnitMap>;
+
 export interface UsageChartPoint {
   blocked: number;
   date: string;
@@ -1609,6 +1698,54 @@ export const UsageDetailResponseTimeSeriesList = /*@__PURE__*/ S.Array(
   UsageChartPoint,
 ) as any as S.Schema<UsageDetailResponseTimeSeriesList>;
 
+export type UsageDetailResponseUntrackedUsageUnitsMap = {
+  [key: string]: number | undefined;
+};
+export const UsageDetailResponseUntrackedUsageUnitsMap = /*@__PURE__*/ S.Record(
+  S.String,
+  S.Number,
+) as any as S.Schema<UsageDetailResponseUntrackedUsageUnitsMap>;
+
+export type UsageDetailResponseUntrackedUsageUnitsByKeyValueMap = {
+  [key: string]: number | undefined;
+};
+export const UsageDetailResponseUntrackedUsageUnitsByKeyValueMap =
+  /*@__PURE__*/ S.Record(
+    S.String,
+    S.Number,
+  ) as any as S.Schema<UsageDetailResponseUntrackedUsageUnitsByKeyValueMap>;
+
+export type UsageDetailResponseUntrackedUsageUnitsByKeyMap = {
+  [key: string]:
+    | UsageDetailResponseUntrackedUsageUnitsByKeyValueMap
+    | undefined;
+};
+export const UsageDetailResponseUntrackedUsageUnitsByKeyMap =
+  /*@__PURE__*/ S.Record(
+    S.String,
+    UsageDetailResponseUntrackedUsageUnitsByKeyValueMap,
+  ) as any as S.Schema<UsageDetailResponseUntrackedUsageUnitsByKeyMap>;
+
+export type UsageDetailResponseUntrackedUsageUnitsByTeamValueMap = {
+  [key: string]: number | undefined;
+};
+export const UsageDetailResponseUntrackedUsageUnitsByTeamValueMap =
+  /*@__PURE__*/ S.Record(
+    S.String,
+    S.Number,
+  ) as any as S.Schema<UsageDetailResponseUntrackedUsageUnitsByTeamValueMap>;
+
+export type UsageDetailResponseUntrackedUsageUnitsByTeamMap = {
+  [key: string]:
+    | UsageDetailResponseUntrackedUsageUnitsByTeamValueMap
+    | undefined;
+};
+export const UsageDetailResponseUntrackedUsageUnitsByTeamMap =
+  /*@__PURE__*/ S.Record(
+    S.String,
+    UsageDetailResponseUntrackedUsageUnitsByTeamValueMap,
+  ) as any as S.Schema<UsageDetailResponseUntrackedUsageUnitsByTeamMap>;
+
 export type UsageDetailResponseUsageUnitsMap = {
   [key: string]: number | undefined;
 };
@@ -1660,11 +1797,13 @@ export const UsageUnitsDailyPointUnitsMap = /*@__PURE__*/ S.Record(
 ) as any as S.Schema<UsageUnitsDailyPointUnitsMap>;
 
 export interface UsageUnitsDailyPoint {
+  cost: number | null;
   date: string;
   units: UsageUnitsDailyPointUnitsMap;
 }
 export const UsageUnitsDailyPoint = /*@__PURE__*/ S.suspend(() =>
   S.Struct({
+    cost: S.NullOr(S.Number),
     date: S.String,
     units: UsageUnitsDailyPointUnitsMap,
   }),
@@ -1681,6 +1820,10 @@ export const UsageDetailResponseUsageUnitsDailyList = /*@__PURE__*/ S.Array(
 export interface UsageDetailResponse {
   avgLatency: number | null;
   avgScore: number | null;
+  cost: number | null;
+  cost_by_key: UsageDetailResponseCostByKeyMap;
+  cost_by_team: UsageDetailResponseCostByTeamMap;
+  cost_by_unit: UsageDetailResponseCostByUnitMap;
   description: string | null;
   failRate: number;
   guardrail_id: string;
@@ -1691,6 +1834,9 @@ export interface UsageDetailResponse {
   time_series: UsageDetailResponseTimeSeriesList;
   trend: string;
   type: string;
+  untracked_usage_units: UsageDetailResponseUntrackedUsageUnitsMap;
+  untracked_usage_units_by_key: UsageDetailResponseUntrackedUsageUnitsByKeyMap;
+  untracked_usage_units_by_team: UsageDetailResponseUntrackedUsageUnitsByTeamMap;
   usage_units: UsageDetailResponseUsageUnitsMap;
   usage_units_by_key: UsageDetailResponseUsageUnitsByKeyMap;
   usage_units_by_team: UsageDetailResponseUsageUnitsByTeamMap;
@@ -1700,6 +1846,10 @@ export const UsageDetailResponse = /*@__PURE__*/ S.suspend(() =>
   S.Struct({
     avgLatency: S.NullOr(S.Number),
     avgScore: S.NullOr(S.Number),
+    cost: S.NullOr(S.Number),
+    cost_by_key: UsageDetailResponseCostByKeyMap,
+    cost_by_team: UsageDetailResponseCostByTeamMap,
+    cost_by_unit: UsageDetailResponseCostByUnitMap,
     description: S.NullOr(S.String),
     failRate: S.Number,
     guardrail_id: S.String,
@@ -1710,6 +1860,11 @@ export const UsageDetailResponse = /*@__PURE__*/ S.suspend(() =>
     time_series: UsageDetailResponseTimeSeriesList,
     trend: S.String,
     type: S.String,
+    untracked_usage_units: UsageDetailResponseUntrackedUsageUnitsMap,
+    untracked_usage_units_by_key:
+      UsageDetailResponseUntrackedUsageUnitsByKeyMap,
+    untracked_usage_units_by_team:
+      UsageDetailResponseUntrackedUsageUnitsByTeamMap,
     usage_units: UsageDetailResponseUsageUnitsMap,
     usage_units_by_key: UsageDetailResponseUsageUnitsByKeyMap,
     usage_units_by_team: UsageDetailResponseUsageUnitsByTeamMap,
@@ -1815,6 +1970,15 @@ export const UsageOverviewResponseChartList = /*@__PURE__*/ S.Array(
   UsageChartPoint,
 ) as any as S.Schema<UsageOverviewResponseChartList>;
 
+/** The share of usageUnits that cost leaves out: units recorded with no known price, per counter */
+export type UsageOverviewRowUntrackedUsageUnitsMap = {
+  [key: string]: number | undefined;
+};
+export const UsageOverviewRowUntrackedUsageUnitsMap = /*@__PURE__*/ S.Record(
+  S.String,
+  S.Number,
+) as any as S.Schema<UsageOverviewRowUntrackedUsageUnitsMap>;
+
 export type UsageOverviewRowUsageUnitsMap = {
   [key: string]: number | undefined;
 };
@@ -1826,6 +1990,8 @@ export const UsageOverviewRowUsageUnitsMap = /*@__PURE__*/ S.Record(
 export interface UsageOverviewRow {
   avgLatency: number | null;
   avgScore: number | null;
+  /** USD for the priced share of usageUnits over the window; null when no unit was priced */
+  cost: number | null;
   failRate: number;
   id: string;
   name: string;
@@ -1834,12 +2000,15 @@ export interface UsageOverviewRow {
   status: string;
   trend: string;
   type: string;
+  /** The share of usageUnits that cost leaves out: units recorded with no known price, per counter */
+  untrackedUsageUnits: UsageOverviewRowUntrackedUsageUnitsMap;
   usageUnits: UsageOverviewRowUsageUnitsMap;
 }
 export const UsageOverviewRow = /*@__PURE__*/ S.suspend(() =>
   S.Struct({
     avgLatency: S.NullOr(S.Number),
     avgScore: S.NullOr(S.Number),
+    cost: S.NullOr(S.Number),
     failRate: S.Number,
     id: S.String,
     name: S.String,
@@ -1848,6 +2017,7 @@ export const UsageOverviewRow = /*@__PURE__*/ S.suspend(() =>
     status: S.String,
     trend: S.String,
     type: S.String,
+    untrackedUsageUnits: UsageOverviewRowUntrackedUsageUnitsMap,
     usageUnits: UsageOverviewRowUsageUnitsMap,
   }),
 ).annotate({
@@ -1859,6 +2029,15 @@ export const UsageOverviewResponseRowsList = /*@__PURE__*/ S.Array(
   UsageOverviewRow,
 ) as any as S.Schema<UsageOverviewResponseRowsList>;
 
+export type UsageOverviewResponseTotalUntrackedUsageUnitsMap = {
+  [key: string]: number | undefined;
+};
+export const UsageOverviewResponseTotalUntrackedUsageUnitsMap =
+  /*@__PURE__*/ S.Record(
+    S.String,
+    S.Number,
+  ) as any as S.Schema<UsageOverviewResponseTotalUntrackedUsageUnitsMap>;
+
 export type UsageOverviewResponseTotalUsageUnitsMap = {
   [key: string]: number | undefined;
 };
@@ -1872,7 +2051,9 @@ export interface UsageOverviewResponse {
   passRate: number;
   rows: UsageOverviewResponseRowsList;
   totalBlocked: number;
+  totalCost: number | null;
   totalRequests: number;
+  totalUntrackedUsageUnits: UsageOverviewResponseTotalUntrackedUsageUnitsMap;
   totalUsageUnits: UsageOverviewResponseTotalUsageUnitsMap;
 }
 export const UsageOverviewResponse = /*@__PURE__*/ S.suspend(() =>
@@ -1881,7 +2062,9 @@ export const UsageOverviewResponse = /*@__PURE__*/ S.suspend(() =>
     passRate: S.Number,
     rows: UsageOverviewResponseRowsList,
     totalBlocked: S.Number,
+    totalCost: S.NullOr(S.Number),
     totalRequests: S.Number,
+    totalUntrackedUsageUnits: UsageOverviewResponseTotalUntrackedUsageUnitsMap,
     totalUsageUnits: UsageOverviewResponseTotalUsageUnitsMap,
   }),
 ).annotate({
@@ -2045,7 +2228,11 @@ export const BaseLitellmParamsExtraHeadersList = /*@__PURE__*/ S.Array(
 export type BaseLitellmParamsOnSensitiveData = "block" | "route";
 export const BaseLitellmParamsOnSensitiveData = S.String;
 
-export type BaseLitellmParamsOnViolation = "warn" | "end_session";
+export type BaseLitellmParamsOnViolation =
+  | "warn"
+  | "end_session"
+  | "block"
+  | "alert";
 export const BaseLitellmParamsOnViolation = S.String;
 
 export type BaseLitellmParamsPatternsList = Array<ContentFilterPattern>;
@@ -2053,7 +2240,7 @@ export const BaseLitellmParamsPatternsList = /*@__PURE__*/ S.Array(
   ContentFilterPattern,
 ) as any as S.Schema<BaseLitellmParamsPatternsList>;
 
-/** Behavior when a guardrail endpoint is unreachable due to network errors. Implemented by guardrail='generic_guardrail_api', 'akto', 'vigil_guard', 'repelloai', 'headroom', and 'compresr'. 'fail_closed' raises an error (default). 'fail_open' logs a critical error and allows the request to proceed. */
+/** Behavior when a guardrail endpoint is unreachable due to network errors. Implemented by guardrail='generic_guardrail_api', 'agent_365', 'akto', 'vigil_guard', 'repelloai', 'headroom', 'compresr', and 'typesafe'. 'fail_closed' raises an error (default). 'fail_open' logs a critical error and allows the request to proceed. */
 export type BaseLitellmParamsUnreachableFallback = "fail_closed" | "fail_open";
 export const BaseLitellmParamsUnreachableFallback = S.String;
 
@@ -2092,6 +2279,8 @@ export interface BaseLitellmParams {
   fail_on_error?: boolean | null;
   /** Name of the guardrail in guardrails.ai */
   guard_name?: string | null;
+  /** When True, the Aim and Cato Networks guardrails send /embeddings `input` to the vendor as user messages. Off by default because embedding input is documents being indexed, not a conversation. */
+  inspect_embeddings?: boolean | null;
   /** Tag to use for keyword redaction */
   keyword_redaction_tag?: string | null;
   /** Google Cloud location/region (e.g., us-central1) */
@@ -2104,7 +2293,7 @@ export interface BaseLitellmParams {
   model?: string | null;
   /** Action to take when sensitive data is detected. 'block' raises an exception (default behavior). 'route' reroutes the request to the model specified in sensitive_data_route_to_model. */
   on_sensitive_data?: BaseLitellmParamsOnSensitiveData | (string & {}) | null;
-  /** For /v1/realtime sessions: 'warn' speaks the violation message and continues; 'end_session' speaks the message and closes the connection. */
+  /** For /v1/realtime sessions: 'warn' speaks the violation message and continues; 'end_session' speaks the message and closes the connection. For guardrail='mcp_security': 'block' rejects the request; 'alert' only logs a warning. */
   on_violation?: BaseLitellmParamsOnViolation | (string & {}) | null;
   /** When True, the guardrail only scans messages that have not already been scanned earlier in the same session (identified by litellm_session_id / session_id). Message content is hashed per session and cached; only the diff (new or edited messages) is sent to the guardrail provider on follow-up calls. Falls back to a full scan when the request has no session id or the cache is unavailable. Intended for blocking/detection guardrails; not applied when mask_request_content is set. */
   only_scan_new_messages?: boolean | null;
@@ -2142,7 +2331,7 @@ export interface BaseLitellmParams {
   template_id?: string | null;
   /** Per-request timeout for the guardrail provider API call (seconds). Accepts int, float, or numeric string; coerced to float on load. Each guardrail handler chooses its own default when unset. */
   timeout?: number | null;
-  /** Behavior when a guardrail endpoint is unreachable due to network errors. Implemented by guardrail='generic_guardrail_api', 'akto', 'vigil_guard', 'repelloai', 'headroom', and 'compresr'. 'fail_closed' raises an error (default). 'fail_open' logs a critical error and allows the request to proceed. */
+  /** Behavior when a guardrail endpoint is unreachable due to network errors. Implemented by guardrail='generic_guardrail_api', 'agent_365', 'akto', 'vigil_guard', 'repelloai', 'headroom', 'compresr', and 'typesafe'. 'fail_closed' raises an error (default). 'fail_open' logs a critical error and allows the request to proceed. */
   unreachable_fallback?: BaseLitellmParamsUnreachableFallback | (string & {});
   /** Custom message when a guardrail blocks an action. Supports placeholders like {tool_name}, {rule_id}, and {default_message}. */
   violation_message_template?: string | null;
@@ -2170,6 +2359,7 @@ export const BaseLitellmParams = /*@__PURE__*/ S.suspend(() =>
     extra_headers: S.optional(S.NullOr(BaseLitellmParamsExtraHeadersList)),
     fail_on_error: S.optional(S.NullOr(S.Boolean)),
     guard_name: S.optional(S.NullOr(S.String)),
+    inspect_embeddings: S.optional(S.NullOr(S.Boolean)),
     keyword_redaction_tag: S.optional(S.NullOr(S.String)),
     location: S.optional(S.NullOr(S.String)),
     mask_request_content: S.optional(S.NullOr(S.Boolean)),

package/src/services/health.ts

@@ -1,4 +1,4 @@
-// AUTO-GENERATED by scripts/generate.ts from .generated-specs (pinned litellm-openapi @ 1.100.0 — see README). Do not edit.
+// AUTO-GENERATED by scripts/generate.ts from .generated-specs (pinned litellm-openapi @ 1.103.0 — see README). Do not edit.
 import * as S from "@distilled.cloud/core/schema";
 import * as API from "@distilled.cloud/core/api";
 import * as C from "@distilled.cloud/core/category";
@@ -287,6 +287,7 @@ export type GetHealthServicesEndpointHealthServiceRequestServiceCase0 =
   | "arize"
   | "galileo"
   | "newrelic"
+  | "pointfive"
   | "sqs";
 export const GetHealthServicesEndpointHealthServiceRequestServiceCase0 =
   S.String;
@@ -698,7 +699,7 @@ export const getTestEndpointTest: API.OperationMethod<
 export type PostTestModelConnectionHealthTestConnectionError =
   | UnprocessableEntity
   | LitellmOpError;
-/** Test Model Connection Test a direct connection to a specific model. This endpoint allows you to verify if your proxy can successfully connect to a specific model. It's useful for troubleshooting model connectivity issues without going through the full proxy routing. Example: ```bash # If model is configured in proxy_config.yaml, you only need to specify the model name: curl -X POST 'http://localhost:4000/health/test_connection' \ -H 'Authorization: Bearer sk-1234' \ -H 'Content-Type: application/json' \ -d '{ "litellm_params": { "model": "gpt-4o" }, "mode": "chat" }' # The endpoint will automatically use api_key, api_base, etc. from proxy_config.yaml # You can also override specific params or test with custom credentials: curl -X POST 'http://localhost:4000/health/test_connection' \ -H 'Authorization: Bearer sk-1234' \ -H 'Content-Type: application/json' \ -d '{ "litellm_params": { "model": "azure/gpt-4o", "api_key": "os.environ/AZURE_OPENAI_API_KEY", "api_base": "os.environ/AZURE_OPENAI_ENDPOINT", "api_version": "2024-10-21" }, "mode": "chat" }' ``` Note: - If the model is configured in proxy_config.yaml, credentials (api_key, api_base, etc.) will be automatically loaded from the config (with resolved environment variables). - You can override specific params by including them in the request. - You can use `os.environ/VARIABLE_NAME` syntax to reference environment variables, which will be resolved automatically (same as in proxy_config.yaml). Returns: dict: A dictionary containing the health check result with either success information or error details. */
+/** Test Model Connection Test a direct connection to a specific model. This endpoint allows you to verify if your proxy can successfully connect to a specific model. It's useful for troubleshooting model connectivity issues without going through the full proxy routing. Example: ```bash # If model is configured in proxy_config.yaml, you only need to specify the model name: curl -X POST 'http://localhost:4000/health/test_connection' \ -H 'Authorization: Bearer sk-1234' \ -H 'Content-Type: application/json' \ -d '{ "litellm_params": { "model": "gpt-4o" }, "mode": "chat" }' # The endpoint will automatically use api_key, api_base, etc. from proxy_config.yaml # You can also override specific params or test with custom credentials: curl -X POST 'http://localhost:4000/health/test_connection' \ -H 'Authorization: Bearer sk-1234' \ -H 'Content-Type: application/json' \ -d '{ "litellm_params": { "model": "azure/gpt-4o", "api_key": "os.environ/AZURE_OPENAI_API_KEY", "api_base": "os.environ/AZURE_OPENAI_ENDPOINT", "api_version": "2024-10-21" }, "mode": "chat" }' ``` Note: - If the model is configured in proxy_config.yaml, credentials (api_key, api_base, etc.) will be automatically loaded from the config (with resolved environment variables). - A request naming a stored credential (`litellm_credential_name`) that the configuration does not name is probed with that credential instead, and inherits no credentials from the configuration its model string happened to match. - You can override specific params by including them in the request. - You can use `os.environ/VARIABLE_NAME` syntax to reference environment variables, which will be resolved automatically (same as in proxy_config.yaml). Returns: dict: A dictionary containing the health check result with either success information or error details. */
 export const postTestModelConnectionHealthTestConnection: API.OperationMethod<
   PostTestModelConnectionHealthTestConnectionRequest,
   PostTestModelConnectionHealthTestConnectionResponse,

package/src/services/images.ts

@@ -1,4 +1,4 @@
-// AUTO-GENERATED by scripts/generate.ts from .generated-specs (pinned litellm-openapi @ 1.100.0 — see README). Do not edit.
+// AUTO-GENERATED by scripts/generate.ts from .generated-specs (pinned litellm-openapi @ 1.103.0 — see README). Do not edit.
 import * as S from "@distilled.cloud/core/schema";
 import * as API from "@distilled.cloud/core/api";
 import * as C from "@distilled.cloud/core/category";