@kolisachint/hoocode-ai 0.4.114 → 0.4.116

package/dist/types.js.map

@@ -1 +1 @@
-{"version":3,"file":"types.js","sourceRoot":"","sources":["../src/types.ts"],"names":[],"mappings":"","sourcesContent":["import type { AssistantMessageDiagnostic } from \"./utils/diagnostics.js\";\nimport type { AssistantMessageEventStream } from \"./utils/event-stream.js\";\n\nexport type { AssistantMessageEventStream } from \"./utils/event-stream.js\";\n\nexport type KnownApi =\n\t| \"openai-completions\"\n\t| \"mistral-conversations\"\n\t| \"openai-responses\"\n\t| \"azure-openai-responses\"\n\t| \"openai-codex-responses\"\n\t| \"anthropic-messages\"\n\t| \"bedrock-converse-stream\"\n\t| \"google-generative-ai\"\n\t| \"google-vertex\";\n\nexport type Api = KnownApi | (string & {});\n\nexport type KnownImagesApi = \"openrouter-images\";\n\nexport type ImagesApi = KnownImagesApi | (string & {});\n\nexport type KnownProvider =\n\t| \"amazon-bedrock\"\n\t| \"anthropic\"\n\t| \"google\"\n\t| \"google-vertex\"\n\t| \"openai\"\n\t| \"azure-openai-responses\"\n\t| \"openai-codex\"\n\t| \"deepseek\"\n\t| \"github-copilot\"\n\t| \"xai\"\n\t| \"groq\"\n\t| \"cerebras\"\n\t| \"openrouter\"\n\t| \"vercel-ai-gateway\"\n\t| \"zai\"\n\t| \"mistral\"\n\t| \"minimax\"\n\t| \"minimax-cn\"\n\t| \"moonshotai\"\n\t| \"moonshotai-cn\"\n\t| \"huggingface\"\n\t| \"fireworks\"\n\t| \"together\"\n\t| \"opencode\"\n\t| \"opencode-go\"\n\t| \"kimi-coding\"\n\t| \"cloudflare-workers-ai\"\n\t| \"cloudflare-ai-gateway\"\n\t| \"xiaomi\"\n\t| \"xiaomi-token-plan-cn\"\n\t| \"xiaomi-token-plan-ams\"\n\t| \"xiaomi-token-plan-sgp\"\n\t| \"nvidia\";\nexport type Provider = KnownProvider | string;\n\nexport type KnownImagesProvider = \"openrouter\";\n\nexport type ImagesProvider = KnownImagesProvider | string;\n\nexport type ThinkingLevel = \"minimal\" | \"low\" | \"medium\" | \"high\" | \"xhigh\";\nexport type ModelThinkingLevel = \"off\" | ThinkingLevel;\nexport type ThinkingLevelMap = Partial<Record<ModelThinkingLevel, string | null>>;\n\n/** Token budgets for each thinking level (token-based providers only) */\nexport interface ThinkingBudgets {\n\tminimal?: number;\n\tlow?: number;\n\tmedium?: number;\n\thigh?: number;\n}\n\n// Base options all providers share\nexport type CacheRetention = \"none\" | \"short\" | \"long\";\n\nexport type Transport = \"sse\" | \"websocket\" | \"websocket-cached\" | \"auto\";\n\nexport interface ProviderResponse {\n\tstatus: number;\n\theaders: Record<string, string>;\n}\n\nexport interface StreamOptions {\n\ttemperature?: number;\n\tmaxTokens?: number;\n\tsignal?: AbortSignal;\n\tapiKey?: string;\n\t/**\n\t * Preferred transport for providers that support multiple transports.\n\t * Providers that do not support this option ignore it.\n\t */\n\ttransport?: Transport;\n\t/**\n\t * Prompt cache retention preference. Providers map this to their supported values.\n\t * Default: \"short\".\n\t */\n\tcacheRetention?: CacheRetention;\n\t/**\n\t * Optional session identifier for providers that support session-based caching.\n\t * Providers can use this to enable prompt caching, request routing, or other\n\t * session-aware features. Ignored by providers that don't support it.\n\t */\n\tsessionId?: string;\n\t/**\n\t * Optional callback for inspecting or replacing provider payloads before sending.\n\t * Return undefined to keep the payload unchanged.\n\t */\n\tonPayload?: (payload: unknown, model: Model<Api>) => unknown | undefined | Promise<unknown | undefined>;\n\t/**\n\t * Optional callback invoked after an HTTP response is received and before\n\t * its body stream is consumed.\n\t */\n\tonResponse?: (response: ProviderResponse, model: Model<Api>) => void | Promise<void>;\n\t/**\n\t * Optional custom HTTP headers to include in API requests.\n\t * Merged with provider defaults; can override default headers.\n\t * Not supported by all providers (e.g., AWS Bedrock uses SDK auth).\n\t */\n\theaders?: Record<string, string>;\n\t/**\n\t * HTTP request timeout in milliseconds for providers/SDKs that support it.\n\t * For example, OpenAI and Anthropic SDK clients default to 10 minutes.\n\t */\n\ttimeoutMs?: number;\n\t/**\n\t * Maximum retry attempts for providers/SDKs that support client-side retries.\n\t * For example, OpenAI and Anthropic SDK clients default to 2.\n\t */\n\tmaxRetries?: number;\n\t/**\n\t * Maximum delay in milliseconds to wait for a retry when the server requests a long wait.\n\t * If the server's requested delay exceeds this value, the request fails immediately\n\t * with an error containing the requested delay, allowing higher-level retry logic\n\t * to handle it with user visibility.\n\t * Default: 60000 (60 seconds). Set to 0 to disable the cap.\n\t */\n\tmaxRetryDelayMs?: number;\n\t/**\n\t * Optional metadata to include in API requests.\n\t * Providers extract the fields they understand and ignore the rest.\n\t * For example, Anthropic uses `user_id` for abuse tracking and rate limiting.\n\t */\n\tmetadata?: Record<string, unknown>;\n}\n\nexport type ProviderStreamOptions = StreamOptions & Record<string, unknown>;\n\nexport interface ImagesOptions {\n\tsignal?: AbortSignal;\n\tapiKey?: string;\n\t/**\n\t * Optional callback for inspecting or replacing provider payloads before sending.\n\t * Return undefined to keep the payload unchanged.\n\t */\n\tonPayload?: (payload: unknown, model: ImagesModel<ImagesApi>) => unknown | undefined | Promise<unknown | undefined>;\n\t/**\n\t * Optional callback invoked after an HTTP response is received.\n\t */\n\tonResponse?: (response: ProviderResponse, model: ImagesModel<ImagesApi>) => void | Promise<void>;\n\t/**\n\t * Optional custom HTTP headers to include in API requests.\n\t * Merged with provider defaults; can override default headers.\n\t */\n\theaders?: Record<string, string>;\n\t/**\n\t * HTTP request timeout in milliseconds for providers/SDKs that support it.\n\t */\n\ttimeoutMs?: number;\n\t/**\n\t * Maximum retry attempts for providers/SDKs that support client-side retries.\n\t */\n\tmaxRetries?: number;\n\t/**\n\t * Maximum delay in milliseconds to wait for a retry when the server requests a long wait.\n\t * If the server's requested delay exceeds this value, the request fails immediately\n\t * with an error containing the requested delay, allowing higher-level retry logic\n\t * to handle it with user visibility.\n\t * Default: 60000 (60 seconds). Set to 0 to disable the cap.\n\t */\n\tmaxRetryDelayMs?: number;\n\t/**\n\t * Optional metadata to include in API requests.\n\t * Providers extract the fields they understand and ignore the rest.\n\t */\n\tmetadata?: Record<string, unknown>;\n}\n\nexport type ProviderImagesOptions = ImagesOptions & Record<string, unknown>;\n\n// Unified options with reasoning passed to streamSimple() and completeSimple()\nexport interface SimpleStreamOptions extends StreamOptions {\n\treasoning?: ThinkingLevel;\n\t/** Custom token budgets for thinking levels (token-based providers only) */\n\tthinkingBudgets?: ThinkingBudgets;\n\t/**\n\t * Controls how thinking content is returned by adaptive-thinking models\n\t * (Anthropic Messages API).\n\t * - \"summarized\": thinking blocks contain summarized thinking text.\n\t * - \"omitted\": thinking blocks come back empty (the encrypted signature still\n\t *   travels back for continuity). The model still reasons at the requested\n\t *   effort; skipping the summary stream lowers time-to-first-token, which\n\t *   speeds up tool-use turns. When unset, the provider picks a per-model\n\t *   default (Opus 4.8 defaults to \"omitted\").\n\t */\n\tthinkingDisplay?: \"summarized\" | \"omitted\";\n}\n\n// Generic StreamFunction with typed options.\n//\n// Contract:\n// - Must return an AssistantMessageEventStream.\n// - Once invoked, request/model/runtime failures should be encoded in the\n//   returned stream, not thrown.\n// - Error termination must produce an AssistantMessage with stopReason\n//   \"error\" or \"aborted\" and errorMessage, emitted via the stream protocol.\nexport type StreamFunction<TApi extends Api = Api, TOptions extends StreamOptions = StreamOptions> = (\n\tmodel: Model<TApi>,\n\tcontext: Context,\n\toptions?: TOptions,\n) => AssistantMessageEventStream;\n\nexport type ImagesFunction<TApi extends ImagesApi = ImagesApi, TOptions extends ImagesOptions = ImagesOptions> = (\n\tmodel: ImagesModel<TApi>,\n\tcontext: ImagesContext,\n\toptions?: TOptions,\n) => Promise<AssistantImages>;\n\nexport interface TextSignatureV1 {\n\tv: 1;\n\tid: string;\n\tphase?: \"commentary\" | \"final_answer\";\n}\n\nexport interface TextContent {\n\ttype: \"text\";\n\ttext: string;\n\ttextSignature?: string; // e.g., for OpenAI responses, message metadata (legacy id string or TextSignatureV1 JSON)\n}\n\nexport interface ThinkingContent {\n\ttype: \"thinking\";\n\tthinking: string;\n\tthinkingSignature?: string; // e.g., for OpenAI responses, the reasoning item ID\n\t/** When true, the thinking content was redacted by safety filters. The opaque\n\t *  encrypted payload is stored in `thinkingSignature` so it can be passed back\n\t *  to the API for multi-turn continuity. */\n\tredacted?: boolean;\n}\n\nexport interface ImageContent {\n\ttype: \"image\";\n\tdata: string; // base64 encoded image data\n\tmimeType: string; // e.g., \"image/jpeg\", \"image/png\"\n}\n\nexport interface ToolCall {\n\ttype: \"toolCall\";\n\tid: string;\n\tname: string;\n\targuments: Record<string, any>;\n\tthoughtSignature?: string; // Google-specific: opaque signature for reusing thought context\n}\n\nexport interface Usage {\n\tinput: number;\n\toutput: number;\n\tcacheRead: number;\n\tcacheWrite: number;\n\ttotalTokens: number;\n\tcost: {\n\t\tinput: number;\n\t\toutput: number;\n\t\tcacheRead: number;\n\t\tcacheWrite: number;\n\t\ttotal: number;\n\t};\n}\n\nexport type StopReason = \"stop\" | \"length\" | \"toolUse\" | \"error\" | \"aborted\";\n\nexport interface UserMessage {\n\trole: \"user\";\n\tcontent: string | (TextContent | ImageContent)[];\n\ttimestamp: number; // Unix timestamp in milliseconds\n}\n\nexport interface AssistantMessage {\n\trole: \"assistant\";\n\tcontent: (TextContent | ThinkingContent | ToolCall)[];\n\tapi: Api;\n\tprovider: Provider;\n\tmodel: string;\n\tresponseModel?: string; // Concrete `chunk.model` when different from the requested `model` (e.g. OpenRouter `auto` -> `anthropic/...`)\n\tresponseId?: string; // Provider-specific response/message identifier when the upstream API exposes one\n\tdiagnostics?: AssistantMessageDiagnostic[]; // Redacted provider/runtime diagnostics for failures and recoveries.\n\tusage: Usage;\n\tstopReason: StopReason;\n\terrorMessage?: string;\n\ttimestamp: number; // Unix timestamp in milliseconds\n}\n\nexport interface ToolResultMessage<TDetails = any> {\n\trole: \"toolResult\";\n\ttoolCallId: string;\n\ttoolName: string;\n\tcontent: (TextContent | ImageContent)[]; // Supports text and images\n\tdetails?: TDetails;\n\tisError: boolean;\n\ttimestamp: number; // Unix timestamp in milliseconds\n}\n\nexport type Message = UserMessage | AssistantMessage | ToolResultMessage;\n\nexport type ImagesInputContent = TextContent | ImageContent;\nexport type ImagesOutputContent = TextContent | ImageContent;\n\nexport interface ImagesContext {\n\tinput: ImagesInputContent[];\n}\n\nexport type ImagesStopReason = \"stop\" | \"error\" | \"aborted\";\n\nexport interface AssistantImages {\n\tapi: ImagesApi;\n\tprovider: ImagesProvider;\n\tmodel: string;\n\toutput: ImagesOutputContent[];\n\tresponseId?: string;\n\tusage?: Usage;\n\tstopReason: ImagesStopReason;\n\terrorMessage?: string;\n\ttimestamp: number; // Unix timestamp in milliseconds\n}\n\nimport type { TSchema } from \"typebox\";\n\nexport interface Tool<TParameters extends TSchema = TSchema> {\n\tname: string;\n\tdescription: string;\n\tparameters: TParameters;\n}\n\nexport interface Context {\n\tsystemPrompt?: string;\n\tmessages: Message[];\n\ttools?: Tool[];\n}\n\n/**\n * Event protocol for AssistantMessageEventStream.\n *\n * Streams should emit `start` before partial updates, then terminate with either:\n * - `done` carrying the final successful AssistantMessage, or\n * - `error` carrying the final AssistantMessage with stopReason \"error\" or \"aborted\"\n *   and errorMessage.\n */\nexport type AssistantMessageEvent =\n\t| { type: \"start\"; partial: AssistantMessage }\n\t| { type: \"text_start\"; contentIndex: number; partial: AssistantMessage }\n\t| { type: \"text_delta\"; contentIndex: number; delta: string; partial: AssistantMessage }\n\t| { type: \"text_end\"; contentIndex: number; content: string; partial: AssistantMessage }\n\t| { type: \"thinking_start\"; contentIndex: number; partial: AssistantMessage }\n\t| { type: \"thinking_delta\"; contentIndex: number; delta: string; partial: AssistantMessage }\n\t| { type: \"thinking_end\"; contentIndex: number; content: string; partial: AssistantMessage }\n\t| { type: \"toolcall_start\"; contentIndex: number; partial: AssistantMessage }\n\t| { type: \"toolcall_delta\"; contentIndex: number; delta: string; partial: AssistantMessage }\n\t| { type: \"toolcall_end\"; contentIndex: number; toolCall: ToolCall; partial: AssistantMessage }\n\t| { type: \"done\"; reason: Extract<StopReason, \"stop\" | \"length\" | \"toolUse\">; message: AssistantMessage }\n\t| { type: \"error\"; reason: Extract<StopReason, \"aborted\" | \"error\">; error: AssistantMessage };\n\n/**\n * Compatibility settings for OpenAI-compatible completions APIs.\n * Use this to override URL-based auto-detection for custom providers.\n */\nexport interface OpenAICompletionsCompat {\n\t/** Whether the provider supports the `store` field. Default: auto-detected from URL. */\n\tsupportsStore?: boolean;\n\t/** Whether the provider supports the `developer` role (vs `system`). Default: auto-detected from URL. */\n\tsupportsDeveloperRole?: boolean;\n\t/** Whether the provider supports `reasoning_effort`. Default: auto-detected from URL. */\n\tsupportsReasoningEffort?: boolean;\n\t/** Whether the provider supports `stream_options: { include_usage: true }` for token usage in streaming responses. Default: true. */\n\tsupportsUsageInStreaming?: boolean;\n\t/** Which field to use for max tokens. Default: auto-detected from URL. */\n\tmaxTokensField?: \"max_completion_tokens\" | \"max_tokens\";\n\t/** Whether tool results require the `name` field. Default: auto-detected from URL. */\n\trequiresToolResultName?: boolean;\n\t/** Whether a user message after tool results requires an assistant message in between. Default: auto-detected from URL. */\n\trequiresAssistantAfterToolResult?: boolean;\n\t/** Whether thinking blocks must be converted to text blocks with <thinking> delimiters. Default: auto-detected from URL. */\n\trequiresThinkingAsText?: boolean;\n\t/** Whether all replayed assistant messages must include an empty reasoning_content field when reasoning is enabled. Default: auto-detected from URL. */\n\trequiresReasoningContentOnAssistantMessages?: boolean;\n\t/** Format for reasoning/thinking parameter. \"openai\" uses reasoning_effort, \"openrouter\" uses reasoning: { effort }, \"deepseek\" uses thinking: { type } plus reasoning_effort, \"together\" uses reasoning: { enabled } plus reasoning_effort when supported, \"zai\" uses top-level enable_thinking: boolean, \"qwen\" uses top-level enable_thinking: boolean, and \"qwen-chat-template\" uses chat_template_kwargs.enable_thinking. Default: \"openai\". */\n\tthinkingFormat?: \"openai\" | \"openrouter\" | \"deepseek\" | \"together\" | \"zai\" | \"qwen\" | \"qwen-chat-template\";\n\t/** OpenRouter-specific routing preferences. Only used when baseUrl points to OpenRouter. */\n\topenRouterRouting?: OpenRouterRouting;\n\t/** Vercel AI Gateway routing preferences. Only used when baseUrl points to Vercel AI Gateway. */\n\tvercelGatewayRouting?: VercelGatewayRouting;\n\t/** Whether z.ai supports top-level `tool_stream: true` for streaming tool call deltas. Default: false. */\n\tzaiToolStream?: boolean;\n\t/** Whether the provider supports the `strict` field in tool definitions. Default: true. */\n\tsupportsStrictMode?: boolean;\n\t/** Cache control convention for prompt caching. \"anthropic\" applies Anthropic-style `cache_control` markers to the system prompt, last tool definition, and last user/assistant text content. */\n\tcacheControlFormat?: \"anthropic\";\n\t/** Whether to send known session-affinity headers (`session_id`, `x-client-request-id`, `x-session-affinity`) from `options.sessionId` when caching is enabled. Default: false. */\n\tsendSessionAffinityHeaders?: boolean;\n\t/** Whether the provider supports long prompt cache retention (`prompt_cache_retention: \"24h\"` or Anthropic-style `cache_control.ttl: \"1h\"`, depending on format). Default: true. */\n\tsupportsLongCacheRetention?: boolean;\n\t/**\n\t * Text appended to the last user message before sending. Used to pass\n\t * inline model directives that some local/OpenAI-compatible runtimes only\n\t * honor in the prompt text (for example Qwen3 `/no_think` on an MLX server,\n\t * which may not respect `chat_template_kwargs.enable_thinking`). Default: none.\n\t */\n\tpromptSuffix?: string;\n}\n\n/** Compatibility settings for OpenAI Responses APIs. */\nexport interface OpenAIResponsesCompat {\n\t/** Whether to send the OpenAI `session_id` cache-affinity header from `options.sessionId` when caching is enabled. Default: true. */\n\tsendSessionIdHeader?: boolean;\n\t/** Whether the provider supports `prompt_cache_retention: \"24h\"`. Default: true. */\n\tsupportsLongCacheRetention?: boolean;\n}\n\n/** Compatibility settings for Anthropic Messages-compatible APIs. */\nexport interface AnthropicMessagesCompat {\n\t/**\n\t * Whether the provider accepts per-tool `eager_input_streaming`.\n\t * When false, the Anthropic provider omits `tools[].eager_input_streaming`\n\t * and sends the legacy `fine-grained-tool-streaming-2025-05-14` beta header\n\t * for tool-enabled requests.\n\t * Default: true.\n\t */\n\tsupportsEagerToolInputStreaming?: boolean;\n\t/** Whether the provider supports Anthropic long cache retention (`cache_control.ttl: \"1h\"`). Default: true. */\n\tsupportsLongCacheRetention?: boolean;\n}\n\n/**\n * OpenRouter provider routing preferences.\n * Controls which upstream providers OpenRouter routes requests to.\n * Sent as the `provider` field in the OpenRouter API request body.\n * @see https://openrouter.ai/docs/guides/routing/provider-selection\n */\nexport interface OpenRouterRouting {\n\t/** Whether to allow backup providers to serve requests. Default: true. */\n\tallow_fallbacks?: boolean;\n\t/** Whether to filter providers to only those that support all parameters in the request. Default: false. */\n\trequire_parameters?: boolean;\n\t/** Data collection setting. \"allow\" (default): allow providers that may store/train on data. \"deny\": only use providers that don't collect user data. */\n\tdata_collection?: \"deny\" | \"allow\";\n\t/** Whether to restrict routing to only ZDR (Zero Data Retention) endpoints. */\n\tzdr?: boolean;\n\t/** Whether to restrict routing to only models that allow text distillation. */\n\tenforce_distillable_text?: boolean;\n\t/** An ordered list of provider names/slugs to try in sequence, falling back to the next if unavailable. */\n\torder?: string[];\n\t/** List of provider names/slugs to exclusively allow for this request. */\n\tonly?: string[];\n\t/** List of provider names/slugs to skip for this request. */\n\tignore?: string[];\n\t/** A list of quantization levels to filter providers by (e.g., [\"fp16\", \"bf16\", \"fp8\", \"fp6\", \"int8\", \"int4\", \"fp4\", \"fp32\"]). */\n\tquantizations?: string[];\n\t/** Sorting strategy. Can be a string (e.g., \"price\", \"throughput\", \"latency\") or an object with `by` and `partition`. */\n\tsort?:\n\t\t| string\n\t\t| {\n\t\t\t\t/** The sorting metric: \"price\", \"throughput\", \"latency\". */\n\t\t\t\tby?: string;\n\t\t\t\t/** Partitioning strategy: \"model\" (default) or \"none\". */\n\t\t\t\tpartition?: string | null;\n\t\t  };\n\t/** Maximum price per million tokens (USD). */\n\tmax_price?: {\n\t\t/** Price per million prompt tokens. */\n\t\tprompt?: number | string;\n\t\t/** Price per million completion tokens. */\n\t\tcompletion?: number | string;\n\t\t/** Price per image. */\n\t\timage?: number | string;\n\t\t/** Price per audio unit. */\n\t\taudio?: number | string;\n\t\t/** Price per request. */\n\t\trequest?: number | string;\n\t};\n\t/** Preferred minimum throughput (tokens/second). Can be a number (applies to p50) or an object with percentile-specific cutoffs. */\n\tpreferred_min_throughput?:\n\t\t| number\n\t\t| {\n\t\t\t\t/** Minimum tokens/second at the 50th percentile. */\n\t\t\t\tp50?: number;\n\t\t\t\t/** Minimum tokens/second at the 75th percentile. */\n\t\t\t\tp75?: number;\n\t\t\t\t/** Minimum tokens/second at the 90th percentile. */\n\t\t\t\tp90?: number;\n\t\t\t\t/** Minimum tokens/second at the 99th percentile. */\n\t\t\t\tp99?: number;\n\t\t  };\n\t/** Preferred maximum latency (seconds). Can be a number (applies to p50) or an object with percentile-specific cutoffs. */\n\tpreferred_max_latency?:\n\t\t| number\n\t\t| {\n\t\t\t\t/** Maximum latency in seconds at the 50th percentile. */\n\t\t\t\tp50?: number;\n\t\t\t\t/** Maximum latency in seconds at the 75th percentile. */\n\t\t\t\tp75?: number;\n\t\t\t\t/** Maximum latency in seconds at the 90th percentile. */\n\t\t\t\tp90?: number;\n\t\t\t\t/** Maximum latency in seconds at the 99th percentile. */\n\t\t\t\tp99?: number;\n\t\t  };\n}\n\n/**\n * Vercel AI Gateway routing preferences.\n * Controls which upstream providers the gateway routes requests to.\n * @see https://vercel.com/docs/ai-gateway/models-and-providers/provider-options\n */\nexport interface VercelGatewayRouting {\n\t/** List of provider slugs to exclusively use for this request (e.g., [\"bedrock\", \"anthropic\"]). */\n\tonly?: string[];\n\t/** List of provider slugs to try in order (e.g., [\"anthropic\", \"openai\"]). */\n\torder?: string[];\n}\n\n// Model interface for the unified model system\nexport interface Model<TApi extends Api> {\n\tid: string;\n\tname: string;\n\tapi: TApi;\n\tprovider: Provider;\n\tbaseUrl: string;\n\treasoning: boolean;\n\t/**\n\t * Maps pi thinking levels to provider/model-specific values.\n\t * Missing keys use provider defaults. null marks a level as unsupported.\n\t */\n\tthinkingLevelMap?: ThinkingLevelMap;\n\tinput: (\"text\" | \"image\")[];\n\tcost: {\n\t\tinput: number; // $/million tokens\n\t\toutput: number; // $/million tokens\n\t\tcacheRead: number; // $/million tokens\n\t\tcacheWrite: number; // $/million tokens\n\t};\n\tcontextWindow: number;\n\tmaxTokens: number;\n\theaders?: Record<string, string>;\n\t/** Compatibility overrides for OpenAI-compatible APIs. If not set, auto-detected from baseUrl. */\n\tcompat?: TApi extends \"openai-completions\"\n\t\t? OpenAICompletionsCompat\n\t\t: TApi extends \"openai-responses\"\n\t\t\t? OpenAIResponsesCompat\n\t\t\t: TApi extends \"anthropic-messages\"\n\t\t\t\t? AnthropicMessagesCompat\n\t\t\t\t: never;\n}\n\nexport interface ImagesModel<TApi extends ImagesApi>\n\textends Omit<Model<Api>, \"api\" | \"provider\" | \"reasoning\" | \"contextWindow\" | \"maxTokens\" | \"compat\"> {\n\tapi: TApi;\n\tprovider: ImagesProvider;\n\toutput: (\"text\" | \"image\")[];\n}\n"]}
+{"version":3,"file":"types.js","sourceRoot":"","sources":["../src/types.ts"],"names":[],"mappings":"","sourcesContent":["import type { AssistantMessageDiagnostic } from \"./utils/diagnostics.js\";\nimport type { AssistantMessageEventStream } from \"./utils/event-stream.js\";\n\nexport type { AssistantMessageEventStream } from \"./utils/event-stream.js\";\n\nexport type KnownApi =\n\t| \"openai-completions\"\n\t| \"openai-responses\"\n\t| \"azure-openai-responses\"\n\t| \"openai-codex-responses\"\n\t| \"anthropic-messages\"\n\t| \"google-generative-ai\"\n\t| \"google-vertex\";\n\nexport type Api = KnownApi | (string & {});\n\nexport type KnownImagesApi = \"openrouter-images\";\n\nexport type ImagesApi = KnownImagesApi | (string & {});\n\nexport type KnownProvider =\n\t| \"anthropic\"\n\t| \"google\"\n\t| \"google-vertex\"\n\t| \"openai\"\n\t| \"azure-openai-responses\"\n\t| \"openai-codex\"\n\t| \"deepseek\"\n\t| \"github-copilot\"\n\t| \"xai\"\n\t| \"groq\"\n\t| \"cerebras\"\n\t| \"openrouter\"\n\t| \"vercel-ai-gateway\"\n\t| \"zai\"\n\t| \"minimax\"\n\t| \"minimax-cn\"\n\t| \"moonshotai\"\n\t| \"moonshotai-cn\"\n\t| \"huggingface\"\n\t| \"fireworks\"\n\t| \"together\"\n\t| \"opencode\"\n\t| \"opencode-go\"\n\t| \"kimi-coding\"\n\t| \"xiaomi\"\n\t| \"xiaomi-token-plan-cn\"\n\t| \"xiaomi-token-plan-ams\"\n\t| \"xiaomi-token-plan-sgp\"\n\t| \"nvidia\";\nexport type Provider = KnownProvider | string;\n\nexport type KnownImagesProvider = \"openrouter\";\n\nexport type ImagesProvider = KnownImagesProvider | string;\n\nexport type ThinkingLevel = \"minimal\" | \"low\" | \"medium\" | \"high\" | \"xhigh\";\nexport type ModelThinkingLevel = \"off\" | ThinkingLevel;\nexport type ThinkingLevelMap = Partial<Record<ModelThinkingLevel, string | null>>;\n\n/** Token budgets for each thinking level (token-based providers only) */\nexport interface ThinkingBudgets {\n\tminimal?: number;\n\tlow?: number;\n\tmedium?: number;\n\thigh?: number;\n}\n\n// Base options all providers share\nexport type CacheRetention = \"none\" | \"short\" | \"long\";\n\nexport type Transport = \"sse\" | \"websocket\" | \"websocket-cached\" | \"auto\";\n\nexport interface ProviderResponse {\n\tstatus: number;\n\theaders: Record<string, string>;\n}\n\nexport interface StreamOptions {\n\ttemperature?: number;\n\tmaxTokens?: number;\n\tsignal?: AbortSignal;\n\tapiKey?: string;\n\t/**\n\t * Preferred transport for providers that support multiple transports.\n\t * Providers that do not support this option ignore it.\n\t */\n\ttransport?: Transport;\n\t/**\n\t * Prompt cache retention preference. Providers map this to their supported values.\n\t * Default: \"short\".\n\t */\n\tcacheRetention?: CacheRetention;\n\t/**\n\t * Optional session identifier for providers that support session-based caching.\n\t * Providers can use this to enable prompt caching, request routing, or other\n\t * session-aware features. Ignored by providers that don't support it.\n\t */\n\tsessionId?: string;\n\t/**\n\t * Optional callback for inspecting or replacing provider payloads before sending.\n\t * Return undefined to keep the payload unchanged.\n\t */\n\tonPayload?: (payload: unknown, model: Model<Api>) => unknown | undefined | Promise<unknown | undefined>;\n\t/**\n\t * Optional callback invoked after an HTTP response is received and before\n\t * its body stream is consumed.\n\t */\n\tonResponse?: (response: ProviderResponse, model: Model<Api>) => void | Promise<void>;\n\t/**\n\t * Optional custom HTTP headers to include in API requests.\n\t * Merged with provider defaults; can override default headers.\n\t * Not supported by all providers (e.g., AWS Bedrock uses SDK auth).\n\t */\n\theaders?: Record<string, string>;\n\t/**\n\t * HTTP request timeout in milliseconds for providers/SDKs that support it.\n\t * For example, OpenAI and Anthropic SDK clients default to 10 minutes.\n\t */\n\ttimeoutMs?: number;\n\t/**\n\t * Maximum retry attempts for providers/SDKs that support client-side retries.\n\t * For example, OpenAI and Anthropic SDK clients default to 2.\n\t */\n\tmaxRetries?: number;\n\t/**\n\t * Maximum delay in milliseconds to wait for a retry when the server requests a long wait.\n\t * If the server's requested delay exceeds this value, the request fails immediately\n\t * with an error containing the requested delay, allowing higher-level retry logic\n\t * to handle it with user visibility.\n\t * Default: 60000 (60 seconds). Set to 0 to disable the cap.\n\t */\n\tmaxRetryDelayMs?: number;\n\t/**\n\t * Optional metadata to include in API requests.\n\t * Providers extract the fields they understand and ignore the rest.\n\t * For example, Anthropic uses `user_id` for abuse tracking and rate limiting.\n\t */\n\tmetadata?: Record<string, unknown>;\n}\n\nexport type ProviderStreamOptions = StreamOptions & Record<string, unknown>;\n\nexport interface ImagesOptions {\n\tsignal?: AbortSignal;\n\tapiKey?: string;\n\t/**\n\t * Optional callback for inspecting or replacing provider payloads before sending.\n\t * Return undefined to keep the payload unchanged.\n\t */\n\tonPayload?: (payload: unknown, model: ImagesModel<ImagesApi>) => unknown | undefined | Promise<unknown | undefined>;\n\t/**\n\t * Optional callback invoked after an HTTP response is received.\n\t */\n\tonResponse?: (response: ProviderResponse, model: ImagesModel<ImagesApi>) => void | Promise<void>;\n\t/**\n\t * Optional custom HTTP headers to include in API requests.\n\t * Merged with provider defaults; can override default headers.\n\t */\n\theaders?: Record<string, string>;\n\t/**\n\t * HTTP request timeout in milliseconds for providers/SDKs that support it.\n\t */\n\ttimeoutMs?: number;\n\t/**\n\t * Maximum retry attempts for providers/SDKs that support client-side retries.\n\t */\n\tmaxRetries?: number;\n\t/**\n\t * Maximum delay in milliseconds to wait for a retry when the server requests a long wait.\n\t * If the server's requested delay exceeds this value, the request fails immediately\n\t * with an error containing the requested delay, allowing higher-level retry logic\n\t * to handle it with user visibility.\n\t * Default: 60000 (60 seconds). Set to 0 to disable the cap.\n\t */\n\tmaxRetryDelayMs?: number;\n\t/**\n\t * Optional metadata to include in API requests.\n\t * Providers extract the fields they understand and ignore the rest.\n\t */\n\tmetadata?: Record<string, unknown>;\n}\n\nexport type ProviderImagesOptions = ImagesOptions & Record<string, unknown>;\n\n// Unified options with reasoning passed to streamSimple() and completeSimple()\nexport interface SimpleStreamOptions extends StreamOptions {\n\treasoning?: ThinkingLevel;\n\t/** Custom token budgets for thinking levels (token-based providers only) */\n\tthinkingBudgets?: ThinkingBudgets;\n\t/**\n\t * Controls how thinking content is returned by adaptive-thinking models\n\t * (Anthropic Messages API).\n\t * - \"summarized\": thinking blocks contain summarized thinking text.\n\t * - \"omitted\": thinking blocks come back empty (the encrypted signature still\n\t *   travels back for continuity). The model still reasons at the requested\n\t *   effort; skipping the summary stream lowers time-to-first-token, which\n\t *   speeds up tool-use turns. When unset, the provider picks a per-model\n\t *   default (Opus 4.8 defaults to \"omitted\").\n\t */\n\tthinkingDisplay?: \"summarized\" | \"omitted\";\n}\n\n// Generic StreamFunction with typed options.\n//\n// Contract:\n// - Must return an AssistantMessageEventStream.\n// - Once invoked, request/model/runtime failures should be encoded in the\n//   returned stream, not thrown.\n// - Error termination must produce an AssistantMessage with stopReason\n//   \"error\" or \"aborted\" and errorMessage, emitted via the stream protocol.\nexport type StreamFunction<TApi extends Api = Api, TOptions extends StreamOptions = StreamOptions> = (\n\tmodel: Model<TApi>,\n\tcontext: Context,\n\toptions?: TOptions,\n) => AssistantMessageEventStream;\n\nexport type ImagesFunction<TApi extends ImagesApi = ImagesApi, TOptions extends ImagesOptions = ImagesOptions> = (\n\tmodel: ImagesModel<TApi>,\n\tcontext: ImagesContext,\n\toptions?: TOptions,\n) => Promise<AssistantImages>;\n\nexport interface TextSignatureV1 {\n\tv: 1;\n\tid: string;\n\tphase?: \"commentary\" | \"final_answer\";\n}\n\nexport interface TextContent {\n\ttype: \"text\";\n\ttext: string;\n\ttextSignature?: string; // e.g., for OpenAI responses, message metadata (legacy id string or TextSignatureV1 JSON)\n}\n\nexport interface ThinkingContent {\n\ttype: \"thinking\";\n\tthinking: string;\n\tthinkingSignature?: string; // e.g., for OpenAI responses, the reasoning item ID\n\t/** When true, the thinking content was redacted by safety filters. The opaque\n\t *  encrypted payload is stored in `thinkingSignature` so it can be passed back\n\t *  to the API for multi-turn continuity. */\n\tredacted?: boolean;\n}\n\nexport interface ImageContent {\n\ttype: \"image\";\n\tdata: string; // base64 encoded image data\n\tmimeType: string; // e.g., \"image/jpeg\", \"image/png\"\n}\n\nexport interface ToolCall {\n\ttype: \"toolCall\";\n\tid: string;\n\tname: string;\n\targuments: Record<string, any>;\n\tthoughtSignature?: string; // Google-specific: opaque signature for reusing thought context\n}\n\nexport interface Usage {\n\tinput: number;\n\toutput: number;\n\tcacheRead: number;\n\tcacheWrite: number;\n\ttotalTokens: number;\n\tcost: {\n\t\tinput: number;\n\t\toutput: number;\n\t\tcacheRead: number;\n\t\tcacheWrite: number;\n\t\ttotal: number;\n\t};\n}\n\nexport type StopReason = \"stop\" | \"length\" | \"toolUse\" | \"error\" | \"aborted\";\n\nexport interface UserMessage {\n\trole: \"user\";\n\tcontent: string | (TextContent | ImageContent)[];\n\ttimestamp: number; // Unix timestamp in milliseconds\n}\n\nexport interface AssistantMessage {\n\trole: \"assistant\";\n\tcontent: (TextContent | ThinkingContent | ToolCall)[];\n\tapi: Api;\n\tprovider: Provider;\n\tmodel: string;\n\tresponseModel?: string; // Concrete `chunk.model` when different from the requested `model` (e.g. OpenRouter `auto` -> `anthropic/...`)\n\tresponseId?: string; // Provider-specific response/message identifier when the upstream API exposes one\n\tdiagnostics?: AssistantMessageDiagnostic[]; // Redacted provider/runtime diagnostics for failures and recoveries.\n\tusage: Usage;\n\tstopReason: StopReason;\n\terrorMessage?: string;\n\ttimestamp: number; // Unix timestamp in milliseconds\n}\n\nexport interface ToolResultMessage<TDetails = any> {\n\trole: \"toolResult\";\n\ttoolCallId: string;\n\ttoolName: string;\n\tcontent: (TextContent | ImageContent)[]; // Supports text and images\n\tdetails?: TDetails;\n\tisError: boolean;\n\ttimestamp: number; // Unix timestamp in milliseconds\n}\n\nexport type Message = UserMessage | AssistantMessage | ToolResultMessage;\n\nexport type ImagesInputContent = TextContent | ImageContent;\nexport type ImagesOutputContent = TextContent | ImageContent;\n\nexport interface ImagesContext {\n\tinput: ImagesInputContent[];\n}\n\nexport type ImagesStopReason = \"stop\" | \"error\" | \"aborted\";\n\nexport interface AssistantImages {\n\tapi: ImagesApi;\n\tprovider: ImagesProvider;\n\tmodel: string;\n\toutput: ImagesOutputContent[];\n\tresponseId?: string;\n\tusage?: Usage;\n\tstopReason: ImagesStopReason;\n\terrorMessage?: string;\n\ttimestamp: number; // Unix timestamp in milliseconds\n}\n\nimport type { TSchema } from \"typebox\";\n\nexport interface Tool<TParameters extends TSchema = TSchema> {\n\tname: string;\n\tdescription: string;\n\tparameters: TParameters;\n}\n\nexport interface Context {\n\tsystemPrompt?: string;\n\tmessages: Message[];\n\ttools?: Tool[];\n}\n\n/**\n * Event protocol for AssistantMessageEventStream.\n *\n * Streams should emit `start` before partial updates, then terminate with either:\n * - `done` carrying the final successful AssistantMessage, or\n * - `error` carrying the final AssistantMessage with stopReason \"error\" or \"aborted\"\n *   and errorMessage.\n */\nexport type AssistantMessageEvent =\n\t| { type: \"start\"; partial: AssistantMessage }\n\t| { type: \"text_start\"; contentIndex: number; partial: AssistantMessage }\n\t| { type: \"text_delta\"; contentIndex: number; delta: string; partial: AssistantMessage }\n\t| { type: \"text_end\"; contentIndex: number; content: string; partial: AssistantMessage }\n\t| { type: \"thinking_start\"; contentIndex: number; partial: AssistantMessage }\n\t| { type: \"thinking_delta\"; contentIndex: number; delta: string; partial: AssistantMessage }\n\t| { type: \"thinking_end\"; contentIndex: number; content: string; partial: AssistantMessage }\n\t| { type: \"toolcall_start\"; contentIndex: number; partial: AssistantMessage }\n\t| { type: \"toolcall_delta\"; contentIndex: number; delta: string; partial: AssistantMessage }\n\t| { type: \"toolcall_end\"; contentIndex: number; toolCall: ToolCall; partial: AssistantMessage }\n\t| { type: \"done\"; reason: Extract<StopReason, \"stop\" | \"length\" | \"toolUse\">; message: AssistantMessage }\n\t| { type: \"error\"; reason: Extract<StopReason, \"aborted\" | \"error\">; error: AssistantMessage };\n\n/**\n * Compatibility settings for OpenAI-compatible completions APIs.\n * Use this to override URL-based auto-detection for custom providers.\n */\nexport interface OpenAICompletionsCompat {\n\t/** Whether the provider supports the `store` field. Default: auto-detected from URL. */\n\tsupportsStore?: boolean;\n\t/** Whether the provider supports the `developer` role (vs `system`). Default: auto-detected from URL. */\n\tsupportsDeveloperRole?: boolean;\n\t/** Whether the provider supports `reasoning_effort`. Default: auto-detected from URL. */\n\tsupportsReasoningEffort?: boolean;\n\t/** Whether the provider supports `stream_options: { include_usage: true }` for token usage in streaming responses. Default: true. */\n\tsupportsUsageInStreaming?: boolean;\n\t/** Which field to use for max tokens. Default: auto-detected from URL. */\n\tmaxTokensField?: \"max_completion_tokens\" | \"max_tokens\";\n\t/** Whether tool results require the `name` field. Default: auto-detected from URL. */\n\trequiresToolResultName?: boolean;\n\t/** Whether a user message after tool results requires an assistant message in between. Default: auto-detected from URL. */\n\trequiresAssistantAfterToolResult?: boolean;\n\t/** Whether thinking blocks must be converted to text blocks with <thinking> delimiters. Default: auto-detected from URL. */\n\trequiresThinkingAsText?: boolean;\n\t/** Whether all replayed assistant messages must include an empty reasoning_content field when reasoning is enabled. Default: auto-detected from URL. */\n\trequiresReasoningContentOnAssistantMessages?: boolean;\n\t/** Format for reasoning/thinking parameter. \"openai\" uses reasoning_effort, \"openrouter\" uses reasoning: { effort }, \"deepseek\" uses thinking: { type } plus reasoning_effort, \"together\" uses reasoning: { enabled } plus reasoning_effort when supported, \"zai\" uses top-level enable_thinking: boolean, \"qwen\" uses top-level enable_thinking: boolean, and \"qwen-chat-template\" uses chat_template_kwargs.enable_thinking. Default: \"openai\". */\n\tthinkingFormat?: \"openai\" | \"openrouter\" | \"deepseek\" | \"together\" | \"zai\" | \"qwen\" | \"qwen-chat-template\";\n\t/** OpenRouter-specific routing preferences. Only used when baseUrl points to OpenRouter. */\n\topenRouterRouting?: OpenRouterRouting;\n\t/** Vercel AI Gateway routing preferences. Only used when baseUrl points to Vercel AI Gateway. */\n\tvercelGatewayRouting?: VercelGatewayRouting;\n\t/** Whether z.ai supports top-level `tool_stream: true` for streaming tool call deltas. Default: false. */\n\tzaiToolStream?: boolean;\n\t/** Whether the provider supports the `strict` field in tool definitions. Default: true. */\n\tsupportsStrictMode?: boolean;\n\t/** Cache control convention for prompt caching. \"anthropic\" applies Anthropic-style `cache_control` markers to the system prompt, last tool definition, and last user/assistant text content. */\n\tcacheControlFormat?: \"anthropic\";\n\t/** Whether to send known session-affinity headers (`session_id`, `x-client-request-id`, `x-session-affinity`) from `options.sessionId` when caching is enabled. Default: false. */\n\tsendSessionAffinityHeaders?: boolean;\n\t/** Whether the provider supports long prompt cache retention (`prompt_cache_retention: \"24h\"` or Anthropic-style `cache_control.ttl: \"1h\"`, depending on format). Default: true. */\n\tsupportsLongCacheRetention?: boolean;\n\t/**\n\t * Text appended to the last user message before sending. Used to pass\n\t * inline model directives that some local/OpenAI-compatible runtimes only\n\t * honor in the prompt text (for example Qwen3 `/no_think` on an MLX server,\n\t * which may not respect `chat_template_kwargs.enable_thinking`). Default: none.\n\t */\n\tpromptSuffix?: string;\n}\n\n/** Compatibility settings for OpenAI Responses APIs. */\nexport interface OpenAIResponsesCompat {\n\t/** Whether to send the OpenAI `session_id` cache-affinity header from `options.sessionId` when caching is enabled. Default: true. */\n\tsendSessionIdHeader?: boolean;\n\t/** Whether the provider supports `prompt_cache_retention: \"24h\"`. Default: true. */\n\tsupportsLongCacheRetention?: boolean;\n}\n\n/** Compatibility settings for Anthropic Messages-compatible APIs. */\nexport interface AnthropicMessagesCompat {\n\t/**\n\t * Whether the provider accepts per-tool `eager_input_streaming`.\n\t * When false, the Anthropic provider omits `tools[].eager_input_streaming`\n\t * and sends the legacy `fine-grained-tool-streaming-2025-05-14` beta header\n\t * for tool-enabled requests.\n\t * Default: true.\n\t */\n\tsupportsEagerToolInputStreaming?: boolean;\n\t/** Whether the provider supports Anthropic long cache retention (`cache_control.ttl: \"1h\"`). Default: true. */\n\tsupportsLongCacheRetention?: boolean;\n}\n\n/**\n * OpenRouter provider routing preferences.\n * Controls which upstream providers OpenRouter routes requests to.\n * Sent as the `provider` field in the OpenRouter API request body.\n * @see https://openrouter.ai/docs/guides/routing/provider-selection\n */\nexport interface OpenRouterRouting {\n\t/** Whether to allow backup providers to serve requests. Default: true. */\n\tallow_fallbacks?: boolean;\n\t/** Whether to filter providers to only those that support all parameters in the request. Default: false. */\n\trequire_parameters?: boolean;\n\t/** Data collection setting. \"allow\" (default): allow providers that may store/train on data. \"deny\": only use providers that don't collect user data. */\n\tdata_collection?: \"deny\" | \"allow\";\n\t/** Whether to restrict routing to only ZDR (Zero Data Retention) endpoints. */\n\tzdr?: boolean;\n\t/** Whether to restrict routing to only models that allow text distillation. */\n\tenforce_distillable_text?: boolean;\n\t/** An ordered list of provider names/slugs to try in sequence, falling back to the next if unavailable. */\n\torder?: string[];\n\t/** List of provider names/slugs to exclusively allow for this request. */\n\tonly?: string[];\n\t/** List of provider names/slugs to skip for this request. */\n\tignore?: string[];\n\t/** A list of quantization levels to filter providers by (e.g., [\"fp16\", \"bf16\", \"fp8\", \"fp6\", \"int8\", \"int4\", \"fp4\", \"fp32\"]). */\n\tquantizations?: string[];\n\t/** Sorting strategy. Can be a string (e.g., \"price\", \"throughput\", \"latency\") or an object with `by` and `partition`. */\n\tsort?:\n\t\t| string\n\t\t| {\n\t\t\t\t/** The sorting metric: \"price\", \"throughput\", \"latency\". */\n\t\t\t\tby?: string;\n\t\t\t\t/** Partitioning strategy: \"model\" (default) or \"none\". */\n\t\t\t\tpartition?: string | null;\n\t\t  };\n\t/** Maximum price per million tokens (USD). */\n\tmax_price?: {\n\t\t/** Price per million prompt tokens. */\n\t\tprompt?: number | string;\n\t\t/** Price per million completion tokens. */\n\t\tcompletion?: number | string;\n\t\t/** Price per image. */\n\t\timage?: number | string;\n\t\t/** Price per audio unit. */\n\t\taudio?: number | string;\n\t\t/** Price per request. */\n\t\trequest?: number | string;\n\t};\n\t/** Preferred minimum throughput (tokens/second). Can be a number (applies to p50) or an object with percentile-specific cutoffs. */\n\tpreferred_min_throughput?:\n\t\t| number\n\t\t| {\n\t\t\t\t/** Minimum tokens/second at the 50th percentile. */\n\t\t\t\tp50?: number;\n\t\t\t\t/** Minimum tokens/second at the 75th percentile. */\n\t\t\t\tp75?: number;\n\t\t\t\t/** Minimum tokens/second at the 90th percentile. */\n\t\t\t\tp90?: number;\n\t\t\t\t/** Minimum tokens/second at the 99th percentile. */\n\t\t\t\tp99?: number;\n\t\t  };\n\t/** Preferred maximum latency (seconds). Can be a number (applies to p50) or an object with percentile-specific cutoffs. */\n\tpreferred_max_latency?:\n\t\t| number\n\t\t| {\n\t\t\t\t/** Maximum latency in seconds at the 50th percentile. */\n\t\t\t\tp50?: number;\n\t\t\t\t/** Maximum latency in seconds at the 75th percentile. */\n\t\t\t\tp75?: number;\n\t\t\t\t/** Maximum latency in seconds at the 90th percentile. */\n\t\t\t\tp90?: number;\n\t\t\t\t/** Maximum latency in seconds at the 99th percentile. */\n\t\t\t\tp99?: number;\n\t\t  };\n}\n\n/**\n * Vercel AI Gateway routing preferences.\n * Controls which upstream providers the gateway routes requests to.\n * @see https://vercel.com/docs/ai-gateway/models-and-providers/provider-options\n */\nexport interface VercelGatewayRouting {\n\t/** List of provider slugs to exclusively use for this request (e.g., [\"bedrock\", \"anthropic\"]). */\n\tonly?: string[];\n\t/** List of provider slugs to try in order (e.g., [\"anthropic\", \"openai\"]). */\n\torder?: string[];\n}\n\n// Model interface for the unified model system\nexport interface Model<TApi extends Api> {\n\tid: string;\n\tname: string;\n\tapi: TApi;\n\tprovider: Provider;\n\tbaseUrl: string;\n\treasoning: boolean;\n\t/**\n\t * Maps pi thinking levels to provider/model-specific values.\n\t * Missing keys use provider defaults. null marks a level as unsupported.\n\t */\n\tthinkingLevelMap?: ThinkingLevelMap;\n\tinput: (\"text\" | \"image\")[];\n\tcost: {\n\t\tinput: number; // $/million tokens\n\t\toutput: number; // $/million tokens\n\t\tcacheRead: number; // $/million tokens\n\t\tcacheWrite: number; // $/million tokens\n\t};\n\tcontextWindow: number;\n\tmaxTokens: number;\n\theaders?: Record<string, string>;\n\t/** Compatibility overrides for OpenAI-compatible APIs. If not set, auto-detected from baseUrl. */\n\tcompat?: TApi extends \"openai-completions\"\n\t\t? OpenAICompletionsCompat\n\t\t: TApi extends \"openai-responses\"\n\t\t\t? OpenAIResponsesCompat\n\t\t\t: TApi extends \"anthropic-messages\"\n\t\t\t\t? AnthropicMessagesCompat\n\t\t\t\t: never;\n}\n\nexport interface ImagesModel<TApi extends ImagesApi>\n\textends Omit<Model<Api>, \"api\" | \"provider\" | \"reasoning\" | \"contextWindow\" | \"maxTokens\" | \"compat\"> {\n\tapi: TApi;\n\tprovider: ImagesProvider;\n\toutput: (\"text\" | \"image\")[];\n}\n"]}

package/package.json

@@ -1,6 +1,6 @@
 {
 	"name": "@kolisachint/hoocode-ai",
-	"version": "0.4.114",
+	"version": "0.4.116",
 	"description": "Unified LLM API with automatic model discovery and provider configuration",
 	"type": "module",
 	"main": "./dist/index.js",
@@ -26,10 +26,6 @@
 			"types": "./dist/providers/google-vertex.d.ts",
 			"import": "./dist/providers/google-vertex.js"
 		},
-		"./mistral": {
-			"types": "./dist/providers/mistral.d.ts",
-			"import": "./dist/providers/mistral.js"
-		},
 		"./openai-codex-responses": {
 			"types": "./dist/providers/openai-codex-responses.d.ts",
 			"import": "./dist/providers/openai-codex-responses.js"
@@ -45,10 +41,6 @@
 		"./oauth": {
 			"types": "./dist/oauth.d.ts",
 			"import": "./dist/oauth.js"
-		},
-		"./bedrock-provider": {
-			"types": "./dist/bedrock-provider.d.ts",
-			"import": "./dist/bedrock-provider.js"
 		}
 	},
 	"bin": {
@@ -72,11 +64,7 @@
 	},
 	"dependencies": {
 		"@anthropic-ai/sdk": "^0.91.1",
-		"@aws-sdk/client-bedrock-runtime": "^3.1030.0",
 		"@google/genai": "^1.40.0",
-		"@mistralai/mistralai": "^2.2.0",
-		"@smithy/node-http-handler": "^4.7.5",
-		"@smithy/types": "^4.14.2",
 		"typebox": "^1.1.24",
 		"chalk": "^5.6.2",
 		"openai": "6.26.0",

package/dist/bedrock-provider.d.ts

@@ -1,5 +0,0 @@
-export declare const bedrockProviderModule: {
-    streamBedrock: import("./types.js").StreamFunction<"bedrock-converse-stream", import("./providers/amazon-bedrock.js").BedrockOptions>;
-    streamSimpleBedrock: import("./types.js").StreamFunction<"bedrock-converse-stream", import("./types.js").SimpleStreamOptions>;
-};
-//# sourceMappingURL=bedrock-provider.d.ts.map

package/dist/bedrock-provider.d.ts.map

@@ -1 +0,0 @@
-{"version":3,"file":"bedrock-provider.d.ts","sourceRoot":"","sources":["../src/bedrock-provider.ts"],"names":[],"mappings":"AAEA,eAAO,MAAM,qBAAqB;;;CAGjC,CAAC","sourcesContent":["import { streamBedrock, streamSimpleBedrock } from \"./providers/amazon-bedrock.js\";\n\nexport const bedrockProviderModule = {\n\tstreamBedrock,\n\tstreamSimpleBedrock,\n};\n"]}

package/dist/bedrock-provider.js

@@ -1,6 +0,0 @@
-import { streamBedrock, streamSimpleBedrock } from "./providers/amazon-bedrock.js";
-export const bedrockProviderModule = {
-    streamBedrock,
-    streamSimpleBedrock,
-};
-//# sourceMappingURL=bedrock-provider.js.map

package/dist/bedrock-provider.js.map

@@ -1 +0,0 @@
-{"version":3,"file":"bedrock-provider.js","sourceRoot":"","sources":["../src/bedrock-provider.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,aAAa,EAAE,mBAAmB,EAAE,MAAM,+BAA+B,CAAC;AAEnF,MAAM,CAAC,MAAM,qBAAqB,GAAG;IACpC,aAAa;IACb,mBAAmB;CACnB,CAAC","sourcesContent":["import { streamBedrock, streamSimpleBedrock } from \"./providers/amazon-bedrock.js\";\n\nexport const bedrockProviderModule = {\n\tstreamBedrock,\n\tstreamSimpleBedrock,\n};\n"]}

package/dist/providers/amazon-bedrock.d.ts

@@ -1,38 +0,0 @@
-import type { SimpleStreamOptions, StreamFunction, StreamOptions, ThinkingBudgets, ThinkingLevel } from "../types.js";
-export type BedrockThinkingDisplay = "summarized" | "omitted";
-export interface BedrockOptions extends StreamOptions {
-    region?: string;
-    profile?: string;
-    toolChoice?: "auto" | "any" | "none" | {
-        type: "tool";
-        name: string;
-    };
-    reasoning?: ThinkingLevel;
-    thinkingBudgets?: ThinkingBudgets;
-    interleavedThinking?: boolean;
-    /**
-     * Controls how Claude's thinking content is returned in responses.
-     * - "summarized": Thinking blocks contain summarized thinking text (default here).
-     * - "omitted": Thinking content is redacted but the signature still travels back
-     *   for multi-turn continuity, reducing time-to-first-text-token.
-     *
-     * Note: Anthropic's API default for Claude Opus 4.7 and Mythos Preview is
-     * "omitted". We default to "summarized" here to keep behavior consistent with
-     * older Claude 4 models. Only applies to Claude models on Bedrock.
-     */
-    thinkingDisplay?: BedrockThinkingDisplay;
-    /** Key-value pairs attached to the inference request for cost allocation tagging.
-     * Keys: max 64 chars, no `aws:` prefix. Values: max 256 chars. Max 50 pairs.
-     * Tags appear in AWS Cost Explorer split cost allocation data.
-     * @see https://docs.aws.amazon.com/bedrock/latest/APIReference/API_runtime_ConverseStream.html */
-    requestMetadata?: Record<string, string>;
-    /** Bearer token for Bedrock API key authentication.
-     * When set, bypasses SigV4 signing and sends Authorization: Bearer <token> instead.
-     * Requires `bedrock:CallWithBearerToken` IAM permission on the token's identity.
-     * Set via AWS_BEARER_TOKEN_BEDROCK env var or pass directly.
-     * @see https://docs.aws.amazon.com/service-authorization/latest/reference/list_amazonbedrock.html */
-    bearerToken?: string;
-}
-export declare const streamBedrock: StreamFunction<"bedrock-converse-stream", BedrockOptions>;
-export declare const streamSimpleBedrock: StreamFunction<"bedrock-converse-stream", SimpleStreamOptions>;
-//# sourceMappingURL=amazon-bedrock.d.ts.map

package/dist/providers/amazon-bedrock.d.ts.map

@@ -1 +0,0 @@
-{"version":3,"file":"amazon-bedrock.d.ts","sourceRoot":"","sources":["../../src/providers/amazon-bedrock.ts"],"names":[],"mappings":"AAwBA,OAAO,KAAK,EAMX,mBAAmB,EAEnB,cAAc,EACd,aAAa,EAEb,eAAe,EAEf,aAAa,EAIb,MAAM,aAAa,CAAC;AAQrB,MAAM,MAAM,sBAAsB,GAAG,YAAY,GAAG,SAAS,CAAC;AAE9D,MAAM,WAAW,cAAe,SAAQ,aAAa;IACpD,MAAM,CAAC,EAAE,MAAM,CAAC;IAChB,OAAO,CAAC,EAAE,MAAM,CAAC;IACjB,UAAU,CAAC,EAAE,MAAM,GAAG,KAAK,GAAG,MAAM,GAAG;QAAE,IAAI,EAAE,MAAM,CAAC;QAAC,IAAI,EAAE,MAAM,CAAA;KAAE,CAAC;IAEtE,SAAS,CAAC,EAAE,aAAa,CAAC;IAE1B,eAAe,CAAC,EAAE,eAAe,CAAC;IAElC,mBAAmB,CAAC,EAAE,OAAO,CAAC;IAC9B;;;;;;;;;OASG;IACH,eAAe,CAAC,EAAE,sBAAsB,CAAC;IACzC;;;sGAGkG;IAClG,eAAe,CAAC,EAAE,MAAM,CAAC,MAAM,EAAE,MAAM,CAAC,CAAC;IACzC;;;;yGAIqG;IACrG,WAAW,CAAC,EAAE,MAAM,CAAC;CACrB;AAID,eAAO,MAAM,aAAa,EAAE,cAAc,CAAC,yBAAyB,EAAE,cAAc,CAgMnF,CAAC;AAgCF,eAAO,MAAM,mBAAmB,EAAE,cAAc,CAAC,yBAAyB,EAAE,mBAAmB,CA0C9F,CAAC","sourcesContent":["import {\n\tBedrockRuntimeClient,\n\ttype BedrockRuntimeClientConfig,\n\tBedrockRuntimeServiceException,\n\tStopReason as BedrockStopReason,\n\ttype Tool as BedrockTool,\n\tCachePointType,\n\tCacheTTL,\n\ttype ContentBlock,\n\ttype ContentBlockDeltaEvent,\n\ttype ContentBlockStartEvent,\n\ttype ContentBlockStopEvent,\n\tConversationRole,\n\tConverseStreamCommand,\n\ttype ConverseStreamMetadataEvent,\n\tImageFormat,\n\ttype Message,\n\ttype SystemContentBlock,\n\ttype ToolChoice,\n\ttype ToolConfiguration,\n\tToolResultStatus,\n} from \"@aws-sdk/client-bedrock-runtime\";\nimport type { DocumentType } from \"@smithy/types\";\nimport { calculateCost } from \"../models.js\";\nimport type {\n\tApi,\n\tAssistantMessage,\n\tCacheRetention,\n\tContext,\n\tModel,\n\tSimpleStreamOptions,\n\tStopReason,\n\tStreamFunction,\n\tStreamOptions,\n\tTextContent,\n\tThinkingBudgets,\n\tThinkingContent,\n\tThinkingLevel,\n\tTool,\n\tToolCall,\n\tToolResultMessage,\n} from \"../types.js\";\nimport { AssistantMessageEventStream } from \"../utils/event-stream.js\";\nimport { parseStreamingJson } from \"../utils/json-parse.js\";\nimport { sanitizeSurrogates } from \"../utils/sanitize-unicode.js\";\nimport { resolveCacheRetention } from \"./cache-retention.js\";\nimport { adjustMaxTokensForThinking, buildBaseOptions, clampReasoning } from \"./simple-options.js\";\nimport { transformMessages } from \"./transform-messages.js\";\n\nexport type BedrockThinkingDisplay = \"summarized\" | \"omitted\";\n\nexport interface BedrockOptions extends StreamOptions {\n\tregion?: string;\n\tprofile?: string;\n\ttoolChoice?: \"auto\" | \"any\" | \"none\" | { type: \"tool\"; name: string };\n\t/* See https://docs.aws.amazon.com/bedrock/latest/userguide/inference-reasoning.html for supported models. */\n\treasoning?: ThinkingLevel;\n\t/* Custom token budgets per thinking level. Overrides default budgets. */\n\tthinkingBudgets?: ThinkingBudgets;\n\t/* Only supported by Claude 4.x models, see https://docs.aws.amazon.com/bedrock/latest/userguide/claude-messages-extended-thinking.html#claude-messages-extended-thinking-tool-use-interleaved */\n\tinterleavedThinking?: boolean;\n\t/**\n\t * Controls how Claude's thinking content is returned in responses.\n\t * - \"summarized\": Thinking blocks contain summarized thinking text (default here).\n\t * - \"omitted\": Thinking content is redacted but the signature still travels back\n\t *   for multi-turn continuity, reducing time-to-first-text-token.\n\t *\n\t * Note: Anthropic's API default for Claude Opus 4.7 and Mythos Preview is\n\t * \"omitted\". We default to \"summarized\" here to keep behavior consistent with\n\t * older Claude 4 models. Only applies to Claude models on Bedrock.\n\t */\n\tthinkingDisplay?: BedrockThinkingDisplay;\n\t/** Key-value pairs attached to the inference request for cost allocation tagging.\n\t * Keys: max 64 chars, no `aws:` prefix. Values: max 256 chars. Max 50 pairs.\n\t * Tags appear in AWS Cost Explorer split cost allocation data.\n\t * @see https://docs.aws.amazon.com/bedrock/latest/APIReference/API_runtime_ConverseStream.html */\n\trequestMetadata?: Record<string, string>;\n\t/** Bearer token for Bedrock API key authentication.\n\t * When set, bypasses SigV4 signing and sends Authorization: Bearer <token> instead.\n\t * Requires `bedrock:CallWithBearerToken` IAM permission on the token's identity.\n\t * Set via AWS_BEARER_TOKEN_BEDROCK env var or pass directly.\n\t * @see https://docs.aws.amazon.com/service-authorization/latest/reference/list_amazonbedrock.html */\n\tbearerToken?: string;\n}\n\ntype Block = (TextContent | ThinkingContent | ToolCall) & { index?: number; partialJson?: string };\n\nexport const streamBedrock: StreamFunction<\"bedrock-converse-stream\", BedrockOptions> = (\n\tmodel: Model<\"bedrock-converse-stream\">,\n\tcontext: Context,\n\toptions: BedrockOptions = {},\n): AssistantMessageEventStream => {\n\tconst stream = new AssistantMessageEventStream();\n\n\t(async () => {\n\t\tconst output: AssistantMessage = {\n\t\t\trole: \"assistant\",\n\t\t\tcontent: [],\n\t\t\tapi: \"bedrock-converse-stream\" as Api,\n\t\t\tprovider: model.provider,\n\t\t\tmodel: model.id,\n\t\t\tusage: {\n\t\t\t\tinput: 0,\n\t\t\t\toutput: 0,\n\t\t\t\tcacheRead: 0,\n\t\t\t\tcacheWrite: 0,\n\t\t\t\ttotalTokens: 0,\n\t\t\t\tcost: { input: 0, output: 0, cacheRead: 0, cacheWrite: 0, total: 0 },\n\t\t\t},\n\t\t\tstopReason: \"stop\",\n\t\t\ttimestamp: Date.now(),\n\t\t};\n\n\t\tconst blocks = output.content as Block[];\n\n\t\tconst config: BedrockRuntimeClientConfig = {\n\t\t\tprofile: options.profile,\n\t\t};\n\t\tconst configuredRegion = getConfiguredBedrockRegion(options);\n\t\tconst hasConfiguredProfile = hasConfiguredBedrockProfile();\n\t\tconst endpointRegion = getStandardBedrockEndpointRegion(model.baseUrl);\n\t\tconst useExplicitEndpoint = shouldUseExplicitBedrockEndpoint(\n\t\t\tmodel.baseUrl,\n\t\t\tconfiguredRegion,\n\t\t\thasConfiguredProfile,\n\t\t);\n\n\t\t// Only pin standard AWS Bedrock runtime endpoints when no region/profile is configured.\n\t\t// This preserves custom endpoints (VPC/proxy) from #3402 without forcing built-in\n\t\t// catalog defaults such as us-east-1 to override AWS_REGION/AWS_PROFILE.\n\t\tif (useExplicitEndpoint) {\n\t\t\tconfig.endpoint = model.baseUrl;\n\t\t}\n\n\t\t// Resolve bearer token for Bedrock API key auth.\n\t\tconst bearerToken = options.bearerToken || process.env.AWS_BEARER_TOKEN_BEDROCK || undefined;\n\t\tconst useBearerToken = bearerToken !== undefined && process.env.AWS_BEDROCK_SKIP_AUTH !== \"1\";\n\n\t\t// in Node.js/Bun environment only\n\t\tif (typeof process !== \"undefined\" && (process.versions?.node || process.versions?.bun)) {\n\t\t\t// Region resolution: explicit option > env vars > SDK default chain.\n\t\t\t// When AWS_PROFILE is set, we leave region undefined so the SDK can\n\t\t\t// resovle it from aws profile configs. Otherwise fall back to us-east-1.\n\t\t\tif (configuredRegion) {\n\t\t\t\tconfig.region = configuredRegion;\n\t\t\t} else if (endpointRegion && useExplicitEndpoint) {\n\t\t\t\tconfig.region = endpointRegion;\n\t\t\t} else if (!hasConfiguredProfile) {\n\t\t\t\tconfig.region = \"us-east-1\";\n\t\t\t}\n\n\t\t\t// Support proxies that don't need authentication\n\t\t\tif (process.env.AWS_BEDROCK_SKIP_AUTH === \"1\") {\n\t\t\t\tconfig.credentials = {\n\t\t\t\t\taccessKeyId: \"dummy-access-key\",\n\t\t\t\t\tsecretAccessKey: \"dummy-secret-key\",\n\t\t\t\t};\n\t\t\t}\n\n\t\t\tif (\n\t\t\t\tprocess.env.HTTP_PROXY ||\n\t\t\t\tprocess.env.HTTPS_PROXY ||\n\t\t\t\tprocess.env.NO_PROXY ||\n\t\t\t\tprocess.env.http_proxy ||\n\t\t\t\tprocess.env.https_proxy ||\n\t\t\t\tprocess.env.no_proxy\n\t\t\t) {\n\t\t\t\tconst nodeHttpHandler = await import(\"@smithy/node-http-handler\");\n\t\t\t\tconst proxyAgent = await import(\"proxy-agent\");\n\n\t\t\t\tconst agent = new proxyAgent.ProxyAgent();\n\n\t\t\t\t// Bedrock runtime uses NodeHttp2Handler by default since v3.798.0, which is based\n\t\t\t\t// on `http2` module and has no support for http agent.\n\t\t\t\t// Use NodeHttpHandler to support http agent.\n\t\t\t\tconfig.requestHandler = new nodeHttpHandler.NodeHttpHandler({\n\t\t\t\t\thttpAgent: agent,\n\t\t\t\t\thttpsAgent: agent,\n\t\t\t\t});\n\t\t\t} else if (process.env.AWS_BEDROCK_FORCE_HTTP1 === \"1\") {\n\t\t\t\t// Some custom endpoints require HTTP/1.1 instead of HTTP/2\n\t\t\t\tconst nodeHttpHandler = await import(\"@smithy/node-http-handler\");\n\t\t\t\tconfig.requestHandler = new nodeHttpHandler.NodeHttpHandler();\n\t\t\t}\n\t\t} else {\n\t\t\t// Non-Node environment (browser): fall back to us-east-1 since\n\t\t\t// there's no config file resolution available.\n\t\t\tconfig.region =\n\t\t\t\tconfiguredRegion || (endpointRegion && useExplicitEndpoint ? endpointRegion : undefined) || \"us-east-1\";\n\t\t}\n\n\t\tif (useBearerToken) {\n\t\t\tconfig.token = { token: bearerToken };\n\t\t\tconfig.authSchemePreference = [\"httpBearerAuth\"];\n\t\t}\n\n\t\ttry {\n\t\t\tconst client = new BedrockRuntimeClient(config);\n\t\t\tconst cacheRetention = resolveCacheRetention(options.cacheRetention);\n\t\t\tlet commandInput = {\n\t\t\t\tmodelId: model.id,\n\t\t\t\tmessages: convertMessages(context, model, cacheRetention),\n\t\t\t\tsystem: buildSystemPrompt(context.systemPrompt, model, cacheRetention),\n\t\t\t\tinferenceConfig: {\n\t\t\t\t\t...(options.maxTokens !== undefined && { maxTokens: options.maxTokens }),\n\t\t\t\t\t...(options.temperature !== undefined && { temperature: options.temperature }),\n\t\t\t\t},\n\t\t\t\ttoolConfig: convertToolConfig(context.tools, options.toolChoice),\n\t\t\t\tadditionalModelRequestFields: buildAdditionalModelRequestFields(model, options),\n\t\t\t\t...(options.requestMetadata !== undefined && { requestMetadata: options.requestMetadata }),\n\t\t\t};\n\t\t\tconst nextCommandInput = await options?.onPayload?.(commandInput, model);\n\t\t\tif (nextCommandInput !== undefined) {\n\t\t\t\tcommandInput = nextCommandInput as typeof commandInput;\n\t\t\t}\n\t\t\tconst command = new ConverseStreamCommand(commandInput);\n\n\t\t\tconst response = await client.send(command, { abortSignal: options.signal });\n\t\t\tif (response.$metadata.httpStatusCode !== undefined) {\n\t\t\t\tconst responseHeaders: Record<string, string> = {};\n\t\t\t\tif (response.$metadata.requestId) {\n\t\t\t\t\tresponseHeaders[\"x-amzn-requestid\"] = response.$metadata.requestId;\n\t\t\t\t}\n\t\t\t\tawait options?.onResponse?.({ status: response.$metadata.httpStatusCode, headers: responseHeaders }, model);\n\t\t\t}\n\n\t\t\tfor await (const item of response.stream!) {\n\t\t\t\tif (item.messageStart) {\n\t\t\t\t\tif (item.messageStart.role !== ConversationRole.ASSISTANT) {\n\t\t\t\t\t\tthrow new Error(\"Unexpected assistant message start but got user message start instead\");\n\t\t\t\t\t}\n\t\t\t\t\tstream.push({ type: \"start\", partial: output });\n\t\t\t\t} else if (item.contentBlockStart) {\n\t\t\t\t\thandleContentBlockStart(item.contentBlockStart, blocks, output, stream);\n\t\t\t\t} else if (item.contentBlockDelta) {\n\t\t\t\t\thandleContentBlockDelta(item.contentBlockDelta, blocks, output, stream);\n\t\t\t\t} else if (item.contentBlockStop) {\n\t\t\t\t\thandleContentBlockStop(item.contentBlockStop, blocks, output, stream);\n\t\t\t\t} else if (item.messageStop) {\n\t\t\t\t\toutput.stopReason = mapStopReason(item.messageStop.stopReason);\n\t\t\t\t} else if (item.metadata) {\n\t\t\t\t\thandleMetadata(item.metadata, model, output);\n\t\t\t\t} else if (item.internalServerException) {\n\t\t\t\t\tthrow item.internalServerException;\n\t\t\t\t} else if (item.modelStreamErrorException) {\n\t\t\t\t\tthrow item.modelStreamErrorException;\n\t\t\t\t} else if (item.validationException) {\n\t\t\t\t\tthrow item.validationException;\n\t\t\t\t} else if (item.throttlingException) {\n\t\t\t\t\tthrow item.throttlingException;\n\t\t\t\t} else if (item.serviceUnavailableException) {\n\t\t\t\t\tthrow item.serviceUnavailableException;\n\t\t\t\t}\n\t\t\t}\n\n\t\t\tif (options.signal?.aborted) {\n\t\t\t\tthrow new Error(\"Request was aborted\");\n\t\t\t}\n\n\t\t\tif (output.stopReason === \"error\" || output.stopReason === \"aborted\") {\n\t\t\t\tthrow new Error(\"An unknown error occurred\");\n\t\t\t}\n\n\t\t\tstream.push({ type: \"done\", reason: output.stopReason, message: output });\n\t\t\tstream.end();\n\t\t} catch (error) {\n\t\t\tfor (const block of output.content) {\n\t\t\t\tdelete (block as Block).index;\n\t\t\t\t// partialJson is only a streaming scratch buffer; never persist it.\n\t\t\t\tdelete (block as Block).partialJson;\n\t\t\t}\n\t\t\toutput.stopReason = options.signal?.aborted ? \"aborted\" : \"error\";\n\t\t\toutput.errorMessage = formatBedrockError(error);\n\t\t\tstream.push({ type: \"error\", reason: output.stopReason, error: output });\n\t\t\tstream.end();\n\t\t}\n\t})();\n\n\treturn stream;\n};\n\n/**\n * Human-readable prefixes for Bedrock SDK exception names.\n * The downstream retry logic in agent-session matches patterns like\n * `server.?error` and `service.?unavailable`, so we preserve the legacy\n * prefix format rather than using the raw SDK exception name.\n */\nconst BEDROCK_ERROR_PREFIXES: Record<string, string> = {\n\tInternalServerException: \"Internal server error\",\n\tModelStreamErrorException: \"Model stream error\",\n\tValidationException: \"Validation error\",\n\tThrottlingException: \"Throttling error\",\n\tServiceUnavailableException: \"Service unavailable\",\n};\n\n/**\n * Format a Bedrock error with a human-readable prefix.\n * AWS SDK exceptions (both from `client.send()` and from stream event items)\n * extend BedrockRuntimeServiceException. We map the `.name` to a stable\n * human-readable prefix so downstream consumers (retry logic, context-overflow\n * detection) can distinguish error categories via simple string matching.\n */\nfunction formatBedrockError(error: unknown): string {\n\tconst message = error instanceof Error ? error.message : JSON.stringify(error);\n\tif (error instanceof BedrockRuntimeServiceException) {\n\t\tconst prefix = BEDROCK_ERROR_PREFIXES[error.name] ?? error.name;\n\t\treturn `${prefix}: ${message}`;\n\t}\n\treturn message;\n}\n\nexport const streamSimpleBedrock: StreamFunction<\"bedrock-converse-stream\", SimpleStreamOptions> = (\n\tmodel: Model<\"bedrock-converse-stream\">,\n\tcontext: Context,\n\toptions?: SimpleStreamOptions,\n): AssistantMessageEventStream => {\n\tconst base = buildBaseOptions(model, options, undefined);\n\tif (!options?.reasoning) {\n\t\treturn streamBedrock(model, context, { ...base, reasoning: undefined } satisfies BedrockOptions);\n\t}\n\n\tif (isAnthropicClaudeModel(model)) {\n\t\tif (supportsAdaptiveThinking(model.id, model.name)) {\n\t\t\treturn streamBedrock(model, context, {\n\t\t\t\t...base,\n\t\t\t\treasoning: options.reasoning,\n\t\t\t\tthinkingBudgets: options.thinkingBudgets,\n\t\t\t} satisfies BedrockOptions);\n\t\t}\n\n\t\tconst adjusted = adjustMaxTokensForThinking(\n\t\t\tbase.maxTokens || 0,\n\t\t\tmodel.maxTokens,\n\t\t\toptions.reasoning,\n\t\t\toptions.thinkingBudgets,\n\t\t);\n\n\t\treturn streamBedrock(model, context, {\n\t\t\t...base,\n\t\t\tmaxTokens: adjusted.maxTokens,\n\t\t\treasoning: options.reasoning,\n\t\t\tthinkingBudgets: {\n\t\t\t\t...(options.thinkingBudgets || {}),\n\t\t\t\t[clampReasoning(options.reasoning)!]: adjusted.thinkingBudget,\n\t\t\t},\n\t\t} satisfies BedrockOptions);\n\t}\n\n\treturn streamBedrock(model, context, {\n\t\t...base,\n\t\treasoning: options.reasoning,\n\t\tthinkingBudgets: options.thinkingBudgets,\n\t} satisfies BedrockOptions);\n};\n\nfunction handleContentBlockStart(\n\tevent: ContentBlockStartEvent,\n\tblocks: Block[],\n\toutput: AssistantMessage,\n\tstream: AssistantMessageEventStream,\n): void {\n\tconst index = event.contentBlockIndex!;\n\tconst start = event.start;\n\n\tif (start?.toolUse) {\n\t\tconst block: Block = {\n\t\t\ttype: \"toolCall\",\n\t\t\tid: start.toolUse.toolUseId || \"\",\n\t\t\tname: start.toolUse.name || \"\",\n\t\t\targuments: {},\n\t\t\tpartialJson: \"\",\n\t\t\tindex,\n\t\t};\n\t\toutput.content.push(block);\n\t\tstream.push({ type: \"toolcall_start\", contentIndex: blocks.length - 1, partial: output });\n\t}\n}\n\nfunction handleContentBlockDelta(\n\tevent: ContentBlockDeltaEvent,\n\tblocks: Block[],\n\toutput: AssistantMessage,\n\tstream: AssistantMessageEventStream,\n): void {\n\tconst contentBlockIndex = event.contentBlockIndex!;\n\tconst delta = event.delta;\n\tlet index = blocks.findIndex((b) => b.index === contentBlockIndex);\n\tlet block = blocks[index];\n\n\tif (delta?.text !== undefined) {\n\t\t// If no text block exists yet, create one, as `handleContentBlockStart` is not sent for text blocks\n\t\tif (!block) {\n\t\t\tconst newBlock: Block = { type: \"text\", text: \"\", index: contentBlockIndex };\n\t\t\toutput.content.push(newBlock);\n\t\t\tindex = blocks.length - 1;\n\t\t\tblock = blocks[index];\n\t\t\tstream.push({ type: \"text_start\", contentIndex: index, partial: output });\n\t\t}\n\t\tif (block.type === \"text\") {\n\t\t\tblock.text += delta.text;\n\t\t\tstream.push({ type: \"text_delta\", contentIndex: index, delta: delta.text, partial: output });\n\t\t}\n\t} else if (delta?.toolUse && block?.type === \"toolCall\") {\n\t\tblock.partialJson = (block.partialJson || \"\") + (delta.toolUse.input || \"\");\n\t\tblock.arguments = parseStreamingJson(block.partialJson);\n\t\tstream.push({ type: \"toolcall_delta\", contentIndex: index, delta: delta.toolUse.input || \"\", partial: output });\n\t} else if (delta?.reasoningContent) {\n\t\tlet thinkingBlock = block;\n\t\tlet thinkingIndex = index;\n\n\t\tif (!thinkingBlock) {\n\t\t\tconst newBlock: Block = { type: \"thinking\", thinking: \"\", thinkingSignature: \"\", index: contentBlockIndex };\n\t\t\toutput.content.push(newBlock);\n\t\t\tthinkingIndex = blocks.length - 1;\n\t\t\tthinkingBlock = blocks[thinkingIndex];\n\t\t\tstream.push({ type: \"thinking_start\", contentIndex: thinkingIndex, partial: output });\n\t\t}\n\n\t\tif (thinkingBlock?.type === \"thinking\") {\n\t\t\tif (delta.reasoningContent.text) {\n\t\t\t\tthinkingBlock.thinking += delta.reasoningContent.text;\n\t\t\t\tstream.push({\n\t\t\t\t\ttype: \"thinking_delta\",\n\t\t\t\t\tcontentIndex: thinkingIndex,\n\t\t\t\t\tdelta: delta.reasoningContent.text,\n\t\t\t\t\tpartial: output,\n\t\t\t\t});\n\t\t\t}\n\t\t\tif (delta.reasoningContent.signature) {\n\t\t\t\tthinkingBlock.thinkingSignature =\n\t\t\t\t\t(thinkingBlock.thinkingSignature || \"\") + delta.reasoningContent.signature;\n\t\t\t}\n\t\t}\n\t}\n}\n\nfunction handleMetadata(\n\tevent: ConverseStreamMetadataEvent,\n\tmodel: Model<\"bedrock-converse-stream\">,\n\toutput: AssistantMessage,\n): void {\n\tif (event.usage) {\n\t\toutput.usage.input = event.usage.inputTokens || 0;\n\t\toutput.usage.output = event.usage.outputTokens || 0;\n\t\toutput.usage.cacheRead = event.usage.cacheReadInputTokens || 0;\n\t\toutput.usage.cacheWrite = event.usage.cacheWriteInputTokens || 0;\n\t\toutput.usage.totalTokens = event.usage.totalTokens || output.usage.input + output.usage.output;\n\t\tcalculateCost(model, output.usage);\n\t}\n}\n\nfunction handleContentBlockStop(\n\tevent: ContentBlockStopEvent,\n\tblocks: Block[],\n\toutput: AssistantMessage,\n\tstream: AssistantMessageEventStream,\n): void {\n\tconst index = blocks.findIndex((b) => b.index === event.contentBlockIndex);\n\tconst block = blocks[index];\n\tif (!block) return;\n\tdelete (block as Block).index;\n\n\tswitch (block.type) {\n\t\tcase \"text\":\n\t\t\tstream.push({ type: \"text_end\", contentIndex: index, content: block.text, partial: output });\n\t\t\tbreak;\n\t\tcase \"thinking\":\n\t\t\tstream.push({ type: \"thinking_end\", contentIndex: index, content: block.thinking, partial: output });\n\t\t\tbreak;\n\t\tcase \"toolCall\":\n\t\t\tblock.arguments = parseStreamingJson(block.partialJson);\n\t\t\t// Finalize in-place and strip the scratch buffer so replay only\n\t\t\t// carries parsed arguments.\n\t\t\tdelete (block as Block).partialJson;\n\t\t\tstream.push({ type: \"toolcall_end\", contentIndex: index, toolCall: block, partial: output });\n\t\t\tbreak;\n\t}\n}\n\n/**\n * Check if the model supports adaptive thinking (Opus 4.6+, Sonnet 4.6).\n * Checks both model ID and model name to support application inference profiles\n * whose ARNs don't contain the model name.\n */\nfunction getModelMatchCandidates(modelId: string, modelName?: string): string[] {\n\tconst values = modelName ? [modelId, modelName] : [modelId];\n\treturn values.flatMap((value) => {\n\t\tconst lower = value.toLowerCase();\n\t\treturn [lower, lower.replace(/[\\s_.:]+/g, \"-\")];\n\t});\n}\n\nfunction supportsAdaptiveThinking(modelId: string, modelName?: string): boolean {\n\tconst candidates = getModelMatchCandidates(modelId, modelName);\n\treturn candidates.some(\n\t\t(s) => s.includes(\"opus-4-6\") || s.includes(\"opus-4-7\") || s.includes(\"opus-4-8\") || s.includes(\"sonnet-4-6\"),\n\t);\n}\n\nfunction supportsNativeXhighEffort(model: Model<\"bedrock-converse-stream\">): boolean {\n\tconst candidates = getModelMatchCandidates(model.id, model.name);\n\treturn candidates.some((s) => s.includes(\"opus-4-7\") || s.includes(\"opus-4-8\"));\n}\n\nfunction mapThinkingLevelToEffort(\n\tmodel: Model<\"bedrock-converse-stream\">,\n\tlevel: SimpleStreamOptions[\"reasoning\"],\n): \"low\" | \"medium\" | \"high\" | \"xhigh\" | \"max\" {\n\tif (level === \"xhigh\" && supportsNativeXhighEffort(model)) return \"xhigh\";\n\n\tconst mapped = level ? model.thinkingLevelMap?.[level] : undefined;\n\tif (typeof mapped === \"string\") return mapped as \"low\" | \"medium\" | \"high\" | \"xhigh\" | \"max\";\n\n\tconst bedrockThinkingLevelToEffortMap: Record<string, \"low\" | \"medium\" | \"high\"> = {\n\t\tminimal: \"low\",\n\t\tlow: \"low\",\n\t\tmedium: \"medium\",\n\t\thigh: \"high\",\n\t};\n\treturn level ? (bedrockThinkingLevelToEffortMap[level] ?? \"high\") : \"high\";\n}\n\n/**\n * Check if the model is an Anthropic Claude model on Bedrock.\n * Checks both model ID and model name to support application inference profiles\n * whose ARNs don't contain the model name.\n */\nfunction isAnthropicClaudeModel(model: Model<\"bedrock-converse-stream\">): boolean {\n\tconst id = model.id.toLowerCase();\n\tconst name = model.name?.toLowerCase() ?? \"\";\n\treturn (\n\t\tid.includes(\"anthropic.claude\") ||\n\t\tid.includes(\"anthropic/claude\") ||\n\t\tname.includes(\"anthropic.claude\") ||\n\t\tname.includes(\"anthropic/claude\") ||\n\t\tname.includes(\"claude\")\n\t);\n}\n\n/**\n * Check if the model supports prompt caching.\n * Supported: Claude 3.5 Haiku, Claude 3.7 Sonnet, Claude 4.x models\n *\n * For base models and system-defined inference profiles the model ID / ARN\n * contains the model name, so we can decide locally.\n *\n * For application inference profiles (whose ARNs don't contain the model name),\n * also checks model.name which is user-controlled via models.json or registerProvider.\n * As a last resort, set AWS_BEDROCK_FORCE_CACHE=1 to enable cache points.\n * Amazon Nova models have automatic caching and don't need explicit cache points.\n */\nfunction supportsPromptCaching(model: Model<\"bedrock-converse-stream\">): boolean {\n\tconst candidates = getModelMatchCandidates(model.id, model.name);\n\n\tconst hasClaudeRef = candidates.some((s) => s.includes(\"claude\"));\n\tif (!hasClaudeRef) {\n\t\t// Application inference profiles don't contain the model name in the ARN.\n\t\t// Allow users to force cache points via environment variable.\n\t\tif (typeof process !== \"undefined\" && process.env.AWS_BEDROCK_FORCE_CACHE === \"1\") return true;\n\t\treturn false;\n\t}\n\t// Claude 4.x models (opus-4, sonnet-4, haiku-4)\n\tif (candidates.some((s) => s.includes(\"-4-\"))) return true;\n\t// Claude 3.7 Sonnet\n\tif (candidates.some((s) => s.includes(\"claude-3-7-sonnet\"))) return true;\n\t// Claude 3.5 Haiku\n\tif (candidates.some((s) => s.includes(\"claude-3-5-haiku\"))) return true;\n\treturn false;\n}\n\n/**\n * Check if the model supports thinking signatures in reasoningContent.\n * Only Anthropic Claude models support the signature field.\n * Other models (OpenAI, Qwen, Minimax, Moonshot, etc.) reject it with:\n * \"This model doesn't support the reasoningContent.reasoningText.signature field\"\n *\n * Checks both model ID and model name to support application inference profiles.\n */\nfunction supportsThinkingSignature(model: Model<\"bedrock-converse-stream\">): boolean {\n\treturn isAnthropicClaudeModel(model);\n}\n\nfunction buildSystemPrompt(\n\tsystemPrompt: string | undefined,\n\tmodel: Model<\"bedrock-converse-stream\">,\n\tcacheRetention: CacheRetention,\n): SystemContentBlock[] | undefined {\n\tif (!systemPrompt) return undefined;\n\n\tconst blocks: SystemContentBlock[] = [{ text: sanitizeSurrogates(systemPrompt) }];\n\n\t// Add cache point for supported Claude models when caching is enabled\n\tif (cacheRetention !== \"none\" && supportsPromptCaching(model)) {\n\t\tblocks.push({\n\t\t\tcachePoint: { type: CachePointType.DEFAULT, ...(cacheRetention === \"long\" ? { ttl: CacheTTL.ONE_HOUR } : {}) },\n\t\t});\n\t}\n\n\treturn blocks;\n}\n\nfunction normalizeToolCallId(id: string): string {\n\tconst sanitized = id.replace(/[^a-zA-Z0-9_-]/g, \"_\");\n\treturn sanitized.length > 64 ? sanitized.slice(0, 64) : sanitized;\n}\n\nfunction convertMessages(\n\tcontext: Context,\n\tmodel: Model<\"bedrock-converse-stream\">,\n\tcacheRetention: CacheRetention,\n): Message[] {\n\tconst result: Message[] = [];\n\tconst transformedMessages = transformMessages(context.messages, model, normalizeToolCallId);\n\n\tfor (let i = 0; i < transformedMessages.length; i++) {\n\t\tconst m = transformedMessages[i];\n\n\t\tswitch (m.role) {\n\t\t\tcase \"user\":\n\t\t\t\tresult.push({\n\t\t\t\t\trole: ConversationRole.USER,\n\t\t\t\t\tcontent:\n\t\t\t\t\t\ttypeof m.content === \"string\"\n\t\t\t\t\t\t\t? [{ text: sanitizeSurrogates(m.content) }]\n\t\t\t\t\t\t\t: m.content.map((c) => {\n\t\t\t\t\t\t\t\t\tswitch (c.type) {\n\t\t\t\t\t\t\t\t\t\tcase \"text\":\n\t\t\t\t\t\t\t\t\t\t\treturn { text: sanitizeSurrogates(c.text) };\n\t\t\t\t\t\t\t\t\t\tcase \"image\":\n\t\t\t\t\t\t\t\t\t\t\treturn { image: createImageBlock(c.mimeType, c.data) };\n\t\t\t\t\t\t\t\t\t\tdefault:\n\t\t\t\t\t\t\t\t\t\t\tthrow new Error(\"Unknown user content type\");\n\t\t\t\t\t\t\t\t\t}\n\t\t\t\t\t\t\t\t}),\n\t\t\t\t});\n\t\t\t\tbreak;\n\t\t\tcase \"assistant\": {\n\t\t\t\t// Skip assistant messages with empty content (e.g., from aborted requests)\n\t\t\t\t// Bedrock rejects messages with empty content arrays\n\t\t\t\tif (m.content.length === 0) {\n\t\t\t\t\tcontinue;\n\t\t\t\t}\n\t\t\t\tconst contentBlocks: ContentBlock[] = [];\n\t\t\t\tfor (const c of m.content) {\n\t\t\t\t\tswitch (c.type) {\n\t\t\t\t\t\tcase \"text\":\n\t\t\t\t\t\t\t// Skip empty text blocks\n\t\t\t\t\t\t\tif (c.text.trim().length === 0) continue;\n\t\t\t\t\t\t\tcontentBlocks.push({ text: sanitizeSurrogates(c.text) });\n\t\t\t\t\t\t\tbreak;\n\t\t\t\t\t\tcase \"toolCall\":\n\t\t\t\t\t\t\tcontentBlocks.push({\n\t\t\t\t\t\t\t\ttoolUse: { toolUseId: c.id, name: c.name, input: c.arguments },\n\t\t\t\t\t\t\t});\n\t\t\t\t\t\t\tbreak;\n\t\t\t\t\t\tcase \"thinking\":\n\t\t\t\t\t\t\t// Skip empty thinking blocks\n\t\t\t\t\t\t\tif (c.thinking.trim().length === 0) continue;\n\t\t\t\t\t\t\t// Only Anthropic models support the signature field in reasoningText.\n\t\t\t\t\t\t\t// For other models, we omit the signature to avoid errors like:\n\t\t\t\t\t\t\t// \"This model doesn't support the reasoningContent.reasoningText.signature field\"\n\t\t\t\t\t\t\tif (supportsThinkingSignature(model)) {\n\t\t\t\t\t\t\t\t// Signatures arrive after thinking deltas. If a partial or externally\n\t\t\t\t\t\t\t\t// persisted message lacks a signature, Bedrock rejects the replayed\n\t\t\t\t\t\t\t\t// reasoning block. Fall back to plain text, matching Anthropic.\n\t\t\t\t\t\t\t\tif (!c.thinkingSignature || c.thinkingSignature.trim().length === 0) {\n\t\t\t\t\t\t\t\t\tcontentBlocks.push({ text: sanitizeSurrogates(c.thinking) });\n\t\t\t\t\t\t\t\t} else {\n\t\t\t\t\t\t\t\t\tcontentBlocks.push({\n\t\t\t\t\t\t\t\t\t\treasoningContent: {\n\t\t\t\t\t\t\t\t\t\t\treasoningText: {\n\t\t\t\t\t\t\t\t\t\t\t\ttext: sanitizeSurrogates(c.thinking),\n\t\t\t\t\t\t\t\t\t\t\t\tsignature: c.thinkingSignature,\n\t\t\t\t\t\t\t\t\t\t\t},\n\t\t\t\t\t\t\t\t\t\t},\n\t\t\t\t\t\t\t\t\t});\n\t\t\t\t\t\t\t\t}\n\t\t\t\t\t\t\t} else {\n\t\t\t\t\t\t\t\tcontentBlocks.push({\n\t\t\t\t\t\t\t\t\treasoningContent: {\n\t\t\t\t\t\t\t\t\t\treasoningText: { text: sanitizeSurrogates(c.thinking) },\n\t\t\t\t\t\t\t\t\t},\n\t\t\t\t\t\t\t\t});\n\t\t\t\t\t\t\t}\n\t\t\t\t\t\t\tbreak;\n\t\t\t\t\t\tdefault:\n\t\t\t\t\t\t\tthrow new Error(\"Unknown assistant content type\");\n\t\t\t\t\t}\n\t\t\t\t}\n\t\t\t\t// Skip if all content blocks were filtered out\n\t\t\t\tif (contentBlocks.length === 0) {\n\t\t\t\t\tcontinue;\n\t\t\t\t}\n\t\t\t\tresult.push({\n\t\t\t\t\trole: ConversationRole.ASSISTANT,\n\t\t\t\t\tcontent: contentBlocks,\n\t\t\t\t});\n\t\t\t\tbreak;\n\t\t\t}\n\t\t\tcase \"toolResult\": {\n\t\t\t\t// Collect all consecutive toolResult messages into a single user message\n\t\t\t\t// Bedrock requires all tool results to be in one message\n\t\t\t\tconst toolResults: ContentBlock.ToolResultMember[] = [];\n\n\t\t\t\t// Add current tool result with all content blocks combined\n\t\t\t\ttoolResults.push({\n\t\t\t\t\ttoolResult: {\n\t\t\t\t\t\ttoolUseId: m.toolCallId,\n\t\t\t\t\t\tcontent: m.content.map((c) =>\n\t\t\t\t\t\t\tc.type === \"image\"\n\t\t\t\t\t\t\t\t? { image: createImageBlock(c.mimeType, c.data) }\n\t\t\t\t\t\t\t\t: { text: sanitizeSurrogates(c.text) },\n\t\t\t\t\t\t),\n\t\t\t\t\t\tstatus: m.isError ? ToolResultStatus.ERROR : ToolResultStatus.SUCCESS,\n\t\t\t\t\t},\n\t\t\t\t});\n\n\t\t\t\t// Look ahead for consecutive toolResult messages\n\t\t\t\tlet j = i + 1;\n\t\t\t\twhile (j < transformedMessages.length && transformedMessages[j].role === \"toolResult\") {\n\t\t\t\t\tconst nextMsg = transformedMessages[j] as ToolResultMessage;\n\t\t\t\t\ttoolResults.push({\n\t\t\t\t\t\ttoolResult: {\n\t\t\t\t\t\t\ttoolUseId: nextMsg.toolCallId,\n\t\t\t\t\t\t\tcontent: nextMsg.content.map((c) =>\n\t\t\t\t\t\t\t\tc.type === \"image\"\n\t\t\t\t\t\t\t\t\t? { image: createImageBlock(c.mimeType, c.data) }\n\t\t\t\t\t\t\t\t\t: { text: sanitizeSurrogates(c.text) },\n\t\t\t\t\t\t\t),\n\t\t\t\t\t\t\tstatus: nextMsg.isError ? ToolResultStatus.ERROR : ToolResultStatus.SUCCESS,\n\t\t\t\t\t\t},\n\t\t\t\t\t});\n\t\t\t\t\tj++;\n\t\t\t\t}\n\n\t\t\t\t// Skip the messages we've already processed\n\t\t\t\ti = j - 1;\n\n\t\t\t\tresult.push({\n\t\t\t\t\trole: ConversationRole.USER,\n\t\t\t\t\tcontent: toolResults,\n\t\t\t\t});\n\t\t\t\tbreak;\n\t\t\t}\n\t\t\tdefault:\n\t\t\t\tthrow new Error(\"Unknown message role\");\n\t\t}\n\t}\n\n\t// Add cache point to the last user message for supported Claude models when caching is enabled\n\tif (cacheRetention !== \"none\" && supportsPromptCaching(model) && result.length > 0) {\n\t\tconst lastMessage = result[result.length - 1];\n\t\tif (lastMessage.role === ConversationRole.USER && lastMessage.content) {\n\t\t\t(lastMessage.content as ContentBlock[]).push({\n\t\t\t\tcachePoint: {\n\t\t\t\t\ttype: CachePointType.DEFAULT,\n\t\t\t\t\t...(cacheRetention === \"long\" ? { ttl: CacheTTL.ONE_HOUR } : {}),\n\t\t\t\t},\n\t\t\t});\n\t\t}\n\t}\n\n\treturn result;\n}\n\nfunction convertToolConfig(\n\ttools: Tool[] | undefined,\n\ttoolChoice: BedrockOptions[\"toolChoice\"],\n): ToolConfiguration | undefined {\n\tif (!tools?.length || toolChoice === \"none\") return undefined;\n\n\tconst bedrockTools: BedrockTool[] = tools.map((tool) => ({\n\t\ttoolSpec: {\n\t\t\tname: tool.name,\n\t\t\tdescription: tool.description,\n\t\t\tinputSchema: { json: tool.parameters as unknown as DocumentType },\n\t\t},\n\t}));\n\n\tlet bedrockToolChoice: ToolChoice | undefined;\n\tswitch (toolChoice) {\n\t\tcase \"auto\":\n\t\t\tbedrockToolChoice = { auto: {} };\n\t\t\tbreak;\n\t\tcase \"any\":\n\t\t\tbedrockToolChoice = { any: {} };\n\t\t\tbreak;\n\t\tdefault:\n\t\t\tif (toolChoice?.type === \"tool\") {\n\t\t\t\tbedrockToolChoice = { tool: { name: toolChoice.name } };\n\t\t\t}\n\t}\n\n\treturn { tools: bedrockTools, toolChoice: bedrockToolChoice };\n}\n\nfunction mapStopReason(reason: string | undefined): StopReason {\n\tconst bedrockStopReasonMap: Record<string, StopReason> = {\n\t\t[BedrockStopReason.END_TURN]: \"stop\",\n\t\t[BedrockStopReason.STOP_SEQUENCE]: \"stop\",\n\t\t[BedrockStopReason.MAX_TOKENS]: \"length\",\n\t\t[BedrockStopReason.MODEL_CONTEXT_WINDOW_EXCEEDED]: \"length\",\n\t\t[BedrockStopReason.TOOL_USE]: \"toolUse\",\n\t};\n\treturn reason ? (bedrockStopReasonMap[reason] ?? \"error\") : \"error\";\n}\n\nfunction getConfiguredBedrockRegion(options: BedrockOptions): string | undefined {\n\tif (typeof process === \"undefined\") {\n\t\treturn options.region;\n\t}\n\n\treturn options.region || process.env.AWS_REGION || process.env.AWS_DEFAULT_REGION || undefined;\n}\n\nfunction hasConfiguredBedrockProfile(): boolean {\n\tif (typeof process === \"undefined\") {\n\t\treturn false;\n\t}\n\n\treturn Boolean(process.env.AWS_PROFILE);\n}\n\nfunction getStandardBedrockEndpointRegion(baseUrl: string | undefined): string | undefined {\n\tif (!baseUrl) {\n\t\treturn undefined;\n\t}\n\n\ttry {\n\t\tconst { hostname } = new URL(baseUrl);\n\t\tconst match = hostname.toLowerCase().match(/^bedrock-runtime(?:-fips)?\\.([a-z0-9-]+)\\.amazonaws\\.com(?:\\.cn)?$/);\n\t\treturn match?.[1];\n\t} catch {\n\t\treturn undefined;\n\t}\n}\n\nfunction shouldUseExplicitBedrockEndpoint(\n\tbaseUrl: string,\n\tconfiguredRegion: string | undefined,\n\thasConfiguredProfile: boolean,\n): boolean {\n\tconst endpointRegion = getStandardBedrockEndpointRegion(baseUrl);\n\tif (!endpointRegion) {\n\t\treturn true;\n\t}\n\n\treturn !configuredRegion && !hasConfiguredProfile;\n}\n\nfunction isGovCloudBedrockTarget(model: Model<\"bedrock-converse-stream\">, options: BedrockOptions): boolean {\n\tconst region = getConfiguredBedrockRegion(options);\n\tif (region?.toLowerCase().startsWith(\"us-gov-\")) {\n\t\treturn true;\n\t}\n\n\tconst modelId = model.id.toLowerCase();\n\treturn modelId.startsWith(\"us-gov.\") || modelId.startsWith(\"arn:aws-us-gov:\");\n}\n\nfunction buildAdditionalModelRequestFields(\n\tmodel: Model<\"bedrock-converse-stream\">,\n\toptions: BedrockOptions,\n): Record<string, any> | undefined {\n\tif (!options.reasoning || !model.reasoning) {\n\t\treturn undefined;\n\t}\n\n\tif (isAnthropicClaudeModel(model)) {\n\t\t// GovCloud Bedrock currently rejects the Claude thinking.display field.\n\t\t// Omit it there until the GovCloud Converse schema catches up.\n\t\tconst display = isGovCloudBedrockTarget(model, options) ? undefined : (options.thinkingDisplay ?? \"summarized\");\n\t\tconst result: Record<string, any> = supportsAdaptiveThinking(model.id, model.name)\n\t\t\t? {\n\t\t\t\t\tthinking: { type: \"adaptive\", ...(display !== undefined ? { display } : {}) },\n\t\t\t\t\toutput_config: { effort: mapThinkingLevelToEffort(model, options.reasoning) },\n\t\t\t\t}\n\t\t\t: (() => {\n\t\t\t\t\tconst defaultBudgets: Record<ThinkingLevel, number> = {\n\t\t\t\t\t\tminimal: 1024,\n\t\t\t\t\t\tlow: 2048,\n\t\t\t\t\t\tmedium: 8192,\n\t\t\t\t\t\thigh: 16384,\n\t\t\t\t\t\txhigh: 16384, // Claude doesn't support xhigh, clamp to high\n\t\t\t\t\t};\n\n\t\t\t\t\t// Custom budgets override defaults (xhigh not in ThinkingBudgets, use high)\n\t\t\t\t\tconst level = options.reasoning === \"xhigh\" ? \"high\" : options.reasoning;\n\t\t\t\t\tconst budget = options.thinkingBudgets?.[level] ?? defaultBudgets[options.reasoning];\n\n\t\t\t\t\treturn {\n\t\t\t\t\t\tthinking: {\n\t\t\t\t\t\t\ttype: \"enabled\",\n\t\t\t\t\t\t\tbudget_tokens: budget,\n\t\t\t\t\t\t\t...(display !== undefined ? { display } : {}),\n\t\t\t\t\t\t},\n\t\t\t\t\t};\n\t\t\t\t})();\n\n\t\tif (!supportsAdaptiveThinking(model.id, model.name) && (options.interleavedThinking ?? true)) {\n\t\t\tresult.anthropic_beta = [\"interleaved-thinking-2025-05-14\"];\n\t\t}\n\n\t\treturn result;\n\t}\n\n\treturn undefined;\n}\n\nfunction createImageBlock(mimeType: string, data: string) {\n\tconst bedrockImageFormatMap: Record<string, ImageFormat> = {\n\t\t\"image/jpeg\": ImageFormat.JPEG,\n\t\t\"image/jpg\": ImageFormat.JPEG,\n\t\t\"image/png\": ImageFormat.PNG,\n\t\t\"image/gif\": ImageFormat.GIF,\n\t\t\"image/webp\": ImageFormat.WEBP,\n\t};\n\tconst format = bedrockImageFormatMap[mimeType];\n\tif (!format) {\n\t\tthrow new Error(`Unknown image type: ${mimeType}`);\n\t}\n\n\tconst binaryString = atob(data);\n\tconst bytes = new Uint8Array(binaryString.length);\n\tfor (let i = 0; i < binaryString.length; i++) {\n\t\tbytes[i] = binaryString.charCodeAt(i);\n\t}\n\n\treturn { source: { bytes }, format };\n}\n"]}