@leanandmean/ai 0.74.1-scramjet.46 → 0.74.1-scramjet.47

package/dist/providers/anthropic.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"anthropic.d.ts","sourceRoot":"","sources":["../../src/providers/anthropic.ts"],"names":[],"mappings":"AAAA,OAAO,SAAS,MAAM,mBAAmB,CAAC;AAW1C,OAAO,KAAK,EASX,mBAAmB,EAEnB,cAAc,EACd,aAAa,EAOb,MAAM,aAAa,CAAC;AAqJrB,MAAM,MAAM,eAAe,GAAG,KAAK,GAAG,QAAQ,GAAG,MAAM,GAAG,OAAO,GAAG,KAAK,CAAC;AAE1E,MAAM,MAAM,wBAAwB,GAAG,YAAY,GAAG,SAAS,CAAC;AAsBhE,MAAM,WAAW,gBAAiB,SAAQ,aAAa;IACtD;;;;OAIG;IACH,eAAe,CAAC,EAAE,OAAO,CAAC;IAC1B;;;OAGG;IACH,oBAAoB,CAAC,EAAE,MAAM,CAAC;IAC9B;;;;;;;;;OASG;IACH,MAAM,CAAC,EAAE,eAAe,CAAC;IACzB;;;;;;;;;;OAUG;IACH,eAAe,CAAC,EAAE,wBAAwB,CAAC;IAC3C,mBAAmB,CAAC,EAAE,OAAO,CAAC;IAC9B,UAAU,CAAC,EAAE,MAAM,GAAG,KAAK,GAAG,MAAM,GAAG;QAAE,IAAI,EAAE,MAAM,CAAC;QAAC,IAAI,EAAE,MAAM,CAAA;KAAE,CAAC;IACtE;;;;OAIG;IACH,MAAM,CAAC,EAAE,SAAS,CAAC;CACnB;AAiPD,eAAO,MAAM,eAAe,EAAE,cAAc,CAAC,oBAAoB,EAAE,gBAAgB,CA0RlF,CAAC;AAwCF,eAAO,MAAM,qBAAqB,EAAE,cAAc,CAAC,oBAAoB,EAAE,mBAAmB,CA2C3F,CAAC","sourcesContent":["import Anthropic from \"@anthropic-ai/sdk\";\nimport type {\n\tCacheControlEphemeral,\n\tContentBlockParam,\n\tMessageCreateParamsStreaming,\n\tMessageParam,\n\tRawMessageStreamEvent,\n\tTextBlockParam,\n} from \"@anthropic-ai/sdk/resources/messages.js\";\nimport { getEnvApiKey } from \"../env-api-keys.js\";\nimport { calculateCost } from \"../models.js\";\nimport type {\n\tAnthropicMessagesCompat,\n\tApi,\n\tAssistantMessage,\n\tCacheRetention,\n\tContext,\n\tImageContent,\n\tMessage,\n\tModel,\n\tSimpleStreamOptions,\n\tStopReason,\n\tStreamFunction,\n\tStreamOptions,\n\tSystemPromptSection,\n\tTextContent,\n\tThinkingContent,\n\tTool,\n\tToolCall,\n\tToolResultMessage,\n} from \"../types.js\";\nimport { AssistantMessageEventStream } from \"../utils/event-stream.js\";\nimport {\n\tappendBuiltinFailure,\n\tfailureFromProviderError,\n\tinvokeProviderCallback,\n\tRequestFailureError,\n} from \"../utils/failure-evidence.js\";\nimport { headersToRecord } from \"../utils/headers.js\";\nimport { parseJsonWithRepair, parseStreamingJson } from \"../utils/json-parse.js\";\nimport { sanitizeSurrogates } from \"../utils/sanitize-unicode.js\";\nimport {\n\tADAPTIVE_THINKING_PATTERNS,\n\tNATIVE_XHIGH_EFFORT_PATTERNS,\n\tnormalizeForPatternMatch,\n} from \"./anthropic-model-patterns.js\";\n\nimport { resolveCloudflareBaseUrl } from \"./cloudflare.js\";\nimport { buildCopilotDynamicHeaders, hasCopilotVisionInput } from \"./github-copilot-headers.js\";\nimport { adjustMaxTokensForThinking, buildBaseOptions } from \"./simple-options.js\";\nimport { transformMessages } from \"./transform-messages.js\";\n\n/**\n * Resolve cache retention preference.\n *\n * Defaults to \"short\" and only honors PI_CACHE_RETENTION === \"long\" — this is\n * the low-level provider fallback for callers that hit the adapter directly.\n * In the coding-agent path this default rarely fires: the SDK's own\n * resolveCacheRetention (packages/coding-agent/src/core/sdk.ts, which defaults\n * to \"long\") resolves first and passes a concrete value down, so `cacheRetention`\n * is normally already set when it reaches here. Keep the two defaults in sync\n * intentionally — they differ on purpose (backward-compat vs. SDK default).\n */\nfunction resolveCacheRetention(cacheRetention?: CacheRetention): CacheRetention {\n\tif (cacheRetention) {\n\t\treturn cacheRetention;\n\t}\n\tif (typeof process !== \"undefined\" && process.env.PI_CACHE_RETENTION === \"long\") {\n\t\treturn \"long\";\n\t}\n\treturn \"short\";\n}\n\nfunction getCacheControl(\n\tmodel: Model<\"anthropic-messages\">,\n\tcacheRetention?: CacheRetention,\n): { retention: CacheRetention; cacheControl?: CacheControlEphemeral } {\n\tconst retention = resolveCacheRetention(cacheRetention);\n\tif (retention === \"none\") {\n\t\treturn { retention };\n\t}\n\tconst ttl = retention === \"long\" && getAnthropicCompat(model).supportsLongCacheRetention ? \"1h\" : undefined;\n\treturn {\n\t\tretention,\n\t\tcacheControl: { type: \"ephemeral\", ...(ttl && { ttl }) },\n\t};\n}\n\n// Stealth mode: Mimic Claude Code's tool naming exactly\nconst claudeCodeVersion = \"2.1.75\";\n\n// Claude Code 2.x tool names (canonical casing)\n// Source: https://cchistory.mariozechner.at/data/prompts-2.1.11.md\n// To update: https://github.com/badlogic/cchistory\nconst claudeCodeTools = [\n\t\"Read\",\n\t\"Write\",\n\t\"Edit\",\n\t\"Bash\",\n\t\"Grep\",\n\t\"Glob\",\n\t\"AskUserQuestion\",\n\t\"EnterPlanMode\",\n\t\"ExitPlanMode\",\n\t\"KillShell\",\n\t\"NotebookEdit\",\n\t\"Skill\",\n\t\"Task\",\n\t\"TaskOutput\",\n\t\"TodoWrite\",\n\t\"WebFetch\",\n\t\"WebSearch\",\n];\n\nconst ccToolLookup = new Map(claudeCodeTools.map((t) => [t.toLowerCase(), t]));\n\n// Convert tool name to CC canonical casing if it matches (case-insensitive)\nconst toClaudeCodeName = (name: string) => ccToolLookup.get(name.toLowerCase()) ?? name;\nconst fromClaudeCodeName = (name: string, tools?: Tool[]) => {\n\tif (tools && tools.length > 0) {\n\t\tconst lowerName = name.toLowerCase();\n\t\tconst matchedTool = tools.find((tool) => tool.name.toLowerCase() === lowerName);\n\t\tif (matchedTool) return matchedTool.name;\n\t}\n\treturn name;\n};\n\n/**\n * Convert content blocks to Anthropic API format\n */\nfunction convertContentBlocks(content: (TextContent | ImageContent)[]):\n\t| string\n\t| Array<\n\t\t\t| { type: \"text\"; text: string }\n\t\t\t| {\n\t\t\t\t\ttype: \"image\";\n\t\t\t\t\tsource: {\n\t\t\t\t\t\ttype: \"base64\";\n\t\t\t\t\t\tmedia_type: \"image/jpeg\" | \"image/png\" | \"image/gif\" | \"image/webp\";\n\t\t\t\t\t\tdata: string;\n\t\t\t\t\t};\n\t\t\t  }\n\t  > {\n\t// If only text blocks, return as concatenated string for simplicity\n\tconst hasImages = content.some((c) => c.type === \"image\");\n\tif (!hasImages) {\n\t\treturn sanitizeSurrogates(content.map((c) => (c as TextContent).text).join(\"\\n\"));\n\t}\n\n\t// If we have images, convert to content block array\n\tconst blocks = content.map((block) => {\n\t\tif (block.type === \"text\") {\n\t\t\treturn {\n\t\t\t\ttype: \"text\" as const,\n\t\t\t\ttext: sanitizeSurrogates(block.text),\n\t\t\t};\n\t\t}\n\t\treturn {\n\t\t\ttype: \"image\" as const,\n\t\t\tsource: {\n\t\t\t\ttype: \"base64\" as const,\n\t\t\t\tmedia_type: block.mimeType as \"image/jpeg\" | \"image/png\" | \"image/gif\" | \"image/webp\",\n\t\t\t\tdata: block.data,\n\t\t\t},\n\t\t};\n\t});\n\n\t// If only images (no text), add placeholder text block\n\tconst hasText = blocks.some((b) => b.type === \"text\");\n\tif (!hasText) {\n\t\tblocks.unshift({\n\t\t\ttype: \"text\" as const,\n\t\t\ttext: \"(see attached image)\",\n\t\t});\n\t}\n\n\treturn blocks;\n}\n\nexport type AnthropicEffort = \"low\" | \"medium\" | \"high\" | \"xhigh\" | \"max\";\n\nexport type AnthropicThinkingDisplay = \"summarized\" | \"omitted\";\n\nconst FINE_GRAINED_TOOL_STREAMING_BETA = \"fine-grained-tool-streaming-2025-05-14\";\nconst INTERLEAVED_THINKING_BETA = \"interleaved-thinking-2025-05-14\";\n\nfunction getAnthropicCompat(model: Model<\"anthropic-messages\">): Required<AnthropicMessagesCompat> {\n\t// Auto-detect session affinity and cache control support from provider\n\tconst isFireworks = model.provider === \"fireworks\";\n\tconst isCloudflareAiGatewayAnthropic =\n\t\tmodel.provider === \"cloudflare-ai-gateway\" && model.baseUrl.includes(\"anthropic\");\n\t// SCRAMJET-DIVERGENCE: supportsTemperature and forceAdaptiveThinking defaults added for Opus 4.8/Fable 5/Sonnet 5 support\n\treturn {\n\t\tsupportsEagerToolInputStreaming: model.compat?.supportsEagerToolInputStreaming ?? !isFireworks,\n\t\tsupportsLongCacheRetention: model.compat?.supportsLongCacheRetention ?? !isFireworks,\n\t\tsendSessionAffinityHeaders:\n\t\t\tmodel.compat?.sendSessionAffinityHeaders ?? !!(isFireworks || isCloudflareAiGatewayAnthropic),\n\t\tsupportsCacheControlOnTools: model.compat?.supportsCacheControlOnTools ?? !isFireworks,\n\t\tsupportsTemperature: model.compat?.supportsTemperature ?? true,\n\t\tforceAdaptiveThinking: model.compat?.forceAdaptiveThinking ?? false,\n\t};\n}\n\nexport interface AnthropicOptions extends StreamOptions {\n\t/**\n\t * Enable extended thinking.\n\t * For adaptive thinking models (Opus 4.6+, Sonnet 4.6+, Fable 5): model decides when/how much to think.\n\t * For older models: uses budget-based thinking with thinkingBudgetTokens.\n\t */\n\tthinkingEnabled?: boolean;\n\t/**\n\t * Token budget for extended thinking (older models only).\n\t * Ignored for adaptive thinking models (Opus 4.6+, Sonnet 4.6+, Fable 5).\n\t */\n\tthinkingBudgetTokens?: number;\n\t/**\n\t * Effort level for adaptive thinking (Opus 4.6+, Sonnet 4.6+, Fable 5).\n\t * Controls how much thinking Claude allocates:\n\t * - \"max\": Always thinks with no constraints (Opus 4.6 only)\n\t * - \"xhigh\": Highest reasoning level (Opus 4.7+, Fable 5)\n\t * - \"high\": Always thinks, deep reasoning (default)\n\t * - \"medium\": Moderate thinking, may skip for simple queries\n\t * - \"low\": Minimal thinking, skips for simple tasks\n\t * Ignored for older models.\n\t */\n\teffort?: AnthropicEffort;\n\t/**\n\t * Controls how thinking content is returned in API responses.\n\t * - \"summarized\": Thinking blocks contain summarized thinking text (default here).\n\t * - \"omitted\": Thinking blocks return an empty thinking field; the encrypted\n\t *   signature still travels back for multi-turn continuity. Use for faster\n\t *   time-to-first-text-token when your UI does not surface thinking.\n\t *\n\t * Note: Anthropic's API default for Claude Opus 4.7 and Claude Mythos Preview\n\t * is \"omitted\". We default to \"summarized\" here to keep behavior consistent\n\t * with older Claude 4 models. Set this explicitly to \"omitted\" to opt in.\n\t */\n\tthinkingDisplay?: AnthropicThinkingDisplay;\n\tinterleavedThinking?: boolean;\n\ttoolChoice?: \"auto\" | \"any\" | \"none\" | { type: \"tool\"; name: string };\n\t/**\n\t * Pre-built Anthropic client instance. When provided, skips internal client\n\t * construction entirely. Use this to inject alternative SDK clients such as\n\t * `AnthropicVertex` that shares the same messaging API.\n\t */\n\tclient?: Anthropic;\n}\n\nfunction mergeHeaders(...headerSources: (Record<string, string | null> | undefined)[]): Record<string, string | null> {\n\tconst merged: Record<string, string | null> = {};\n\tfor (const headers of headerSources) {\n\t\tif (headers) {\n\t\t\tObject.assign(merged, headers);\n\t\t}\n\t}\n\treturn merged;\n}\n\ninterface ServerSentEvent {\n\tevent: string | null;\n\tdata: string;\n\traw: string[];\n}\n\ninterface SseDecoderState {\n\tevent: string | null;\n\tdata: string[];\n\traw: string[];\n}\n\nconst ANTHROPIC_MESSAGE_EVENTS: ReadonlySet<string> = new Set([\n\t\"message_start\",\n\t\"message_delta\",\n\t\"message_stop\",\n\t\"content_block_start\",\n\t\"content_block_delta\",\n\t\"content_block_stop\",\n]);\n\nfunction flushSseEvent(state: SseDecoderState): ServerSentEvent | null {\n\tif (!state.event && state.data.length === 0) {\n\t\treturn null;\n\t}\n\n\tconst event: ServerSentEvent = {\n\t\tevent: state.event,\n\t\tdata: state.data.join(\"\\n\"),\n\t\traw: [...state.raw],\n\t};\n\tstate.event = null;\n\tstate.data = [];\n\tstate.raw = [];\n\treturn event;\n}\n\nfunction decodeSseLine(line: string, state: SseDecoderState): ServerSentEvent | null {\n\tif (line === \"\") {\n\t\treturn flushSseEvent(state);\n\t}\n\n\tstate.raw.push(line);\n\tif (line.startsWith(\":\")) {\n\t\treturn null;\n\t}\n\n\tconst delimiterIndex = line.indexOf(\":\");\n\tconst fieldName = delimiterIndex === -1 ? line : line.slice(0, delimiterIndex);\n\tlet value = delimiterIndex === -1 ? \"\" : line.slice(delimiterIndex + 1);\n\tif (value.startsWith(\" \")) {\n\t\tvalue = value.slice(1);\n\t}\n\n\tif (fieldName === \"event\") {\n\t\tstate.event = value;\n\t} else if (fieldName === \"data\") {\n\t\tstate.data.push(value);\n\t}\n\n\treturn null;\n}\n\nfunction nextLineBreakIndex(text: string): number {\n\tconst carriageReturnIndex = text.indexOf(\"\\r\");\n\tconst newlineIndex = text.indexOf(\"\\n\");\n\tif (carriageReturnIndex === -1) {\n\t\treturn newlineIndex;\n\t}\n\tif (newlineIndex === -1) {\n\t\treturn carriageReturnIndex;\n\t}\n\treturn Math.min(carriageReturnIndex, newlineIndex);\n}\n\nfunction consumeLine(text: string): { line: string; rest: string } | null {\n\tconst lineBreakIndex = nextLineBreakIndex(text);\n\tif (lineBreakIndex === -1) {\n\t\treturn null;\n\t}\n\n\tlet nextIndex = lineBreakIndex + 1;\n\tif (text[lineBreakIndex] === \"\\r\" && text[nextIndex] === \"\\n\") {\n\t\tnextIndex += 1;\n\t}\n\n\treturn {\n\t\tline: text.slice(0, lineBreakIndex),\n\t\trest: text.slice(nextIndex),\n\t};\n}\n\nasync function* iterateSseMessages(\n\tbody: ReadableStream<Uint8Array>,\n\tsignal?: AbortSignal,\n): AsyncGenerator<ServerSentEvent> {\n\tconst reader = body.getReader();\n\tconst decoder = new TextDecoder();\n\tconst state: SseDecoderState = { event: null, data: [], raw: [] };\n\tlet buffer = \"\";\n\n\ttry {\n\t\twhile (true) {\n\t\t\tif (signal?.aborted) {\n\t\t\t\tthrow new Error(\"Request was aborted\");\n\t\t\t}\n\n\t\t\tconst { value, done } = await reader.read();\n\t\t\tif (done) {\n\t\t\t\tbreak;\n\t\t\t}\n\n\t\t\tbuffer += decoder.decode(value, { stream: true });\n\t\t\tlet consumed = consumeLine(buffer);\n\t\t\twhile (consumed) {\n\t\t\t\tbuffer = consumed.rest;\n\t\t\t\tconst event = decodeSseLine(consumed.line, state);\n\t\t\t\tif (event) {\n\t\t\t\t\tyield event;\n\t\t\t\t}\n\t\t\t\tconsumed = consumeLine(buffer);\n\t\t\t}\n\t\t}\n\n\t\tbuffer += decoder.decode();\n\t\tlet consumed = consumeLine(buffer);\n\t\twhile (consumed) {\n\t\t\tbuffer = consumed.rest;\n\t\t\tconst event = decodeSseLine(consumed.line, state);\n\t\t\tif (event) {\n\t\t\t\tyield event;\n\t\t\t}\n\t\t\tconsumed = consumeLine(buffer);\n\t\t}\n\n\t\tif (buffer.length > 0) {\n\t\t\tconst event = decodeSseLine(buffer, state);\n\t\t\tif (event) {\n\t\t\t\tyield event;\n\t\t\t}\n\t\t}\n\n\t\tconst trailingEvent = flushSseEvent(state);\n\t\tif (trailingEvent) {\n\t\t\tyield trailingEvent;\n\t\t}\n\t} finally {\n\t\treader.releaseLock();\n\t}\n}\n\n// SCRAMJET-DIVERGENCE: Preserve provider and callback failure facts independently.\nasync function* iterateAnthropicEvents(\n\tresponse: Response,\n\tsignal?: AbortSignal,\n): AsyncGenerator<RawMessageStreamEvent> {\n\tif (!response.body) {\n\t\tthrow new RequestFailureError(\"Anthropic response has no body.\", {\n\t\t\tschemaVersion: 1,\n\t\t\tkind: \"stream\",\n\t\t\treason: \"missing_body\",\n\t\t});\n\t}\n\n\tlet sawMessageStart = false;\n\tlet sawMessageEnd = false;\n\n\tfor await (const sse of iterateSseMessages(response.body, signal)) {\n\t\tif (sse.event === \"error\") {\n\t\t\tlet value: unknown;\n\t\t\ttry {\n\t\t\t\tvalue = JSON.parse(sse.data);\n\t\t\t} catch {\n\t\t\t\tthrow new RequestFailureError(\"Malformed Anthropic error event.\", {\n\t\t\t\t\tschemaVersion: 1,\n\t\t\t\t\tkind: \"stream\",\n\t\t\t\t\treason: \"malformed_event\",\n\t\t\t\t});\n\t\t\t}\n\t\t\tif (\n\t\t\t\t!value ||\n\t\t\t\ttypeof value !== \"object\" ||\n\t\t\t\tArray.isArray(value) ||\n\t\t\t\t!(\"error\" in value) ||\n\t\t\t\t!value.error ||\n\t\t\t\ttypeof value.error !== \"object\" ||\n\t\t\t\tArray.isArray(value.error)\n\t\t\t)\n\t\t\t\tthrow new RequestFailureError(\"Malformed Anthropic error event.\", {\n\t\t\t\t\tschemaVersion: 1,\n\t\t\t\t\tkind: \"stream\",\n\t\t\t\t\treason: \"malformed_event\",\n\t\t\t\t});\n\t\t\tthrow new RequestFailureError(\n\t\t\t\tsse.data,\n\t\t\t\tfailureFromProviderError(value) ?? { schemaVersion: 1, kind: \"provider\", category: \"unknown\" },\n\t\t\t);\n\t\t}\n\n\t\tif (!ANTHROPIC_MESSAGE_EVENTS.has(sse.event ?? \"\")) {\n\t\t\tcontinue;\n\t\t}\n\n\t\ttry {\n\t\t\tconst event = parseJsonWithRepair<RawMessageStreamEvent>(sse.data);\n\t\t\tif (event.type === \"message_start\") {\n\t\t\t\tsawMessageStart = true;\n\t\t\t} else if (event.type === \"message_stop\") {\n\t\t\t\tsawMessageEnd = true;\n\t\t\t}\n\t\t\tyield event;\n\t\t} catch (error) {\n\t\t\tconst message = error instanceof Error ? error.message : String(error);\n\t\t\tthrow new RequestFailureError(\n\t\t\t\t`Could not parse Anthropic SSE event ${sse.event}: ${message}; data=${sse.data}; raw=${sse.raw.join(\"\\\\n\")}`,\n\t\t\t\t{ schemaVersion: 1, kind: \"stream\", reason: \"malformed_event\" },\n\t\t\t);\n\t\t}\n\t}\n\n\tif (sawMessageStart && !sawMessageEnd) {\n\t\tthrow new RequestFailureError(\"Anthropic stream ended before message_stop\", {\n\t\t\tschemaVersion: 1,\n\t\t\tkind: \"stream\",\n\t\t\treason: \"missing_terminal_event\",\n\t\t});\n\t}\n}\n\nexport const streamAnthropic: StreamFunction<\"anthropic-messages\", AnthropicOptions> = (\n\tmodel: Model<\"anthropic-messages\">,\n\tcontext: Context,\n\toptions?: AnthropicOptions,\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: model.api 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\tlet requestStarted = false;\n\t\ttry {\n\t\t\tlet client: Anthropic;\n\t\t\tlet isOAuth: boolean;\n\n\t\t\tif (options?.client) {\n\t\t\t\tclient = options.client;\n\t\t\t\tisOAuth = false;\n\t\t\t} else {\n\t\t\t\tconst apiKey = options?.apiKey ?? getEnvApiKey(model.provider) ?? \"\";\n\n\t\t\t\tlet copilotDynamicHeaders: Record<string, string> | undefined;\n\t\t\t\tif (model.provider === \"github-copilot\") {\n\t\t\t\t\tconst hasImages = hasCopilotVisionInput(context.messages);\n\t\t\t\t\tcopilotDynamicHeaders = buildCopilotDynamicHeaders({\n\t\t\t\t\t\tmessages: context.messages,\n\t\t\t\t\t\thasImages,\n\t\t\t\t\t});\n\t\t\t\t}\n\n\t\t\t\tconst cacheRetention = options?.cacheRetention ?? resolveCacheRetention();\n\t\t\t\tconst cacheSessionId = cacheRetention === \"none\" ? undefined : options?.sessionId;\n\n\t\t\t\tconst created = createClient(\n\t\t\t\t\tmodel,\n\t\t\t\t\tapiKey,\n\t\t\t\t\toptions?.interleavedThinking ?? true,\n\t\t\t\t\tshouldUseFineGrainedToolStreamingBeta(model, context),\n\t\t\t\t\toptions?.headers,\n\t\t\t\t\tcopilotDynamicHeaders,\n\t\t\t\t\tcacheSessionId,\n\t\t\t\t\t(input, init) => {\n\t\t\t\t\t\trequestStarted = true;\n\t\t\t\t\t\treturn fetch(input, init);\n\t\t\t\t\t},\n\t\t\t\t);\n\t\t\t\tclient = created.client;\n\t\t\t\tisOAuth = created.isOAuthToken;\n\t\t\t}\n\t\t\tlet params = buildParams(model, context, isOAuth, options);\n\t\t\tconst nextParams = await invokeProviderCallback(\"onPayload\", () => options?.onPayload?.(params, model));\n\t\t\tif (nextParams !== undefined) {\n\t\t\t\tparams = nextParams as MessageCreateParamsStreaming;\n\t\t\t}\n\t\t\tconst requestOptions = {\n\t\t\t\t...(options?.signal ? { signal: options.signal } : {}),\n\t\t\t\t...(options?.timeoutMs !== undefined ? { timeout: options.timeoutMs } : {}),\n\t\t\t\t...(options?.maxRetries !== undefined ? { maxRetries: options.maxRetries } : {}),\n\t\t\t};\n\t\t\tif (options?.client) requestStarted = true;\n\t\t\tconst response = await client.messages.create({ ...params, stream: true }, requestOptions).asResponse();\n\t\t\tawait invokeProviderCallback(\"onResponse\", () =>\n\t\t\t\toptions?.onResponse?.({ status: response.status, headers: headersToRecord(response.headers) }, model),\n\t\t\t);\n\t\t\tstream.push({ type: \"start\", partial: output });\n\n\t\t\ttype Block = (ThinkingContent | TextContent | (ToolCall & { partialJson: string })) & { index: number };\n\t\t\tconst blocks = output.content as Block[];\n\n\t\t\tfor await (const event of iterateAnthropicEvents(response, options?.signal)) {\n\t\t\t\tif (event.type === \"message_start\") {\n\t\t\t\t\toutput.responseId = event.message.id;\n\t\t\t\t\t// Capture initial token usage from message_start event\n\t\t\t\t\t// This ensures we have input token counts even if the stream is aborted early\n\t\t\t\t\toutput.usage.input = event.message.usage.input_tokens || 0;\n\t\t\t\t\toutput.usage.output = event.message.usage.output_tokens || 0;\n\t\t\t\t\toutput.usage.cacheRead = event.message.usage.cache_read_input_tokens || 0;\n\t\t\t\t\toutput.usage.cacheWrite = event.message.usage.cache_creation_input_tokens || 0;\n\t\t\t\t\t// Anthropic doesn't provide total_tokens, compute from components\n\t\t\t\t\toutput.usage.totalTokens =\n\t\t\t\t\t\toutput.usage.input + output.usage.output + output.usage.cacheRead + output.usage.cacheWrite;\n\t\t\t\t\tcalculateCost(model, output.usage);\n\t\t\t\t} else if (event.type === \"content_block_start\") {\n\t\t\t\t\tif (event.content_block.type === \"text\") {\n\t\t\t\t\t\tconst block: Block = {\n\t\t\t\t\t\t\ttype: \"text\",\n\t\t\t\t\t\t\ttext: \"\",\n\t\t\t\t\t\t\tindex: event.index,\n\t\t\t\t\t\t};\n\t\t\t\t\t\toutput.content.push(block);\n\t\t\t\t\t\tstream.push({ type: \"text_start\", contentIndex: output.content.length - 1, partial: output });\n\t\t\t\t\t} else if (event.content_block.type === \"thinking\") {\n\t\t\t\t\t\tconst block: Block = {\n\t\t\t\t\t\t\ttype: \"thinking\",\n\t\t\t\t\t\t\tthinking: \"\",\n\t\t\t\t\t\t\tthinkingSignature: \"\",\n\t\t\t\t\t\t\tindex: event.index,\n\t\t\t\t\t\t};\n\t\t\t\t\t\toutput.content.push(block);\n\t\t\t\t\t\tstream.push({ type: \"thinking_start\", contentIndex: output.content.length - 1, partial: output });\n\t\t\t\t\t} else if (event.content_block.type === \"redacted_thinking\") {\n\t\t\t\t\t\tconst block: Block = {\n\t\t\t\t\t\t\ttype: \"thinking\",\n\t\t\t\t\t\t\tthinking: \"[Reasoning redacted]\",\n\t\t\t\t\t\t\tthinkingSignature: event.content_block.data,\n\t\t\t\t\t\t\tredacted: true,\n\t\t\t\t\t\t\tindex: event.index,\n\t\t\t\t\t\t};\n\t\t\t\t\t\toutput.content.push(block);\n\t\t\t\t\t\tstream.push({ type: \"thinking_start\", contentIndex: output.content.length - 1, partial: output });\n\t\t\t\t\t} else if (event.content_block.type === \"tool_use\") {\n\t\t\t\t\t\tconst block: Block = {\n\t\t\t\t\t\t\ttype: \"toolCall\",\n\t\t\t\t\t\t\tid: event.content_block.id,\n\t\t\t\t\t\t\tname: isOAuth\n\t\t\t\t\t\t\t\t? fromClaudeCodeName(event.content_block.name, context.tools)\n\t\t\t\t\t\t\t\t: event.content_block.name,\n\t\t\t\t\t\t\targuments: (event.content_block.input as Record<string, any>) ?? {},\n\t\t\t\t\t\t\tpartialJson: \"\",\n\t\t\t\t\t\t\tindex: event.index,\n\t\t\t\t\t\t};\n\t\t\t\t\t\toutput.content.push(block);\n\t\t\t\t\t\tstream.push({ type: \"toolcall_start\", contentIndex: output.content.length - 1, partial: output });\n\t\t\t\t\t}\n\t\t\t\t} else if (event.type === \"content_block_delta\") {\n\t\t\t\t\tif (event.delta.type === \"text_delta\") {\n\t\t\t\t\t\tconst index = blocks.findIndex((b) => b.index === event.index);\n\t\t\t\t\t\tconst block = blocks[index];\n\t\t\t\t\t\tif (block && block.type === \"text\") {\n\t\t\t\t\t\t\tblock.text += event.delta.text;\n\t\t\t\t\t\t\tstream.push({\n\t\t\t\t\t\t\t\ttype: \"text_delta\",\n\t\t\t\t\t\t\t\tcontentIndex: index,\n\t\t\t\t\t\t\t\tdelta: event.delta.text,\n\t\t\t\t\t\t\t\tpartial: output,\n\t\t\t\t\t\t\t});\n\t\t\t\t\t\t}\n\t\t\t\t\t} else if (event.delta.type === \"thinking_delta\") {\n\t\t\t\t\t\tconst index = blocks.findIndex((b) => b.index === event.index);\n\t\t\t\t\t\tconst block = blocks[index];\n\t\t\t\t\t\tif (block && block.type === \"thinking\") {\n\t\t\t\t\t\t\tblock.thinking += event.delta.thinking;\n\t\t\t\t\t\t\tstream.push({\n\t\t\t\t\t\t\t\ttype: \"thinking_delta\",\n\t\t\t\t\t\t\t\tcontentIndex: index,\n\t\t\t\t\t\t\t\tdelta: event.delta.thinking,\n\t\t\t\t\t\t\t\tpartial: output,\n\t\t\t\t\t\t\t});\n\t\t\t\t\t\t}\n\t\t\t\t\t} else if (event.delta.type === \"input_json_delta\") {\n\t\t\t\t\t\tconst index = blocks.findIndex((b) => b.index === event.index);\n\t\t\t\t\t\tconst block = blocks[index];\n\t\t\t\t\t\tif (block && block.type === \"toolCall\") {\n\t\t\t\t\t\t\tblock.partialJson += event.delta.partial_json;\n\t\t\t\t\t\t\tblock.arguments = parseStreamingJson(block.partialJson);\n\t\t\t\t\t\t\tstream.push({\n\t\t\t\t\t\t\t\ttype: \"toolcall_delta\",\n\t\t\t\t\t\t\t\tcontentIndex: index,\n\t\t\t\t\t\t\t\tdelta: event.delta.partial_json,\n\t\t\t\t\t\t\t\tpartial: output,\n\t\t\t\t\t\t\t});\n\t\t\t\t\t\t}\n\t\t\t\t\t} else if (event.delta.type === \"signature_delta\") {\n\t\t\t\t\t\tconst index = blocks.findIndex((b) => b.index === event.index);\n\t\t\t\t\t\tconst block = blocks[index];\n\t\t\t\t\t\tif (block && block.type === \"thinking\") {\n\t\t\t\t\t\t\tblock.thinkingSignature = block.thinkingSignature || \"\";\n\t\t\t\t\t\t\tblock.thinkingSignature += event.delta.signature;\n\t\t\t\t\t\t}\n\t\t\t\t\t}\n\t\t\t\t} else if (event.type === \"content_block_stop\") {\n\t\t\t\t\tconst index = blocks.findIndex((b) => b.index === event.index);\n\t\t\t\t\tconst block = blocks[index];\n\t\t\t\t\tif (block) {\n\t\t\t\t\t\tdelete (block as any).index;\n\t\t\t\t\t\tif (block.type === \"text\") {\n\t\t\t\t\t\t\tstream.push({\n\t\t\t\t\t\t\t\ttype: \"text_end\",\n\t\t\t\t\t\t\t\tcontentIndex: index,\n\t\t\t\t\t\t\t\tcontent: block.text,\n\t\t\t\t\t\t\t\tpartial: output,\n\t\t\t\t\t\t\t});\n\t\t\t\t\t\t} else if (block.type === \"thinking\") {\n\t\t\t\t\t\t\tstream.push({\n\t\t\t\t\t\t\t\ttype: \"thinking_end\",\n\t\t\t\t\t\t\t\tcontentIndex: index,\n\t\t\t\t\t\t\t\tcontent: block.thinking,\n\t\t\t\t\t\t\t\tpartial: output,\n\t\t\t\t\t\t\t});\n\t\t\t\t\t\t} else if (block.type === \"toolCall\") {\n\t\t\t\t\t\t\tblock.arguments = parseStreamingJson(block.partialJson);\n\t\t\t\t\t\t\t// Finalize in-place and strip the scratch buffer so replay only\n\t\t\t\t\t\t\t// carries parsed arguments.\n\t\t\t\t\t\t\tdelete (block as { partialJson?: string }).partialJson;\n\t\t\t\t\t\t\tstream.push({\n\t\t\t\t\t\t\t\ttype: \"toolcall_end\",\n\t\t\t\t\t\t\t\tcontentIndex: index,\n\t\t\t\t\t\t\t\ttoolCall: block,\n\t\t\t\t\t\t\t\tpartial: output,\n\t\t\t\t\t\t\t});\n\t\t\t\t\t\t}\n\t\t\t\t\t}\n\t\t\t\t} else if (event.type === \"message_delta\") {\n\t\t\t\t\tif (event.delta.stop_reason) {\n\t\t\t\t\t\tif (output.stopReason !== \"error\") output.stopReason = mapStopReason(event.delta.stop_reason);\n\t\t\t\t\t}\n\t\t\t\t\t// Only update usage fields if present (not null).\n\t\t\t\t\t// Preserves input_tokens from message_start when proxies omit it in message_delta.\n\t\t\t\t\tif (event.usage.input_tokens != null) {\n\t\t\t\t\t\toutput.usage.input = event.usage.input_tokens;\n\t\t\t\t\t}\n\t\t\t\t\tif (event.usage.output_tokens != null) {\n\t\t\t\t\t\toutput.usage.output = event.usage.output_tokens;\n\t\t\t\t\t}\n\t\t\t\t\tif (event.usage.cache_read_input_tokens != null) {\n\t\t\t\t\t\toutput.usage.cacheRead = event.usage.cache_read_input_tokens;\n\t\t\t\t\t}\n\t\t\t\t\tif (event.usage.cache_creation_input_tokens != null) {\n\t\t\t\t\t\toutput.usage.cacheWrite = event.usage.cache_creation_input_tokens;\n\t\t\t\t\t}\n\t\t\t\t\t// Anthropic doesn't provide total_tokens, compute from components\n\t\t\t\t\toutput.usage.totalTokens =\n\t\t\t\t\t\toutput.usage.input + output.usage.output + output.usage.cacheRead + output.usage.cacheWrite;\n\t\t\t\t\tcalculateCost(model, output.usage);\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\")\n\t\t\t\tthrow new RequestFailureError(\"Anthropic rejected the output.\", {\n\t\t\t\t\tschemaVersion: 1,\n\t\t\t\t\tkind: \"provider\",\n\t\t\t\t\tcategory: \"content_rejection\",\n\t\t\t\t});\n\t\t\tif (output.stopReason === \"aborted\") throw new Error(\"Request was aborted\");\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 { index?: number }).index;\n\t\t\t\t// partialJson is only a streaming scratch buffer; never persist it.\n\t\t\t\tdelete (block as { partialJson?: string }).partialJson;\n\t\t\t}\n\t\t\toutput.stopReason = options?.signal?.aborted ? \"aborted\" : \"error\";\n\t\t\toutput.errorMessage = error instanceof Error ? error.message : JSON.stringify(error);\n\t\t\tappendBuiltinFailure(\n\t\t\t\toutput,\n\t\t\t\trequestStarted && !options?.signal?.aborted && error instanceof Anthropic.APIConnectionError\n\t\t\t\t\t? new RequestFailureError(output.errorMessage, {\n\t\t\t\t\t\t\tschemaVersion: 1,\n\t\t\t\t\t\t\tkind: \"provider\",\n\t\t\t\t\t\t\tcategory: error instanceof Anthropic.APIConnectionTimeoutError ? \"timeout\" : \"transport\",\n\t\t\t\t\t\t})\n\t\t\t\t\t: error,\n\t\t\t\t!requestStarted,\n\t\t\t);\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// SCRAMJET-DIVERGENCE: extended with opus-4-8, fable-5, sonnet-5 patterns for new model support\nfunction supportsAdaptiveThinking(modelId: string): boolean {\n\tconst normalized = normalizeForPatternMatch(modelId);\n\treturn ADAPTIVE_THINKING_PATTERNS.some((p) => normalized.includes(p));\n}\n\n// SCRAMJET-DIVERGENCE: native xhigh effort support for Opus 4.8/Fable 5 models\nfunction supportsNativeXhighEffort(modelId: string): boolean {\n\tconst normalized = normalizeForPatternMatch(modelId);\n\treturn NATIVE_XHIGH_EFFORT_PATTERNS.some((p) => normalized.includes(p));\n}\n\n/**\n * Map ThinkingLevel to Anthropic effort levels for adaptive thinking.\n * Note: effort \"max\" is only valid on Opus 4.6; Opus 4.7+ and Fable 5 support \"xhigh\".\n */\nfunction mapThinkingLevelToEffort(\n\tmodel: Model<\"anthropic-messages\">,\n\tlevel: SimpleStreamOptions[\"reasoning\"],\n): AnthropicEffort {\n\tconst mapped = level ? model.thinkingLevelMap?.[level] : undefined;\n\tif (typeof mapped === \"string\") return mapped as AnthropicEffort;\n\n\tswitch (level) {\n\t\tcase \"minimal\":\n\t\tcase \"low\":\n\t\t\treturn \"low\";\n\t\tcase \"medium\":\n\t\t\treturn \"medium\";\n\t\tcase \"high\":\n\t\t\treturn \"high\";\n\t\tcase \"xhigh\":\n\t\t\treturn supportsNativeXhighEffort(model.id) ? \"xhigh\" : \"high\";\n\t\tdefault:\n\t\t\treturn \"high\";\n\t}\n}\n\nexport const streamSimpleAnthropic: StreamFunction<\"anthropic-messages\", SimpleStreamOptions> = (\n\tmodel: Model<\"anthropic-messages\">,\n\tcontext: Context,\n\toptions?: SimpleStreamOptions,\n): AssistantMessageEventStream => {\n\tconst apiKey = options?.apiKey || getEnvApiKey(model.provider);\n\tif (!apiKey) {\n\t\tthrow new Error(`No API key for provider: ${model.provider}`);\n\t}\n\n\tconst base = buildBaseOptions(model, options, apiKey);\n\tif (!options?.reasoning) {\n\t\treturn streamAnthropic(model, context, { ...base, thinkingEnabled: false } satisfies AnthropicOptions);\n\t}\n\n\t// For adaptive-thinking models: use effort level.\n\t// For older models: use budget-based thinking.\n\tif (supportsAdaptiveThinking(model.id)) {\n\t\tconst effort = mapThinkingLevelToEffort(model, options.reasoning);\n\t\treturn streamAnthropic(model, context, {\n\t\t\t...base,\n\t\t\tthinkingEnabled: true,\n\t\t\teffort,\n\t\t} satisfies AnthropicOptions);\n\t}\n\n\tconst adjusted = adjustMaxTokensForThinking(\n\t\tbase.maxTokens || 0,\n\t\tMath.min(model.maxTokens, options.maxTokens ?? Infinity),\n\t\toptions.reasoning,\n\t\toptions.thinkingBudgets,\n\t);\n\n\t// SCRAMJET-DIVERGENCE: An explicit output ceiling includes thinking; never resurrect a zero budget.\n\tif (adjusted.thinkingBudget < 1024) {\n\t\tthrow new Error(\"Insufficient output space for Anthropic thinking; reduce input or request more output space.\");\n\t}\n\treturn streamAnthropic(model, context, {\n\t\t...base,\n\t\tmaxTokens: adjusted.maxTokens,\n\t\tthinkingEnabled: true,\n\t\tthinkingBudgetTokens: adjusted.thinkingBudget,\n\t} satisfies AnthropicOptions);\n};\n\nfunction isOAuthToken(apiKey: string): boolean {\n\treturn apiKey.includes(\"sk-ant-oat\");\n}\n\nfunction createClient(\n\tmodel: Model<\"anthropic-messages\">,\n\tapiKey: string,\n\tinterleavedThinking: boolean,\n\tuseFineGrainedToolStreamingBeta: boolean,\n\toptionsHeaders?: Record<string, string>,\n\tdynamicHeaders?: Record<string, string>,\n\tsessionId?: string,\n\tfetchImplementation?: typeof fetch,\n): { client: Anthropic; isOAuthToken: boolean } {\n\t// Adaptive-thinking models have interleaved thinking built in, so skip the beta header.\n\tconst needsInterleavedBeta = interleavedThinking && !supportsAdaptiveThinking(model.id);\n\tconst betaFeatures: string[] = [];\n\tif (useFineGrainedToolStreamingBeta) {\n\t\tbetaFeatures.push(FINE_GRAINED_TOOL_STREAMING_BETA);\n\t}\n\tif (needsInterleavedBeta) {\n\t\tbetaFeatures.push(INTERLEAVED_THINKING_BETA);\n\t}\n\n\tif (model.provider === \"cloudflare-ai-gateway\") {\n\t\tconst client = new Anthropic({\n\t\t\tapiKey: null,\n\t\t\tauthToken: null,\n\t\t\tbaseURL: resolveCloudflareBaseUrl(model),\n\t\t\tfetch: fetchImplementation,\n\t\t\tdangerouslyAllowBrowser: true,\n\t\t\tdefaultHeaders: mergeHeaders(\n\t\t\t\t{\n\t\t\t\t\taccept: \"application/json\",\n\t\t\t\t\t\"anthropic-dangerous-direct-browser-access\": \"true\",\n\t\t\t\t\t\"cf-aig-authorization\": `Bearer ${apiKey}`,\n\t\t\t\t\t\"x-api-key\": null,\n\t\t\t\t\tAuthorization: null,\n\t\t\t\t\t...(betaFeatures.length > 0 ? { \"anthropic-beta\": betaFeatures.join(\",\") } : {}),\n\t\t\t\t},\n\t\t\t\tmodel.headers,\n\t\t\t\toptionsHeaders,\n\t\t\t),\n\t\t});\n\n\t\treturn { client, isOAuthToken: false };\n\t}\n\n\t// Copilot: Bearer auth, selective betas.\n\tif (model.provider === \"github-copilot\") {\n\t\tconst client = new Anthropic({\n\t\t\tfetch: fetchImplementation,\n\t\t\tapiKey: null,\n\t\t\tauthToken: apiKey,\n\t\t\tbaseURL: model.baseUrl,\n\t\t\tdangerouslyAllowBrowser: true,\n\t\t\tdefaultHeaders: mergeHeaders(\n\t\t\t\t{\n\t\t\t\t\taccept: \"application/json\",\n\t\t\t\t\t\"anthropic-dangerous-direct-browser-access\": \"true\",\n\t\t\t\t\t...(betaFeatures.length > 0 ? { \"anthropic-beta\": betaFeatures.join(\",\") } : {}),\n\t\t\t\t},\n\t\t\t\tmodel.headers,\n\t\t\t\tdynamicHeaders,\n\t\t\t\toptionsHeaders,\n\t\t\t),\n\t\t});\n\n\t\treturn { client, isOAuthToken: false };\n\t}\n\n\t// OAuth: Bearer auth, Claude Code identity headers\n\tif (isOAuthToken(apiKey)) {\n\t\tconst client = new Anthropic({\n\t\t\tfetch: fetchImplementation,\n\t\t\tapiKey: null,\n\t\t\tauthToken: apiKey,\n\t\t\tbaseURL: model.baseUrl,\n\t\t\tdangerouslyAllowBrowser: true,\n\t\t\tdefaultHeaders: mergeHeaders(\n\t\t\t\t{\n\t\t\t\t\taccept: \"application/json\",\n\t\t\t\t\t\"anthropic-dangerous-direct-browser-access\": \"true\",\n\t\t\t\t\t\"anthropic-beta\": [\"claude-code-20250219\", \"oauth-2025-04-20\", ...betaFeatures].join(\",\"),\n\t\t\t\t\t\"user-agent\": `claude-cli/${claudeCodeVersion}`,\n\t\t\t\t\t\"x-app\": \"cli\",\n\t\t\t\t},\n\t\t\t\tmodel.headers,\n\t\t\t\toptionsHeaders,\n\t\t\t),\n\t\t});\n\n\t\treturn { client, isOAuthToken: true };\n\t}\n\n\t// API key auth\n\tconst sessionAffinityHeaders: Record<string, string | null> =\n\t\tsessionId && getAnthropicCompat(model).sendSessionAffinityHeaders ? { \"x-session-affinity\": sessionId } : {};\n\tconst client = new Anthropic({\n\t\tapiKey,\n\t\tbaseURL: model.baseUrl,\n\t\tfetch: fetchImplementation,\n\t\tdangerouslyAllowBrowser: true,\n\t\tdefaultHeaders: mergeHeaders(\n\t\t\t{\n\t\t\t\taccept: \"application/json\",\n\t\t\t\t\"anthropic-dangerous-direct-browser-access\": \"true\",\n\t\t\t\t...(betaFeatures.length > 0 ? { \"anthropic-beta\": betaFeatures.join(\",\") } : {}),\n\t\t\t},\n\t\t\tsessionAffinityHeaders,\n\t\t\tmodel.headers,\n\t\t\toptionsHeaders,\n\t\t),\n\t});\n\n\treturn { client, isOAuthToken: false };\n}\n\nconst claudeCodeIdentity = \"You are Claude Code, Anthropic's official CLI for Claude.\";\n\n/**\n * Builds the system block array for a sectioned system prompt.\n *\n * Anthropic allows at most 4 `cache_control` breakpoints per request, and the\n * last tool and last user message already consume two, so the system prompt\n * spends at most one: it lands on a single stable block folding the leading\n * run of sections up to the first `cacheRetention: \"none\"` section (zero\n * breakpoints when retention is \"none\"; when the first section is volatile\n * the breakpoint moves to the OAuth identity block, or is omitted entirely\n * on API-key auth).\n * Everything from the first volatile section onward trails as uncached blocks\n * in array order — anything after the breakpoint is excluded from the cached\n * prefix, and preserving array order keeps the concatenated block text\n * byte-identical (for well-formed text; `sanitizeSurrogates` runs per block)\n * to the flattened prompt other providers send. On OAuth the\n * Claude Code identity block stays first; caching is prefix-based, so the\n * stable block's breakpoint covers it without a second one (unlike the legacy\n * string path, which marks both for back-compat).\n */\nfunction buildSystemBlocks(\n\tsections: SystemPromptSection[],\n\tisOAuthToken: boolean,\n\tcacheControl?: CacheControlEphemeral,\n): TextBlockParam[] {\n\tconst blocks: TextBlockParam[] = [];\n\tconst nonEmpty = sections.filter((s) => s.text);\n\tconst firstVolatile = nonEmpty.findIndex((s) => s.cacheRetention === \"none\");\n\tconst stable = firstVolatile === -1 ? nonEmpty : nonEmpty.slice(0, firstVolatile);\n\tconst tail = firstVolatile === -1 ? [] : nonEmpty.slice(firstVolatile);\n\n\tif (isOAuthToken) {\n\t\tblocks.push({\n\t\t\ttype: \"text\",\n\t\t\ttext: claudeCodeIdentity,\n\t\t\t// Without stable sections the identity block is the whole stable\n\t\t\t// prefix, so the breakpoint lands here (as on the string path).\n\t\t\t...(stable.length === 0 && cacheControl ? { cache_control: cacheControl } : {}),\n\t\t});\n\t}\n\tif (stable.length > 0) {\n\t\tblocks.push({\n\t\t\ttype: \"text\",\n\t\t\ttext: sanitizeSurrogates(stable.map((s) => s.text).join(\"\")),\n\t\t\t...(cacheControl ? { cache_control: cacheControl } : {}),\n\t\t});\n\t}\n\tfor (const section of tail) {\n\t\tblocks.push({ type: \"text\", text: sanitizeSurrogates(section.text) });\n\t}\n\treturn blocks;\n}\n\nfunction buildParams(\n\tmodel: Model<\"anthropic-messages\">,\n\tcontext: Context,\n\tisOAuthToken: boolean,\n\toptions?: AnthropicOptions,\n): MessageCreateParamsStreaming {\n\tconst { cacheControl } = getCacheControl(model, options?.cacheRetention);\n\tconst params: MessageCreateParamsStreaming = {\n\t\tmodel: model.id,\n\t\tmessages: convertMessages(context.messages, model, isOAuthToken, cacheControl),\n\t\tmax_tokens: options?.maxTokens || (model.maxTokens / 3) | 0,\n\t\tstream: true,\n\t};\n\n\tif (Array.isArray(context.systemPrompt)) {\n\t\tconst blocks = buildSystemBlocks(context.systemPrompt, isOAuthToken, cacheControl);\n\t\tif (blocks.length > 0) {\n\t\t\tparams.system = blocks;\n\t\t}\n\t} else {\n\t\tconst systemPrompt = context.systemPrompt;\n\n\t\t// For OAuth tokens, we MUST include Claude Code identity\n\t\tif (isOAuthToken) {\n\t\t\tparams.system = [\n\t\t\t\t{\n\t\t\t\t\ttype: \"text\",\n\t\t\t\t\ttext: claudeCodeIdentity,\n\t\t\t\t\t...(cacheControl ? { cache_control: cacheControl } : {}),\n\t\t\t\t},\n\t\t\t];\n\t\t\tif (systemPrompt) {\n\t\t\t\tparams.system.push({\n\t\t\t\t\ttype: \"text\",\n\t\t\t\t\ttext: sanitizeSurrogates(systemPrompt),\n\t\t\t\t\t...(cacheControl ? { cache_control: cacheControl } : {}),\n\t\t\t\t});\n\t\t\t}\n\t\t} else if (systemPrompt) {\n\t\t\t// Add cache control to system prompt for non-OAuth tokens\n\t\t\tparams.system = [\n\t\t\t\t{\n\t\t\t\t\ttype: \"text\",\n\t\t\t\t\ttext: sanitizeSurrogates(systemPrompt),\n\t\t\t\t\t...(cacheControl ? { cache_control: cacheControl } : {}),\n\t\t\t\t},\n\t\t\t];\n\t\t}\n\t}\n\n\t// Temperature is incompatible with extended thinking (adaptive or budget-based).\n\t// SCRAMJET-DIVERGENCE: gate on supportsTemperature compat (opus-4-7+ rejects non-default temperature)\n\tif (\n\t\toptions?.temperature !== undefined &&\n\t\t!options?.thinkingEnabled &&\n\t\tgetAnthropicCompat(model).supportsTemperature\n\t) {\n\t\tparams.temperature = options.temperature;\n\t}\n\n\tif (context.tools && context.tools.length > 0) {\n\t\tconst compat = getAnthropicCompat(model);\n\t\tparams.tools = convertTools(\n\t\t\tcontext.tools,\n\t\t\tisOAuthToken,\n\t\t\tcompat.supportsEagerToolInputStreaming,\n\t\t\tcompat.supportsCacheControlOnTools ? cacheControl : undefined,\n\t\t);\n\t}\n\n\t// Configure thinking mode: adaptive-thinking models,\n\t// budget-based older models, or explicitly disabled.\n\tif (model.reasoning) {\n\t\tif (options?.thinkingEnabled) {\n\t\t\t// Default to \"summarized\" so Opus 4.7 and Mythos Preview behave like\n\t\t\t// older Claude 4 models (whose API default is also \"summarized\").\n\t\t\tconst display: AnthropicThinkingDisplay = options.thinkingDisplay ?? \"summarized\";\n\t\t\tif (supportsAdaptiveThinking(model.id)) {\n\t\t\t\t// Adaptive thinking: Claude decides when and how much to think.\n\t\t\t\tparams.thinking = { type: \"adaptive\", display };\n\t\t\t\tif (options.effort) {\n\t\t\t\t\t// The Anthropic SDK types can lag newly supported effort values such as \"xhigh\".\n\t\t\t\t\tparams.output_config =\n\t\t\t\t\t\toptions.effort === \"xhigh\"\n\t\t\t\t\t\t\t? ({ effort: options.effort } as unknown as NonNullable<\n\t\t\t\t\t\t\t\t\tMessageCreateParamsStreaming[\"output_config\"]\n\t\t\t\t\t\t\t\t>)\n\t\t\t\t\t\t\t: { effort: options.effort };\n\t\t\t\t}\n\t\t\t} else {\n\t\t\t\t// Budget-based thinking for older models\n\t\t\t\tparams.thinking = {\n\t\t\t\t\ttype: \"enabled\",\n\t\t\t\t\tbudget_tokens: options.thinkingBudgetTokens || 1024,\n\t\t\t\t\tdisplay,\n\t\t\t\t};\n\t\t\t}\n\t\t} else if (options?.thinkingEnabled === false) {\n\t\t\tparams.thinking = { type: \"disabled\" };\n\t\t}\n\t}\n\n\tif (options?.metadata) {\n\t\tconst userId = options.metadata.user_id;\n\t\tif (typeof userId === \"string\") {\n\t\t\tparams.metadata = { user_id: userId };\n\t\t}\n\t}\n\n\tif (options?.toolChoice) {\n\t\tif (typeof options.toolChoice === \"string\") {\n\t\t\tparams.tool_choice = { type: options.toolChoice };\n\t\t} else {\n\t\t\tparams.tool_choice = options.toolChoice;\n\t\t}\n\t}\n\n\treturn params;\n}\n\n// Normalize tool call IDs to match Anthropic's required pattern and length\nfunction normalizeToolCallId(id: string): string {\n\treturn id.replace(/[^a-zA-Z0-9_-]/g, \"_\").slice(0, 64);\n}\n\nfunction convertMessages(\n\tmessages: Message[],\n\tmodel: Model<\"anthropic-messages\">,\n\tisOAuthToken: boolean,\n\tcacheControl?: CacheControlEphemeral,\n): MessageParam[] {\n\tconst params: MessageParam[] = [];\n\n\t// Transform messages for cross-provider compatibility\n\tconst transformedMessages = transformMessages(messages, model, normalizeToolCallId);\n\n\tfor (let i = 0; i < transformedMessages.length; i++) {\n\t\tconst msg = transformedMessages[i];\n\n\t\tif (msg.role === \"user\") {\n\t\t\tif (typeof msg.content === \"string\") {\n\t\t\t\tif (msg.content.trim().length > 0) {\n\t\t\t\t\tparams.push({\n\t\t\t\t\t\trole: \"user\",\n\t\t\t\t\t\tcontent: sanitizeSurrogates(msg.content),\n\t\t\t\t\t});\n\t\t\t\t}\n\t\t\t} else {\n\t\t\t\tconst blocks: ContentBlockParam[] = msg.content.map((item) => {\n\t\t\t\t\tif (item.type === \"text\") {\n\t\t\t\t\t\treturn {\n\t\t\t\t\t\t\ttype: \"text\",\n\t\t\t\t\t\t\ttext: sanitizeSurrogates(item.text),\n\t\t\t\t\t\t};\n\t\t\t\t\t} else {\n\t\t\t\t\t\treturn {\n\t\t\t\t\t\t\ttype: \"image\",\n\t\t\t\t\t\t\tsource: {\n\t\t\t\t\t\t\t\ttype: \"base64\",\n\t\t\t\t\t\t\t\tmedia_type: item.mimeType as \"image/jpeg\" | \"image/png\" | \"image/gif\" | \"image/webp\",\n\t\t\t\t\t\t\t\tdata: item.data,\n\t\t\t\t\t\t\t},\n\t\t\t\t\t\t};\n\t\t\t\t\t}\n\t\t\t\t});\n\t\t\t\tconst filteredBlocks = blocks.filter((b) => {\n\t\t\t\t\tif (b.type === \"text\") {\n\t\t\t\t\t\treturn b.text.trim().length > 0;\n\t\t\t\t\t}\n\t\t\t\t\treturn true;\n\t\t\t\t});\n\t\t\t\tif (filteredBlocks.length === 0) continue;\n\t\t\t\tparams.push({\n\t\t\t\t\trole: \"user\",\n\t\t\t\t\tcontent: filteredBlocks,\n\t\t\t\t});\n\t\t\t}\n\t\t} else if (msg.role === \"assistant\") {\n\t\t\tconst blocks: ContentBlockParam[] = [];\n\n\t\t\tfor (const block of msg.content) {\n\t\t\t\tif (block.type === \"text\") {\n\t\t\t\t\tif (block.text.trim().length === 0) continue;\n\t\t\t\t\tblocks.push({\n\t\t\t\t\t\ttype: \"text\",\n\t\t\t\t\t\ttext: sanitizeSurrogates(block.text),\n\t\t\t\t\t});\n\t\t\t\t} else if (block.type === \"thinking\") {\n\t\t\t\t\t// Redacted thinking: pass the opaque payload back as redacted_thinking\n\t\t\t\t\tif (block.redacted) {\n\t\t\t\t\t\tblocks.push({\n\t\t\t\t\t\t\ttype: \"redacted_thinking\",\n\t\t\t\t\t\t\tdata: block.thinkingSignature!,\n\t\t\t\t\t\t});\n\t\t\t\t\t\tcontinue;\n\t\t\t\t\t}\n\t\t\t\t\tif (block.thinking.trim().length === 0) continue;\n\t\t\t\t\t// If thinking signature is missing/empty (e.g., from aborted stream),\n\t\t\t\t\t// convert to plain text block without <thinking> tags to avoid API rejection\n\t\t\t\t\t// and prevent Claude from mimicking the tags in responses\n\t\t\t\t\tif (!block.thinkingSignature || block.thinkingSignature.trim().length === 0) {\n\t\t\t\t\t\tblocks.push({\n\t\t\t\t\t\t\ttype: \"text\",\n\t\t\t\t\t\t\ttext: sanitizeSurrogates(block.thinking),\n\t\t\t\t\t\t});\n\t\t\t\t\t} else {\n\t\t\t\t\t\tblocks.push({\n\t\t\t\t\t\t\ttype: \"thinking\",\n\t\t\t\t\t\t\tthinking: sanitizeSurrogates(block.thinking),\n\t\t\t\t\t\t\tsignature: block.thinkingSignature,\n\t\t\t\t\t\t});\n\t\t\t\t\t}\n\t\t\t\t} else if (block.type === \"toolCall\") {\n\t\t\t\t\tblocks.push({\n\t\t\t\t\t\ttype: \"tool_use\",\n\t\t\t\t\t\t// SCRAMJET-DIVERGENCE: Sanitize tool-call IDs unconditionally on the outgoing\n\t\t\t\t\t\t// request. `transform-messages.ts` only normalizes IDs for cross-model replay\n\t\t\t\t\t\t// (`!isSameModel`), so a legacy session carrying a provider-invalid ID (e.g. one\n\t\t\t\t\t\t// embedding a raw model id like `gpt-5.5`) replayed to the SAME model would be sent\n\t\t\t\t\t\t// verbatim and rejected by Anthropic's `^[a-zA-Z0-9_-]+$` constraint. Applying the\n\t\t\t\t\t\t// idempotent `normalizeToolCallId` here (and at the matching tool_result sites below)\n\t\t\t\t\t\t// is harmless on already-clean IDs and preserves call/result correlation because the\n\t\t\t\t\t\t// same deterministic function maps both halves to the same value.\n\t\t\t\t\t\tid: normalizeToolCallId(block.id),\n\t\t\t\t\t\tname: isOAuthToken ? toClaudeCodeName(block.name) : block.name,\n\t\t\t\t\t\tinput: block.arguments ?? {},\n\t\t\t\t\t});\n\t\t\t\t}\n\t\t\t}\n\t\t\tif (blocks.length === 0) continue;\n\t\t\tparams.push({\n\t\t\t\trole: \"assistant\",\n\t\t\t\tcontent: blocks,\n\t\t\t});\n\t\t} else if (msg.role === \"toolResult\") {\n\t\t\t// Collect all consecutive toolResult messages, needed for z.ai Anthropic endpoint\n\t\t\tconst toolResults: ContentBlockParam[] = [];\n\n\t\t\t// Add the current tool result\n\t\t\ttoolResults.push({\n\t\t\t\ttype: \"tool_result\",\n\t\t\t\t// SCRAMJET-DIVERGENCE: sanitize to keep correlation with the tool_use id above.\n\t\t\t\ttool_use_id: normalizeToolCallId(msg.toolCallId),\n\t\t\t\tcontent: convertContentBlocks(msg.content),\n\t\t\t\tis_error: msg.isError,\n\t\t\t});\n\n\t\t\t// Look ahead for consecutive toolResult messages\n\t\t\tlet j = i + 1;\n\t\t\twhile (j < transformedMessages.length && transformedMessages[j].role === \"toolResult\") {\n\t\t\t\tconst nextMsg = transformedMessages[j] as ToolResultMessage; // We know it's a toolResult\n\t\t\t\ttoolResults.push({\n\t\t\t\t\ttype: \"tool_result\",\n\t\t\t\t\t// SCRAMJET-DIVERGENCE: sanitize to keep correlation with the tool_use id above.\n\t\t\t\t\ttool_use_id: normalizeToolCallId(nextMsg.toolCallId),\n\t\t\t\t\tcontent: convertContentBlocks(nextMsg.content),\n\t\t\t\t\tis_error: nextMsg.isError,\n\t\t\t\t});\n\t\t\t\tj++;\n\t\t\t}\n\n\t\t\t// Skip the messages we've already processed\n\t\t\ti = j - 1;\n\n\t\t\t// Add a single user message with all tool results\n\t\t\tparams.push({\n\t\t\t\trole: \"user\",\n\t\t\t\tcontent: toolResults,\n\t\t\t});\n\t\t}\n\t}\n\n\t// Add cache_control to the last user message to cache conversation history\n\tif (cacheControl && params.length > 0) {\n\t\tconst lastMessage = params[params.length - 1];\n\t\tif (lastMessage.role === \"user\") {\n\t\t\tif (Array.isArray(lastMessage.content)) {\n\t\t\t\tconst lastBlock = lastMessage.content[lastMessage.content.length - 1];\n\t\t\t\tif (\n\t\t\t\t\tlastBlock &&\n\t\t\t\t\t(lastBlock.type === \"text\" || lastBlock.type === \"image\" || lastBlock.type === \"tool_result\")\n\t\t\t\t) {\n\t\t\t\t\t(lastBlock as any).cache_control = cacheControl;\n\t\t\t\t}\n\t\t\t} else if (typeof lastMessage.content === \"string\") {\n\t\t\t\tlastMessage.content = [\n\t\t\t\t\t{\n\t\t\t\t\t\ttype: \"text\",\n\t\t\t\t\t\ttext: lastMessage.content,\n\t\t\t\t\t\tcache_control: cacheControl,\n\t\t\t\t\t},\n\t\t\t\t] as any;\n\t\t\t}\n\t\t}\n\t}\n\n\treturn params;\n}\n\nfunction shouldUseFineGrainedToolStreamingBeta(model: Model<\"anthropic-messages\">, context: Context): boolean {\n\treturn !!context.tools?.length && !getAnthropicCompat(model).supportsEagerToolInputStreaming;\n}\n\nfunction convertTools(\n\ttools: Tool[],\n\tisOAuthToken: boolean,\n\tsupportsEagerToolInputStreaming: boolean,\n\tcacheControl?: CacheControlEphemeral,\n): Anthropic.Messages.Tool[] {\n\tif (!tools) return [];\n\n\treturn tools.map((tool, index) => {\n\t\tconst schema = tool.parameters as { properties?: unknown; required?: string[] };\n\n\t\treturn {\n\t\t\tname: isOAuthToken ? toClaudeCodeName(tool.name) : tool.name,\n\t\t\tdescription: tool.description,\n\t\t\t...(supportsEagerToolInputStreaming ? { eager_input_streaming: true } : {}),\n\t\t\tinput_schema: {\n\t\t\t\ttype: \"object\",\n\t\t\t\tproperties: schema.properties ?? {},\n\t\t\t\trequired: schema.required ?? [],\n\t\t\t},\n\t\t\t...(cacheControl && index === tools.length - 1 ? { cache_control: cacheControl } : {}),\n\t\t};\n\t});\n}\n\nfunction mapStopReason(reason: Anthropic.Messages.StopReason | string): StopReason {\n\tswitch (reason) {\n\t\tcase \"end_turn\":\n\t\t\treturn \"stop\";\n\t\tcase \"max_tokens\":\n\t\t\treturn \"length\";\n\t\tcase \"tool_use\":\n\t\t\treturn \"toolUse\";\n\t\tcase \"refusal\":\n\t\t\treturn \"error\";\n\t\tcase \"pause_turn\": // Stop is good enough -> resubmit\n\t\t\treturn \"stop\";\n\t\tcase \"stop_sequence\":\n\t\t\treturn \"stop\"; // We don't supply stop sequences, so this should never happen\n\t\tcase \"sensitive\": // Content flagged by safety filters (not yet in SDK types)\n\t\t\treturn \"error\";\n\t\tdefault:\n\t\t\t// Handle unknown stop reasons gracefully (API may add new values)\n\t\t\tthrow new Error(`Unhandled stop reason: ${reason}`);\n\t}\n}\n"]}
+{"version":3,"file":"anthropic.d.ts","sourceRoot":"","sources":["../../src/providers/anthropic.ts"],"names":[],"mappings":"AAAA,OAAO,SAAS,MAAM,mBAAmB,CAAC;AAW1C,OAAO,KAAK,EASX,mBAAmB,EAEnB,cAAc,EACd,aAAa,EAOb,MAAM,aAAa,CAAC;AAsJrB,MAAM,MAAM,eAAe,GAAG,KAAK,GAAG,QAAQ,GAAG,MAAM,GAAG,OAAO,GAAG,KAAK,CAAC;AAE1E,MAAM,MAAM,wBAAwB,GAAG,YAAY,GAAG,SAAS,CAAC;AAsBhE,MAAM,WAAW,gBAAiB,SAAQ,aAAa;IACtD;;;;OAIG;IACH,eAAe,CAAC,EAAE,OAAO,CAAC;IAC1B;;;OAGG;IACH,oBAAoB,CAAC,EAAE,MAAM,CAAC;IAC9B;;;;;;;;;OASG;IACH,MAAM,CAAC,EAAE,eAAe,CAAC;IACzB;;;;;;;;;;OAUG;IACH,eAAe,CAAC,EAAE,wBAAwB,CAAC;IAC3C,mBAAmB,CAAC,EAAE,OAAO,CAAC;IAC9B,UAAU,CAAC,EAAE,MAAM,GAAG,KAAK,GAAG,MAAM,GAAG;QAAE,IAAI,EAAE,MAAM,CAAC;QAAC,IAAI,EAAE,MAAM,CAAA;KAAE,CAAC;IACtE;;;;OAIG;IACH,MAAM,CAAC,EAAE,SAAS,CAAC;CACnB;AAiPD,eAAO,MAAM,eAAe,EAAE,cAAc,CAAC,oBAAoB,EAAE,gBAAgB,CA8RlF,CAAC;AAwCF,eAAO,MAAM,qBAAqB,EAAE,cAAc,CAAC,oBAAoB,EAAE,mBAAmB,CA2C3F,CAAC","sourcesContent":["import Anthropic from \"@anthropic-ai/sdk\";\nimport type {\n\tCacheControlEphemeral,\n\tContentBlockParam,\n\tMessageCreateParamsStreaming,\n\tMessageParam,\n\tRawMessageStreamEvent,\n\tTextBlockParam,\n} from \"@anthropic-ai/sdk/resources/messages.js\";\nimport { getEnvApiKey } from \"../env-api-keys.js\";\nimport { calculateCost } from \"../models.js\";\nimport type {\n\tAnthropicMessagesCompat,\n\tApi,\n\tAssistantMessage,\n\tCacheRetention,\n\tContext,\n\tImageContent,\n\tMessage,\n\tModel,\n\tSimpleStreamOptions,\n\tStopReason,\n\tStreamFunction,\n\tStreamOptions,\n\tSystemPromptSection,\n\tTextContent,\n\tThinkingContent,\n\tTool,\n\tToolCall,\n\tToolResultMessage,\n} from \"../types.js\";\nimport { AssistantMessageEventStream } from \"../utils/event-stream.js\";\nimport {\n\tappendBuiltinFailure,\n\tfailureFromProviderError,\n\tinvokeProviderCallback,\n\tRequestFailureError,\n} from \"../utils/failure-evidence.js\";\nimport { headersToRecord } from \"../utils/headers.js\";\nimport { parseJsonWithRepair, parseStreamingJson } from \"../utils/json-parse.js\";\nimport { createProviderAbortScope } from \"../utils/provider-abort-scope.js\";\nimport { sanitizeSurrogates } from \"../utils/sanitize-unicode.js\";\nimport {\n\tADAPTIVE_THINKING_PATTERNS,\n\tNATIVE_XHIGH_EFFORT_PATTERNS,\n\tnormalizeForPatternMatch,\n} from \"./anthropic-model-patterns.js\";\n\nimport { resolveCloudflareBaseUrl } from \"./cloudflare.js\";\nimport { buildCopilotDynamicHeaders, hasCopilotVisionInput } from \"./github-copilot-headers.js\";\nimport { adjustMaxTokensForThinking, buildBaseOptions } from \"./simple-options.js\";\nimport { transformMessages } from \"./transform-messages.js\";\n\n/**\n * Resolve cache retention preference.\n *\n * Defaults to \"short\" and only honors PI_CACHE_RETENTION === \"long\" — this is\n * the low-level provider fallback for callers that hit the adapter directly.\n * In the coding-agent path this default rarely fires: the SDK's own\n * resolveCacheRetention (packages/coding-agent/src/core/sdk.ts, which defaults\n * to \"long\") resolves first and passes a concrete value down, so `cacheRetention`\n * is normally already set when it reaches here. Keep the two defaults in sync\n * intentionally — they differ on purpose (backward-compat vs. SDK default).\n */\nfunction resolveCacheRetention(cacheRetention?: CacheRetention): CacheRetention {\n\tif (cacheRetention) {\n\t\treturn cacheRetention;\n\t}\n\tif (typeof process !== \"undefined\" && process.env.PI_CACHE_RETENTION === \"long\") {\n\t\treturn \"long\";\n\t}\n\treturn \"short\";\n}\n\nfunction getCacheControl(\n\tmodel: Model<\"anthropic-messages\">,\n\tcacheRetention?: CacheRetention,\n): { retention: CacheRetention; cacheControl?: CacheControlEphemeral } {\n\tconst retention = resolveCacheRetention(cacheRetention);\n\tif (retention === \"none\") {\n\t\treturn { retention };\n\t}\n\tconst ttl = retention === \"long\" && getAnthropicCompat(model).supportsLongCacheRetention ? \"1h\" : undefined;\n\treturn {\n\t\tretention,\n\t\tcacheControl: { type: \"ephemeral\", ...(ttl && { ttl }) },\n\t};\n}\n\n// Stealth mode: Mimic Claude Code's tool naming exactly\nconst claudeCodeVersion = \"2.1.75\";\n\n// Claude Code 2.x tool names (canonical casing)\n// Source: https://cchistory.mariozechner.at/data/prompts-2.1.11.md\n// To update: https://github.com/badlogic/cchistory\nconst claudeCodeTools = [\n\t\"Read\",\n\t\"Write\",\n\t\"Edit\",\n\t\"Bash\",\n\t\"Grep\",\n\t\"Glob\",\n\t\"AskUserQuestion\",\n\t\"EnterPlanMode\",\n\t\"ExitPlanMode\",\n\t\"KillShell\",\n\t\"NotebookEdit\",\n\t\"Skill\",\n\t\"Task\",\n\t\"TaskOutput\",\n\t\"TodoWrite\",\n\t\"WebFetch\",\n\t\"WebSearch\",\n];\n\nconst ccToolLookup = new Map(claudeCodeTools.map((t) => [t.toLowerCase(), t]));\n\n// Convert tool name to CC canonical casing if it matches (case-insensitive)\nconst toClaudeCodeName = (name: string) => ccToolLookup.get(name.toLowerCase()) ?? name;\nconst fromClaudeCodeName = (name: string, tools?: Tool[]) => {\n\tif (tools && tools.length > 0) {\n\t\tconst lowerName = name.toLowerCase();\n\t\tconst matchedTool = tools.find((tool) => tool.name.toLowerCase() === lowerName);\n\t\tif (matchedTool) return matchedTool.name;\n\t}\n\treturn name;\n};\n\n/**\n * Convert content blocks to Anthropic API format\n */\nfunction convertContentBlocks(content: (TextContent | ImageContent)[]):\n\t| string\n\t| Array<\n\t\t\t| { type: \"text\"; text: string }\n\t\t\t| {\n\t\t\t\t\ttype: \"image\";\n\t\t\t\t\tsource: {\n\t\t\t\t\t\ttype: \"base64\";\n\t\t\t\t\t\tmedia_type: \"image/jpeg\" | \"image/png\" | \"image/gif\" | \"image/webp\";\n\t\t\t\t\t\tdata: string;\n\t\t\t\t\t};\n\t\t\t  }\n\t  > {\n\t// If only text blocks, return as concatenated string for simplicity\n\tconst hasImages = content.some((c) => c.type === \"image\");\n\tif (!hasImages) {\n\t\treturn sanitizeSurrogates(content.map((c) => (c as TextContent).text).join(\"\\n\"));\n\t}\n\n\t// If we have images, convert to content block array\n\tconst blocks = content.map((block) => {\n\t\tif (block.type === \"text\") {\n\t\t\treturn {\n\t\t\t\ttype: \"text\" as const,\n\t\t\t\ttext: sanitizeSurrogates(block.text),\n\t\t\t};\n\t\t}\n\t\treturn {\n\t\t\ttype: \"image\" as const,\n\t\t\tsource: {\n\t\t\t\ttype: \"base64\" as const,\n\t\t\t\tmedia_type: block.mimeType as \"image/jpeg\" | \"image/png\" | \"image/gif\" | \"image/webp\",\n\t\t\t\tdata: block.data,\n\t\t\t},\n\t\t};\n\t});\n\n\t// If only images (no text), add placeholder text block\n\tconst hasText = blocks.some((b) => b.type === \"text\");\n\tif (!hasText) {\n\t\tblocks.unshift({\n\t\t\ttype: \"text\" as const,\n\t\t\ttext: \"(see attached image)\",\n\t\t});\n\t}\n\n\treturn blocks;\n}\n\nexport type AnthropicEffort = \"low\" | \"medium\" | \"high\" | \"xhigh\" | \"max\";\n\nexport type AnthropicThinkingDisplay = \"summarized\" | \"omitted\";\n\nconst FINE_GRAINED_TOOL_STREAMING_BETA = \"fine-grained-tool-streaming-2025-05-14\";\nconst INTERLEAVED_THINKING_BETA = \"interleaved-thinking-2025-05-14\";\n\nfunction getAnthropicCompat(model: Model<\"anthropic-messages\">): Required<AnthropicMessagesCompat> {\n\t// Auto-detect session affinity and cache control support from provider\n\tconst isFireworks = model.provider === \"fireworks\";\n\tconst isCloudflareAiGatewayAnthropic =\n\t\tmodel.provider === \"cloudflare-ai-gateway\" && model.baseUrl.includes(\"anthropic\");\n\t// SCRAMJET-DIVERGENCE: supportsTemperature and forceAdaptiveThinking defaults added for Opus 4.8/Fable 5/Sonnet 5 support\n\treturn {\n\t\tsupportsEagerToolInputStreaming: model.compat?.supportsEagerToolInputStreaming ?? !isFireworks,\n\t\tsupportsLongCacheRetention: model.compat?.supportsLongCacheRetention ?? !isFireworks,\n\t\tsendSessionAffinityHeaders:\n\t\t\tmodel.compat?.sendSessionAffinityHeaders ?? !!(isFireworks || isCloudflareAiGatewayAnthropic),\n\t\tsupportsCacheControlOnTools: model.compat?.supportsCacheControlOnTools ?? !isFireworks,\n\t\tsupportsTemperature: model.compat?.supportsTemperature ?? true,\n\t\tforceAdaptiveThinking: model.compat?.forceAdaptiveThinking ?? false,\n\t};\n}\n\nexport interface AnthropicOptions extends StreamOptions {\n\t/**\n\t * Enable extended thinking.\n\t * For adaptive thinking models (Opus 4.6+, Sonnet 4.6+, Fable 5): model decides when/how much to think.\n\t * For older models: uses budget-based thinking with thinkingBudgetTokens.\n\t */\n\tthinkingEnabled?: boolean;\n\t/**\n\t * Token budget for extended thinking (older models only).\n\t * Ignored for adaptive thinking models (Opus 4.6+, Sonnet 4.6+, Fable 5).\n\t */\n\tthinkingBudgetTokens?: number;\n\t/**\n\t * Effort level for adaptive thinking (Opus 4.6+, Sonnet 4.6+, Fable 5).\n\t * Controls how much thinking Claude allocates:\n\t * - \"max\": Always thinks with no constraints (Opus 4.6 only)\n\t * - \"xhigh\": Highest reasoning level (Opus 4.7+, Fable 5)\n\t * - \"high\": Always thinks, deep reasoning (default)\n\t * - \"medium\": Moderate thinking, may skip for simple queries\n\t * - \"low\": Minimal thinking, skips for simple tasks\n\t * Ignored for older models.\n\t */\n\teffort?: AnthropicEffort;\n\t/**\n\t * Controls how thinking content is returned in API responses.\n\t * - \"summarized\": Thinking blocks contain summarized thinking text (default here).\n\t * - \"omitted\": Thinking blocks return an empty thinking field; the encrypted\n\t *   signature still travels back for multi-turn continuity. Use for faster\n\t *   time-to-first-text-token when your UI does not surface thinking.\n\t *\n\t * Note: Anthropic's API default for Claude Opus 4.7 and Claude Mythos Preview\n\t * is \"omitted\". We default to \"summarized\" here to keep behavior consistent\n\t * with older Claude 4 models. Set this explicitly to \"omitted\" to opt in.\n\t */\n\tthinkingDisplay?: AnthropicThinkingDisplay;\n\tinterleavedThinking?: boolean;\n\ttoolChoice?: \"auto\" | \"any\" | \"none\" | { type: \"tool\"; name: string };\n\t/**\n\t * Pre-built Anthropic client instance. When provided, skips internal client\n\t * construction entirely. Use this to inject alternative SDK clients such as\n\t * `AnthropicVertex` that shares the same messaging API.\n\t */\n\tclient?: Anthropic;\n}\n\nfunction mergeHeaders(...headerSources: (Record<string, string | null> | undefined)[]): Record<string, string | null> {\n\tconst merged: Record<string, string | null> = {};\n\tfor (const headers of headerSources) {\n\t\tif (headers) {\n\t\t\tObject.assign(merged, headers);\n\t\t}\n\t}\n\treturn merged;\n}\n\ninterface ServerSentEvent {\n\tevent: string | null;\n\tdata: string;\n\traw: string[];\n}\n\ninterface SseDecoderState {\n\tevent: string | null;\n\tdata: string[];\n\traw: string[];\n}\n\nconst ANTHROPIC_MESSAGE_EVENTS: ReadonlySet<string> = new Set([\n\t\"message_start\",\n\t\"message_delta\",\n\t\"message_stop\",\n\t\"content_block_start\",\n\t\"content_block_delta\",\n\t\"content_block_stop\",\n]);\n\nfunction flushSseEvent(state: SseDecoderState): ServerSentEvent | null {\n\tif (!state.event && state.data.length === 0) {\n\t\treturn null;\n\t}\n\n\tconst event: ServerSentEvent = {\n\t\tevent: state.event,\n\t\tdata: state.data.join(\"\\n\"),\n\t\traw: [...state.raw],\n\t};\n\tstate.event = null;\n\tstate.data = [];\n\tstate.raw = [];\n\treturn event;\n}\n\nfunction decodeSseLine(line: string, state: SseDecoderState): ServerSentEvent | null {\n\tif (line === \"\") {\n\t\treturn flushSseEvent(state);\n\t}\n\n\tstate.raw.push(line);\n\tif (line.startsWith(\":\")) {\n\t\treturn null;\n\t}\n\n\tconst delimiterIndex = line.indexOf(\":\");\n\tconst fieldName = delimiterIndex === -1 ? line : line.slice(0, delimiterIndex);\n\tlet value = delimiterIndex === -1 ? \"\" : line.slice(delimiterIndex + 1);\n\tif (value.startsWith(\" \")) {\n\t\tvalue = value.slice(1);\n\t}\n\n\tif (fieldName === \"event\") {\n\t\tstate.event = value;\n\t} else if (fieldName === \"data\") {\n\t\tstate.data.push(value);\n\t}\n\n\treturn null;\n}\n\nfunction nextLineBreakIndex(text: string): number {\n\tconst carriageReturnIndex = text.indexOf(\"\\r\");\n\tconst newlineIndex = text.indexOf(\"\\n\");\n\tif (carriageReturnIndex === -1) {\n\t\treturn newlineIndex;\n\t}\n\tif (newlineIndex === -1) {\n\t\treturn carriageReturnIndex;\n\t}\n\treturn Math.min(carriageReturnIndex, newlineIndex);\n}\n\nfunction consumeLine(text: string): { line: string; rest: string } | null {\n\tconst lineBreakIndex = nextLineBreakIndex(text);\n\tif (lineBreakIndex === -1) {\n\t\treturn null;\n\t}\n\n\tlet nextIndex = lineBreakIndex + 1;\n\tif (text[lineBreakIndex] === \"\\r\" && text[nextIndex] === \"\\n\") {\n\t\tnextIndex += 1;\n\t}\n\n\treturn {\n\t\tline: text.slice(0, lineBreakIndex),\n\t\trest: text.slice(nextIndex),\n\t};\n}\n\nasync function* iterateSseMessages(\n\tbody: ReadableStream<Uint8Array>,\n\tsignal?: AbortSignal,\n): AsyncGenerator<ServerSentEvent> {\n\tconst reader = body.getReader();\n\tconst decoder = new TextDecoder();\n\tconst state: SseDecoderState = { event: null, data: [], raw: [] };\n\tlet buffer = \"\";\n\n\ttry {\n\t\twhile (true) {\n\t\t\tif (signal?.aborted) {\n\t\t\t\tthrow new Error(\"Request was aborted\");\n\t\t\t}\n\n\t\t\tconst { value, done } = await reader.read();\n\t\t\tif (done) {\n\t\t\t\tbreak;\n\t\t\t}\n\n\t\t\tbuffer += decoder.decode(value, { stream: true });\n\t\t\tlet consumed = consumeLine(buffer);\n\t\t\twhile (consumed) {\n\t\t\t\tbuffer = consumed.rest;\n\t\t\t\tconst event = decodeSseLine(consumed.line, state);\n\t\t\t\tif (event) {\n\t\t\t\t\tyield event;\n\t\t\t\t}\n\t\t\t\tconsumed = consumeLine(buffer);\n\t\t\t}\n\t\t}\n\n\t\tbuffer += decoder.decode();\n\t\tlet consumed = consumeLine(buffer);\n\t\twhile (consumed) {\n\t\t\tbuffer = consumed.rest;\n\t\t\tconst event = decodeSseLine(consumed.line, state);\n\t\t\tif (event) {\n\t\t\t\tyield event;\n\t\t\t}\n\t\t\tconsumed = consumeLine(buffer);\n\t\t}\n\n\t\tif (buffer.length > 0) {\n\t\t\tconst event = decodeSseLine(buffer, state);\n\t\t\tif (event) {\n\t\t\t\tyield event;\n\t\t\t}\n\t\t}\n\n\t\tconst trailingEvent = flushSseEvent(state);\n\t\tif (trailingEvent) {\n\t\t\tyield trailingEvent;\n\t\t}\n\t} finally {\n\t\treader.releaseLock();\n\t}\n}\n\n// SCRAMJET-DIVERGENCE: Preserve provider and callback failure facts independently.\nasync function* iterateAnthropicEvents(\n\tresponse: Response,\n\tsignal?: AbortSignal,\n): AsyncGenerator<RawMessageStreamEvent> {\n\tif (!response.body) {\n\t\tthrow new RequestFailureError(\"Anthropic response has no body.\", {\n\t\t\tschemaVersion: 1,\n\t\t\tkind: \"stream\",\n\t\t\treason: \"missing_body\",\n\t\t});\n\t}\n\n\tlet sawMessageStart = false;\n\tlet sawMessageEnd = false;\n\n\tfor await (const sse of iterateSseMessages(response.body, signal)) {\n\t\tif (sse.event === \"error\") {\n\t\t\tlet value: unknown;\n\t\t\ttry {\n\t\t\t\tvalue = JSON.parse(sse.data);\n\t\t\t} catch {\n\t\t\t\tthrow new RequestFailureError(\"Malformed Anthropic error event.\", {\n\t\t\t\t\tschemaVersion: 1,\n\t\t\t\t\tkind: \"stream\",\n\t\t\t\t\treason: \"malformed_event\",\n\t\t\t\t});\n\t\t\t}\n\t\t\tif (\n\t\t\t\t!value ||\n\t\t\t\ttypeof value !== \"object\" ||\n\t\t\t\tArray.isArray(value) ||\n\t\t\t\t!(\"error\" in value) ||\n\t\t\t\t!value.error ||\n\t\t\t\ttypeof value.error !== \"object\" ||\n\t\t\t\tArray.isArray(value.error)\n\t\t\t)\n\t\t\t\tthrow new RequestFailureError(\"Malformed Anthropic error event.\", {\n\t\t\t\t\tschemaVersion: 1,\n\t\t\t\t\tkind: \"stream\",\n\t\t\t\t\treason: \"malformed_event\",\n\t\t\t\t});\n\t\t\tthrow new RequestFailureError(\n\t\t\t\tsse.data,\n\t\t\t\tfailureFromProviderError(value) ?? { schemaVersion: 1, kind: \"provider\", category: \"unknown\" },\n\t\t\t);\n\t\t}\n\n\t\tif (!ANTHROPIC_MESSAGE_EVENTS.has(sse.event ?? \"\")) {\n\t\t\tcontinue;\n\t\t}\n\n\t\ttry {\n\t\t\tconst event = parseJsonWithRepair<RawMessageStreamEvent>(sse.data);\n\t\t\tif (event.type === \"message_start\") {\n\t\t\t\tsawMessageStart = true;\n\t\t\t} else if (event.type === \"message_stop\") {\n\t\t\t\tsawMessageEnd = true;\n\t\t\t}\n\t\t\tyield event;\n\t\t} catch (error) {\n\t\t\tconst message = error instanceof Error ? error.message : String(error);\n\t\t\tthrow new RequestFailureError(\n\t\t\t\t`Could not parse Anthropic SSE event ${sse.event}: ${message}; data=${sse.data}; raw=${sse.raw.join(\"\\\\n\")}`,\n\t\t\t\t{ schemaVersion: 1, kind: \"stream\", reason: \"malformed_event\" },\n\t\t\t);\n\t\t}\n\t}\n\n\tif (sawMessageStart && !sawMessageEnd) {\n\t\tthrow new RequestFailureError(\"Anthropic stream ended before message_stop\", {\n\t\t\tschemaVersion: 1,\n\t\t\tkind: \"stream\",\n\t\t\treason: \"missing_terminal_event\",\n\t\t});\n\t}\n}\n\nexport const streamAnthropic: StreamFunction<\"anthropic-messages\", AnthropicOptions> = (\n\tmodel: Model<\"anthropic-messages\">,\n\tcontext: Context,\n\toptions?: AnthropicOptions,\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: model.api 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\tlet requestStarted = false;\n\t\tconst abortScope = createProviderAbortScope(options?.signal);\n\t\ttry {\n\t\t\tlet client: Anthropic;\n\t\t\tlet isOAuth: boolean;\n\n\t\t\tif (options?.client) {\n\t\t\t\tclient = options.client;\n\t\t\t\tisOAuth = false;\n\t\t\t} else {\n\t\t\t\tconst apiKey = options?.apiKey ?? getEnvApiKey(model.provider) ?? \"\";\n\n\t\t\t\tlet copilotDynamicHeaders: Record<string, string> | undefined;\n\t\t\t\tif (model.provider === \"github-copilot\") {\n\t\t\t\t\tconst hasImages = hasCopilotVisionInput(context.messages);\n\t\t\t\t\tcopilotDynamicHeaders = buildCopilotDynamicHeaders({\n\t\t\t\t\t\tmessages: context.messages,\n\t\t\t\t\t\thasImages,\n\t\t\t\t\t});\n\t\t\t\t}\n\n\t\t\t\tconst cacheRetention = options?.cacheRetention ?? resolveCacheRetention();\n\t\t\t\tconst cacheSessionId = cacheRetention === \"none\" ? undefined : options?.sessionId;\n\n\t\t\t\tconst created = createClient(\n\t\t\t\t\tmodel,\n\t\t\t\t\tapiKey,\n\t\t\t\t\toptions?.interleavedThinking ?? true,\n\t\t\t\t\tshouldUseFineGrainedToolStreamingBeta(model, context),\n\t\t\t\t\toptions?.headers,\n\t\t\t\t\tcopilotDynamicHeaders,\n\t\t\t\t\tcacheSessionId,\n\t\t\t\t\t(input, init) => {\n\t\t\t\t\t\trequestStarted = true;\n\t\t\t\t\t\treturn fetch(input, init);\n\t\t\t\t\t},\n\t\t\t\t);\n\t\t\t\tclient = created.client;\n\t\t\t\tisOAuth = created.isOAuthToken;\n\t\t\t}\n\t\t\tlet params = buildParams(model, context, isOAuth, options);\n\t\t\tconst nextParams = await invokeProviderCallback(\"onPayload\", () => options?.onPayload?.(params, model));\n\t\t\tif (nextParams !== undefined) {\n\t\t\t\tparams = nextParams as MessageCreateParamsStreaming;\n\t\t\t}\n\t\t\tconst requestOptions = {\n\t\t\t\tsignal: abortScope.signal,\n\t\t\t\t...(options?.timeoutMs !== undefined ? { timeout: options.timeoutMs } : {}),\n\t\t\t\t...(options?.maxRetries !== undefined ? { maxRetries: options.maxRetries } : {}),\n\t\t\t};\n\t\t\tif (options?.client) requestStarted = true;\n\t\t\tconst response = await client.messages.create({ ...params, stream: true }, requestOptions).asResponse();\n\t\t\tawait invokeProviderCallback(\"onResponse\", () =>\n\t\t\t\toptions?.onResponse?.({ status: response.status, headers: headersToRecord(response.headers) }, model),\n\t\t\t);\n\t\t\tstream.push({ type: \"start\", partial: output });\n\n\t\t\ttype Block = (ThinkingContent | TextContent | (ToolCall & { partialJson: string })) & { index: number };\n\t\t\tconst blocks = output.content as Block[];\n\n\t\t\tfor await (const event of iterateAnthropicEvents(response, abortScope.signal)) {\n\t\t\t\tif (event.type === \"message_start\") {\n\t\t\t\t\toutput.responseId = event.message.id;\n\t\t\t\t\t// Capture initial token usage from message_start event\n\t\t\t\t\t// This ensures we have input token counts even if the stream is aborted early\n\t\t\t\t\toutput.usage.input = event.message.usage.input_tokens || 0;\n\t\t\t\t\toutput.usage.output = event.message.usage.output_tokens || 0;\n\t\t\t\t\toutput.usage.cacheRead = event.message.usage.cache_read_input_tokens || 0;\n\t\t\t\t\toutput.usage.cacheWrite = event.message.usage.cache_creation_input_tokens || 0;\n\t\t\t\t\t// Anthropic doesn't provide total_tokens, compute from components\n\t\t\t\t\toutput.usage.totalTokens =\n\t\t\t\t\t\toutput.usage.input + output.usage.output + output.usage.cacheRead + output.usage.cacheWrite;\n\t\t\t\t\tcalculateCost(model, output.usage);\n\t\t\t\t} else if (event.type === \"content_block_start\") {\n\t\t\t\t\tif (event.content_block.type === \"text\") {\n\t\t\t\t\t\tconst block: Block = {\n\t\t\t\t\t\t\ttype: \"text\",\n\t\t\t\t\t\t\ttext: \"\",\n\t\t\t\t\t\t\tindex: event.index,\n\t\t\t\t\t\t};\n\t\t\t\t\t\toutput.content.push(block);\n\t\t\t\t\t\tstream.push({ type: \"text_start\", contentIndex: output.content.length - 1, partial: output });\n\t\t\t\t\t} else if (event.content_block.type === \"thinking\") {\n\t\t\t\t\t\tconst block: Block = {\n\t\t\t\t\t\t\ttype: \"thinking\",\n\t\t\t\t\t\t\tthinking: \"\",\n\t\t\t\t\t\t\tthinkingSignature: \"\",\n\t\t\t\t\t\t\tindex: event.index,\n\t\t\t\t\t\t};\n\t\t\t\t\t\toutput.content.push(block);\n\t\t\t\t\t\tstream.push({ type: \"thinking_start\", contentIndex: output.content.length - 1, partial: output });\n\t\t\t\t\t} else if (event.content_block.type === \"redacted_thinking\") {\n\t\t\t\t\t\tconst block: Block = {\n\t\t\t\t\t\t\ttype: \"thinking\",\n\t\t\t\t\t\t\tthinking: \"[Reasoning redacted]\",\n\t\t\t\t\t\t\tthinkingSignature: event.content_block.data,\n\t\t\t\t\t\t\tredacted: true,\n\t\t\t\t\t\t\tindex: event.index,\n\t\t\t\t\t\t};\n\t\t\t\t\t\toutput.content.push(block);\n\t\t\t\t\t\tstream.push({ type: \"thinking_start\", contentIndex: output.content.length - 1, partial: output });\n\t\t\t\t\t} else if (event.content_block.type === \"tool_use\") {\n\t\t\t\t\t\tconst block: Block = {\n\t\t\t\t\t\t\ttype: \"toolCall\",\n\t\t\t\t\t\t\tid: event.content_block.id,\n\t\t\t\t\t\t\tname: isOAuth\n\t\t\t\t\t\t\t\t? fromClaudeCodeName(event.content_block.name, context.tools)\n\t\t\t\t\t\t\t\t: event.content_block.name,\n\t\t\t\t\t\t\targuments: (event.content_block.input as Record<string, any>) ?? {},\n\t\t\t\t\t\t\tpartialJson: \"\",\n\t\t\t\t\t\t\tindex: event.index,\n\t\t\t\t\t\t};\n\t\t\t\t\t\toutput.content.push(block);\n\t\t\t\t\t\tstream.push({ type: \"toolcall_start\", contentIndex: output.content.length - 1, partial: output });\n\t\t\t\t\t}\n\t\t\t\t} else if (event.type === \"content_block_delta\") {\n\t\t\t\t\tif (event.delta.type === \"text_delta\") {\n\t\t\t\t\t\tconst index = blocks.findIndex((b) => b.index === event.index);\n\t\t\t\t\t\tconst block = blocks[index];\n\t\t\t\t\t\tif (block && block.type === \"text\") {\n\t\t\t\t\t\t\tblock.text += event.delta.text;\n\t\t\t\t\t\t\tstream.push({\n\t\t\t\t\t\t\t\ttype: \"text_delta\",\n\t\t\t\t\t\t\t\tcontentIndex: index,\n\t\t\t\t\t\t\t\tdelta: event.delta.text,\n\t\t\t\t\t\t\t\tpartial: output,\n\t\t\t\t\t\t\t});\n\t\t\t\t\t\t}\n\t\t\t\t\t} else if (event.delta.type === \"thinking_delta\") {\n\t\t\t\t\t\tconst index = blocks.findIndex((b) => b.index === event.index);\n\t\t\t\t\t\tconst block = blocks[index];\n\t\t\t\t\t\tif (block && block.type === \"thinking\") {\n\t\t\t\t\t\t\tblock.thinking += event.delta.thinking;\n\t\t\t\t\t\t\tstream.push({\n\t\t\t\t\t\t\t\ttype: \"thinking_delta\",\n\t\t\t\t\t\t\t\tcontentIndex: index,\n\t\t\t\t\t\t\t\tdelta: event.delta.thinking,\n\t\t\t\t\t\t\t\tpartial: output,\n\t\t\t\t\t\t\t});\n\t\t\t\t\t\t}\n\t\t\t\t\t} else if (event.delta.type === \"input_json_delta\") {\n\t\t\t\t\t\tconst index = blocks.findIndex((b) => b.index === event.index);\n\t\t\t\t\t\tconst block = blocks[index];\n\t\t\t\t\t\tif (block && block.type === \"toolCall\") {\n\t\t\t\t\t\t\tblock.partialJson += event.delta.partial_json;\n\t\t\t\t\t\t\tblock.arguments = parseStreamingJson(block.partialJson);\n\t\t\t\t\t\t\tstream.push({\n\t\t\t\t\t\t\t\ttype: \"toolcall_delta\",\n\t\t\t\t\t\t\t\tcontentIndex: index,\n\t\t\t\t\t\t\t\tdelta: event.delta.partial_json,\n\t\t\t\t\t\t\t\tpartial: output,\n\t\t\t\t\t\t\t});\n\t\t\t\t\t\t}\n\t\t\t\t\t} else if (event.delta.type === \"signature_delta\") {\n\t\t\t\t\t\tconst index = blocks.findIndex((b) => b.index === event.index);\n\t\t\t\t\t\tconst block = blocks[index];\n\t\t\t\t\t\tif (block && block.type === \"thinking\") {\n\t\t\t\t\t\t\tblock.thinkingSignature = block.thinkingSignature || \"\";\n\t\t\t\t\t\t\tblock.thinkingSignature += event.delta.signature;\n\t\t\t\t\t\t}\n\t\t\t\t\t}\n\t\t\t\t} else if (event.type === \"content_block_stop\") {\n\t\t\t\t\tconst index = blocks.findIndex((b) => b.index === event.index);\n\t\t\t\t\tconst block = blocks[index];\n\t\t\t\t\tif (block) {\n\t\t\t\t\t\tdelete (block as any).index;\n\t\t\t\t\t\tif (block.type === \"text\") {\n\t\t\t\t\t\t\tstream.push({\n\t\t\t\t\t\t\t\ttype: \"text_end\",\n\t\t\t\t\t\t\t\tcontentIndex: index,\n\t\t\t\t\t\t\t\tcontent: block.text,\n\t\t\t\t\t\t\t\tpartial: output,\n\t\t\t\t\t\t\t});\n\t\t\t\t\t\t} else if (block.type === \"thinking\") {\n\t\t\t\t\t\t\tstream.push({\n\t\t\t\t\t\t\t\ttype: \"thinking_end\",\n\t\t\t\t\t\t\t\tcontentIndex: index,\n\t\t\t\t\t\t\t\tcontent: block.thinking,\n\t\t\t\t\t\t\t\tpartial: output,\n\t\t\t\t\t\t\t});\n\t\t\t\t\t\t} else if (block.type === \"toolCall\") {\n\t\t\t\t\t\t\tblock.arguments = parseStreamingJson(block.partialJson);\n\t\t\t\t\t\t\t// Finalize in-place and strip the scratch buffer so replay only\n\t\t\t\t\t\t\t// carries parsed arguments.\n\t\t\t\t\t\t\tdelete (block as { partialJson?: string }).partialJson;\n\t\t\t\t\t\t\tstream.push({\n\t\t\t\t\t\t\t\ttype: \"toolcall_end\",\n\t\t\t\t\t\t\t\tcontentIndex: index,\n\t\t\t\t\t\t\t\ttoolCall: block,\n\t\t\t\t\t\t\t\tpartial: output,\n\t\t\t\t\t\t\t});\n\t\t\t\t\t\t}\n\t\t\t\t\t}\n\t\t\t\t} else if (event.type === \"message_delta\") {\n\t\t\t\t\tif (event.delta.stop_reason) {\n\t\t\t\t\t\tif (output.stopReason !== \"error\") output.stopReason = mapStopReason(event.delta.stop_reason);\n\t\t\t\t\t}\n\t\t\t\t\t// Only update usage fields if present (not null).\n\t\t\t\t\t// Preserves input_tokens from message_start when proxies omit it in message_delta.\n\t\t\t\t\tif (event.usage.input_tokens != null) {\n\t\t\t\t\t\toutput.usage.input = event.usage.input_tokens;\n\t\t\t\t\t}\n\t\t\t\t\tif (event.usage.output_tokens != null) {\n\t\t\t\t\t\toutput.usage.output = event.usage.output_tokens;\n\t\t\t\t\t}\n\t\t\t\t\tif (event.usage.cache_read_input_tokens != null) {\n\t\t\t\t\t\toutput.usage.cacheRead = event.usage.cache_read_input_tokens;\n\t\t\t\t\t}\n\t\t\t\t\tif (event.usage.cache_creation_input_tokens != null) {\n\t\t\t\t\t\toutput.usage.cacheWrite = event.usage.cache_creation_input_tokens;\n\t\t\t\t\t}\n\t\t\t\t\t// Anthropic doesn't provide total_tokens, compute from components\n\t\t\t\t\toutput.usage.totalTokens =\n\t\t\t\t\t\toutput.usage.input + output.usage.output + output.usage.cacheRead + output.usage.cacheWrite;\n\t\t\t\t\tcalculateCost(model, output.usage);\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\")\n\t\t\t\tthrow new RequestFailureError(\"Anthropic rejected the output.\", {\n\t\t\t\t\tschemaVersion: 1,\n\t\t\t\t\tkind: \"provider\",\n\t\t\t\t\tcategory: \"content_rejection\",\n\t\t\t\t});\n\t\t\tif (output.stopReason === \"aborted\") throw new Error(\"Request was aborted\");\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 { index?: number }).index;\n\t\t\t\t// partialJson is only a streaming scratch buffer; never persist it.\n\t\t\t\tdelete (block as { partialJson?: string }).partialJson;\n\t\t\t}\n\t\t\toutput.stopReason = options?.signal?.aborted ? \"aborted\" : \"error\";\n\t\t\toutput.errorMessage = error instanceof Error ? error.message : JSON.stringify(error);\n\t\t\tappendBuiltinFailure(\n\t\t\t\toutput,\n\t\t\t\trequestStarted && !options?.signal?.aborted && error instanceof Anthropic.APIConnectionError\n\t\t\t\t\t? new RequestFailureError(output.errorMessage, {\n\t\t\t\t\t\t\tschemaVersion: 1,\n\t\t\t\t\t\t\tkind: \"provider\",\n\t\t\t\t\t\t\tcategory: error instanceof Anthropic.APIConnectionTimeoutError ? \"timeout\" : \"transport\",\n\t\t\t\t\t\t})\n\t\t\t\t\t: error,\n\t\t\t\t!requestStarted,\n\t\t\t);\n\t\t\tstream.push({ type: \"error\", reason: output.stopReason, error: output });\n\t\t\tstream.end();\n\t\t} finally {\n\t\t\t// SCRAMJET-DIVERGENCE: finish cleanup synchronously before terminal consumers resume (#587).\n\t\t\tabortScope.dispose();\n\t\t}\n\t})();\n\n\treturn stream;\n};\n\n// SCRAMJET-DIVERGENCE: extended with opus-4-8, fable-5, sonnet-5 patterns for new model support\nfunction supportsAdaptiveThinking(modelId: string): boolean {\n\tconst normalized = normalizeForPatternMatch(modelId);\n\treturn ADAPTIVE_THINKING_PATTERNS.some((p) => normalized.includes(p));\n}\n\n// SCRAMJET-DIVERGENCE: native xhigh effort support for Opus 4.8/Fable 5 models\nfunction supportsNativeXhighEffort(modelId: string): boolean {\n\tconst normalized = normalizeForPatternMatch(modelId);\n\treturn NATIVE_XHIGH_EFFORT_PATTERNS.some((p) => normalized.includes(p));\n}\n\n/**\n * Map ThinkingLevel to Anthropic effort levels for adaptive thinking.\n * Note: effort \"max\" is only valid on Opus 4.6; Opus 4.7+ and Fable 5 support \"xhigh\".\n */\nfunction mapThinkingLevelToEffort(\n\tmodel: Model<\"anthropic-messages\">,\n\tlevel: SimpleStreamOptions[\"reasoning\"],\n): AnthropicEffort {\n\tconst mapped = level ? model.thinkingLevelMap?.[level] : undefined;\n\tif (typeof mapped === \"string\") return mapped as AnthropicEffort;\n\n\tswitch (level) {\n\t\tcase \"minimal\":\n\t\tcase \"low\":\n\t\t\treturn \"low\";\n\t\tcase \"medium\":\n\t\t\treturn \"medium\";\n\t\tcase \"high\":\n\t\t\treturn \"high\";\n\t\tcase \"xhigh\":\n\t\t\treturn supportsNativeXhighEffort(model.id) ? \"xhigh\" : \"high\";\n\t\tdefault:\n\t\t\treturn \"high\";\n\t}\n}\n\nexport const streamSimpleAnthropic: StreamFunction<\"anthropic-messages\", SimpleStreamOptions> = (\n\tmodel: Model<\"anthropic-messages\">,\n\tcontext: Context,\n\toptions?: SimpleStreamOptions,\n): AssistantMessageEventStream => {\n\tconst apiKey = options?.apiKey || getEnvApiKey(model.provider);\n\tif (!apiKey) {\n\t\tthrow new Error(`No API key for provider: ${model.provider}`);\n\t}\n\n\tconst base = buildBaseOptions(model, options, apiKey);\n\tif (!options?.reasoning) {\n\t\treturn streamAnthropic(model, context, { ...base, thinkingEnabled: false } satisfies AnthropicOptions);\n\t}\n\n\t// For adaptive-thinking models: use effort level.\n\t// For older models: use budget-based thinking.\n\tif (supportsAdaptiveThinking(model.id)) {\n\t\tconst effort = mapThinkingLevelToEffort(model, options.reasoning);\n\t\treturn streamAnthropic(model, context, {\n\t\t\t...base,\n\t\t\tthinkingEnabled: true,\n\t\t\teffort,\n\t\t} satisfies AnthropicOptions);\n\t}\n\n\tconst adjusted = adjustMaxTokensForThinking(\n\t\tbase.maxTokens || 0,\n\t\tMath.min(model.maxTokens, options.maxTokens ?? Infinity),\n\t\toptions.reasoning,\n\t\toptions.thinkingBudgets,\n\t);\n\n\t// SCRAMJET-DIVERGENCE: An explicit output ceiling includes thinking; never resurrect a zero budget.\n\tif (adjusted.thinkingBudget < 1024) {\n\t\tthrow new Error(\"Insufficient output space for Anthropic thinking; reduce input or request more output space.\");\n\t}\n\treturn streamAnthropic(model, context, {\n\t\t...base,\n\t\tmaxTokens: adjusted.maxTokens,\n\t\tthinkingEnabled: true,\n\t\tthinkingBudgetTokens: adjusted.thinkingBudget,\n\t} satisfies AnthropicOptions);\n};\n\nfunction isOAuthToken(apiKey: string): boolean {\n\treturn apiKey.includes(\"sk-ant-oat\");\n}\n\nfunction createClient(\n\tmodel: Model<\"anthropic-messages\">,\n\tapiKey: string,\n\tinterleavedThinking: boolean,\n\tuseFineGrainedToolStreamingBeta: boolean,\n\toptionsHeaders?: Record<string, string>,\n\tdynamicHeaders?: Record<string, string>,\n\tsessionId?: string,\n\tfetchImplementation?: typeof fetch,\n): { client: Anthropic; isOAuthToken: boolean } {\n\t// Adaptive-thinking models have interleaved thinking built in, so skip the beta header.\n\tconst needsInterleavedBeta = interleavedThinking && !supportsAdaptiveThinking(model.id);\n\tconst betaFeatures: string[] = [];\n\tif (useFineGrainedToolStreamingBeta) {\n\t\tbetaFeatures.push(FINE_GRAINED_TOOL_STREAMING_BETA);\n\t}\n\tif (needsInterleavedBeta) {\n\t\tbetaFeatures.push(INTERLEAVED_THINKING_BETA);\n\t}\n\n\tif (model.provider === \"cloudflare-ai-gateway\") {\n\t\tconst client = new Anthropic({\n\t\t\tapiKey: null,\n\t\t\tauthToken: null,\n\t\t\tbaseURL: resolveCloudflareBaseUrl(model),\n\t\t\tfetch: fetchImplementation,\n\t\t\tdangerouslyAllowBrowser: true,\n\t\t\tdefaultHeaders: mergeHeaders(\n\t\t\t\t{\n\t\t\t\t\taccept: \"application/json\",\n\t\t\t\t\t\"anthropic-dangerous-direct-browser-access\": \"true\",\n\t\t\t\t\t\"cf-aig-authorization\": `Bearer ${apiKey}`,\n\t\t\t\t\t\"x-api-key\": null,\n\t\t\t\t\tAuthorization: null,\n\t\t\t\t\t...(betaFeatures.length > 0 ? { \"anthropic-beta\": betaFeatures.join(\",\") } : {}),\n\t\t\t\t},\n\t\t\t\tmodel.headers,\n\t\t\t\toptionsHeaders,\n\t\t\t),\n\t\t});\n\n\t\treturn { client, isOAuthToken: false };\n\t}\n\n\t// Copilot: Bearer auth, selective betas.\n\tif (model.provider === \"github-copilot\") {\n\t\tconst client = new Anthropic({\n\t\t\tfetch: fetchImplementation,\n\t\t\tapiKey: null,\n\t\t\tauthToken: apiKey,\n\t\t\tbaseURL: model.baseUrl,\n\t\t\tdangerouslyAllowBrowser: true,\n\t\t\tdefaultHeaders: mergeHeaders(\n\t\t\t\t{\n\t\t\t\t\taccept: \"application/json\",\n\t\t\t\t\t\"anthropic-dangerous-direct-browser-access\": \"true\",\n\t\t\t\t\t...(betaFeatures.length > 0 ? { \"anthropic-beta\": betaFeatures.join(\",\") } : {}),\n\t\t\t\t},\n\t\t\t\tmodel.headers,\n\t\t\t\tdynamicHeaders,\n\t\t\t\toptionsHeaders,\n\t\t\t),\n\t\t});\n\n\t\treturn { client, isOAuthToken: false };\n\t}\n\n\t// OAuth: Bearer auth, Claude Code identity headers\n\tif (isOAuthToken(apiKey)) {\n\t\tconst client = new Anthropic({\n\t\t\tfetch: fetchImplementation,\n\t\t\tapiKey: null,\n\t\t\tauthToken: apiKey,\n\t\t\tbaseURL: model.baseUrl,\n\t\t\tdangerouslyAllowBrowser: true,\n\t\t\tdefaultHeaders: mergeHeaders(\n\t\t\t\t{\n\t\t\t\t\taccept: \"application/json\",\n\t\t\t\t\t\"anthropic-dangerous-direct-browser-access\": \"true\",\n\t\t\t\t\t\"anthropic-beta\": [\"claude-code-20250219\", \"oauth-2025-04-20\", ...betaFeatures].join(\",\"),\n\t\t\t\t\t\"user-agent\": `claude-cli/${claudeCodeVersion}`,\n\t\t\t\t\t\"x-app\": \"cli\",\n\t\t\t\t},\n\t\t\t\tmodel.headers,\n\t\t\t\toptionsHeaders,\n\t\t\t),\n\t\t});\n\n\t\treturn { client, isOAuthToken: true };\n\t}\n\n\t// API key auth\n\tconst sessionAffinityHeaders: Record<string, string | null> =\n\t\tsessionId && getAnthropicCompat(model).sendSessionAffinityHeaders ? { \"x-session-affinity\": sessionId } : {};\n\tconst client = new Anthropic({\n\t\tapiKey,\n\t\tbaseURL: model.baseUrl,\n\t\tfetch: fetchImplementation,\n\t\tdangerouslyAllowBrowser: true,\n\t\tdefaultHeaders: mergeHeaders(\n\t\t\t{\n\t\t\t\taccept: \"application/json\",\n\t\t\t\t\"anthropic-dangerous-direct-browser-access\": \"true\",\n\t\t\t\t...(betaFeatures.length > 0 ? { \"anthropic-beta\": betaFeatures.join(\",\") } : {}),\n\t\t\t},\n\t\t\tsessionAffinityHeaders,\n\t\t\tmodel.headers,\n\t\t\toptionsHeaders,\n\t\t),\n\t});\n\n\treturn { client, isOAuthToken: false };\n}\n\nconst claudeCodeIdentity = \"You are Claude Code, Anthropic's official CLI for Claude.\";\n\n/**\n * Builds the system block array for a sectioned system prompt.\n *\n * Anthropic allows at most 4 `cache_control` breakpoints per request, and the\n * last tool and last user message already consume two, so the system prompt\n * spends at most one: it lands on a single stable block folding the leading\n * run of sections up to the first `cacheRetention: \"none\"` section (zero\n * breakpoints when retention is \"none\"; when the first section is volatile\n * the breakpoint moves to the OAuth identity block, or is omitted entirely\n * on API-key auth).\n * Everything from the first volatile section onward trails as uncached blocks\n * in array order — anything after the breakpoint is excluded from the cached\n * prefix, and preserving array order keeps the concatenated block text\n * byte-identical (for well-formed text; `sanitizeSurrogates` runs per block)\n * to the flattened prompt other providers send. On OAuth the\n * Claude Code identity block stays first; caching is prefix-based, so the\n * stable block's breakpoint covers it without a second one (unlike the legacy\n * string path, which marks both for back-compat).\n */\nfunction buildSystemBlocks(\n\tsections: SystemPromptSection[],\n\tisOAuthToken: boolean,\n\tcacheControl?: CacheControlEphemeral,\n): TextBlockParam[] {\n\tconst blocks: TextBlockParam[] = [];\n\tconst nonEmpty = sections.filter((s) => s.text);\n\tconst firstVolatile = nonEmpty.findIndex((s) => s.cacheRetention === \"none\");\n\tconst stable = firstVolatile === -1 ? nonEmpty : nonEmpty.slice(0, firstVolatile);\n\tconst tail = firstVolatile === -1 ? [] : nonEmpty.slice(firstVolatile);\n\n\tif (isOAuthToken) {\n\t\tblocks.push({\n\t\t\ttype: \"text\",\n\t\t\ttext: claudeCodeIdentity,\n\t\t\t// Without stable sections the identity block is the whole stable\n\t\t\t// prefix, so the breakpoint lands here (as on the string path).\n\t\t\t...(stable.length === 0 && cacheControl ? { cache_control: cacheControl } : {}),\n\t\t});\n\t}\n\tif (stable.length > 0) {\n\t\tblocks.push({\n\t\t\ttype: \"text\",\n\t\t\ttext: sanitizeSurrogates(stable.map((s) => s.text).join(\"\")),\n\t\t\t...(cacheControl ? { cache_control: cacheControl } : {}),\n\t\t});\n\t}\n\tfor (const section of tail) {\n\t\tblocks.push({ type: \"text\", text: sanitizeSurrogates(section.text) });\n\t}\n\treturn blocks;\n}\n\nfunction buildParams(\n\tmodel: Model<\"anthropic-messages\">,\n\tcontext: Context,\n\tisOAuthToken: boolean,\n\toptions?: AnthropicOptions,\n): MessageCreateParamsStreaming {\n\tconst { cacheControl } = getCacheControl(model, options?.cacheRetention);\n\tconst params: MessageCreateParamsStreaming = {\n\t\tmodel: model.id,\n\t\tmessages: convertMessages(context.messages, model, isOAuthToken, cacheControl),\n\t\tmax_tokens: options?.maxTokens || (model.maxTokens / 3) | 0,\n\t\tstream: true,\n\t};\n\n\tif (Array.isArray(context.systemPrompt)) {\n\t\tconst blocks = buildSystemBlocks(context.systemPrompt, isOAuthToken, cacheControl);\n\t\tif (blocks.length > 0) {\n\t\t\tparams.system = blocks;\n\t\t}\n\t} else {\n\t\tconst systemPrompt = context.systemPrompt;\n\n\t\t// For OAuth tokens, we MUST include Claude Code identity\n\t\tif (isOAuthToken) {\n\t\t\tparams.system = [\n\t\t\t\t{\n\t\t\t\t\ttype: \"text\",\n\t\t\t\t\ttext: claudeCodeIdentity,\n\t\t\t\t\t...(cacheControl ? { cache_control: cacheControl } : {}),\n\t\t\t\t},\n\t\t\t];\n\t\t\tif (systemPrompt) {\n\t\t\t\tparams.system.push({\n\t\t\t\t\ttype: \"text\",\n\t\t\t\t\ttext: sanitizeSurrogates(systemPrompt),\n\t\t\t\t\t...(cacheControl ? { cache_control: cacheControl } : {}),\n\t\t\t\t});\n\t\t\t}\n\t\t} else if (systemPrompt) {\n\t\t\t// Add cache control to system prompt for non-OAuth tokens\n\t\t\tparams.system = [\n\t\t\t\t{\n\t\t\t\t\ttype: \"text\",\n\t\t\t\t\ttext: sanitizeSurrogates(systemPrompt),\n\t\t\t\t\t...(cacheControl ? { cache_control: cacheControl } : {}),\n\t\t\t\t},\n\t\t\t];\n\t\t}\n\t}\n\n\t// Temperature is incompatible with extended thinking (adaptive or budget-based).\n\t// SCRAMJET-DIVERGENCE: gate on supportsTemperature compat (opus-4-7+ rejects non-default temperature)\n\tif (\n\t\toptions?.temperature !== undefined &&\n\t\t!options?.thinkingEnabled &&\n\t\tgetAnthropicCompat(model).supportsTemperature\n\t) {\n\t\tparams.temperature = options.temperature;\n\t}\n\n\tif (context.tools && context.tools.length > 0) {\n\t\tconst compat = getAnthropicCompat(model);\n\t\tparams.tools = convertTools(\n\t\t\tcontext.tools,\n\t\t\tisOAuthToken,\n\t\t\tcompat.supportsEagerToolInputStreaming,\n\t\t\tcompat.supportsCacheControlOnTools ? cacheControl : undefined,\n\t\t);\n\t}\n\n\t// Configure thinking mode: adaptive-thinking models,\n\t// budget-based older models, or explicitly disabled.\n\tif (model.reasoning) {\n\t\tif (options?.thinkingEnabled) {\n\t\t\t// Default to \"summarized\" so Opus 4.7 and Mythos Preview behave like\n\t\t\t// older Claude 4 models (whose API default is also \"summarized\").\n\t\t\tconst display: AnthropicThinkingDisplay = options.thinkingDisplay ?? \"summarized\";\n\t\t\tif (supportsAdaptiveThinking(model.id)) {\n\t\t\t\t// Adaptive thinking: Claude decides when and how much to think.\n\t\t\t\tparams.thinking = { type: \"adaptive\", display };\n\t\t\t\tif (options.effort) {\n\t\t\t\t\t// The Anthropic SDK types can lag newly supported effort values such as \"xhigh\".\n\t\t\t\t\tparams.output_config =\n\t\t\t\t\t\toptions.effort === \"xhigh\"\n\t\t\t\t\t\t\t? ({ effort: options.effort } as unknown as NonNullable<\n\t\t\t\t\t\t\t\t\tMessageCreateParamsStreaming[\"output_config\"]\n\t\t\t\t\t\t\t\t>)\n\t\t\t\t\t\t\t: { effort: options.effort };\n\t\t\t\t}\n\t\t\t} else {\n\t\t\t\t// Budget-based thinking for older models\n\t\t\t\tparams.thinking = {\n\t\t\t\t\ttype: \"enabled\",\n\t\t\t\t\tbudget_tokens: options.thinkingBudgetTokens || 1024,\n\t\t\t\t\tdisplay,\n\t\t\t\t};\n\t\t\t}\n\t\t} else if (options?.thinkingEnabled === false) {\n\t\t\tparams.thinking = { type: \"disabled\" };\n\t\t}\n\t}\n\n\tif (options?.metadata) {\n\t\tconst userId = options.metadata.user_id;\n\t\tif (typeof userId === \"string\") {\n\t\t\tparams.metadata = { user_id: userId };\n\t\t}\n\t}\n\n\tif (options?.toolChoice) {\n\t\tif (typeof options.toolChoice === \"string\") {\n\t\t\tparams.tool_choice = { type: options.toolChoice };\n\t\t} else {\n\t\t\tparams.tool_choice = options.toolChoice;\n\t\t}\n\t}\n\n\treturn params;\n}\n\n// Normalize tool call IDs to match Anthropic's required pattern and length\nfunction normalizeToolCallId(id: string): string {\n\treturn id.replace(/[^a-zA-Z0-9_-]/g, \"_\").slice(0, 64);\n}\n\nfunction convertMessages(\n\tmessages: Message[],\n\tmodel: Model<\"anthropic-messages\">,\n\tisOAuthToken: boolean,\n\tcacheControl?: CacheControlEphemeral,\n): MessageParam[] {\n\tconst params: MessageParam[] = [];\n\n\t// Transform messages for cross-provider compatibility\n\tconst transformedMessages = transformMessages(messages, model, normalizeToolCallId);\n\n\tfor (let i = 0; i < transformedMessages.length; i++) {\n\t\tconst msg = transformedMessages[i];\n\n\t\tif (msg.role === \"user\") {\n\t\t\tif (typeof msg.content === \"string\") {\n\t\t\t\tif (msg.content.trim().length > 0) {\n\t\t\t\t\tparams.push({\n\t\t\t\t\t\trole: \"user\",\n\t\t\t\t\t\tcontent: sanitizeSurrogates(msg.content),\n\t\t\t\t\t});\n\t\t\t\t}\n\t\t\t} else {\n\t\t\t\tconst blocks: ContentBlockParam[] = msg.content.map((item) => {\n\t\t\t\t\tif (item.type === \"text\") {\n\t\t\t\t\t\treturn {\n\t\t\t\t\t\t\ttype: \"text\",\n\t\t\t\t\t\t\ttext: sanitizeSurrogates(item.text),\n\t\t\t\t\t\t};\n\t\t\t\t\t} else {\n\t\t\t\t\t\treturn {\n\t\t\t\t\t\t\ttype: \"image\",\n\t\t\t\t\t\t\tsource: {\n\t\t\t\t\t\t\t\ttype: \"base64\",\n\t\t\t\t\t\t\t\tmedia_type: item.mimeType as \"image/jpeg\" | \"image/png\" | \"image/gif\" | \"image/webp\",\n\t\t\t\t\t\t\t\tdata: item.data,\n\t\t\t\t\t\t\t},\n\t\t\t\t\t\t};\n\t\t\t\t\t}\n\t\t\t\t});\n\t\t\t\tconst filteredBlocks = blocks.filter((b) => {\n\t\t\t\t\tif (b.type === \"text\") {\n\t\t\t\t\t\treturn b.text.trim().length > 0;\n\t\t\t\t\t}\n\t\t\t\t\treturn true;\n\t\t\t\t});\n\t\t\t\tif (filteredBlocks.length === 0) continue;\n\t\t\t\tparams.push({\n\t\t\t\t\trole: \"user\",\n\t\t\t\t\tcontent: filteredBlocks,\n\t\t\t\t});\n\t\t\t}\n\t\t} else if (msg.role === \"assistant\") {\n\t\t\tconst blocks: ContentBlockParam[] = [];\n\n\t\t\tfor (const block of msg.content) {\n\t\t\t\tif (block.type === \"text\") {\n\t\t\t\t\tif (block.text.trim().length === 0) continue;\n\t\t\t\t\tblocks.push({\n\t\t\t\t\t\ttype: \"text\",\n\t\t\t\t\t\ttext: sanitizeSurrogates(block.text),\n\t\t\t\t\t});\n\t\t\t\t} else if (block.type === \"thinking\") {\n\t\t\t\t\t// Redacted thinking: pass the opaque payload back as redacted_thinking\n\t\t\t\t\tif (block.redacted) {\n\t\t\t\t\t\tblocks.push({\n\t\t\t\t\t\t\ttype: \"redacted_thinking\",\n\t\t\t\t\t\t\tdata: block.thinkingSignature!,\n\t\t\t\t\t\t});\n\t\t\t\t\t\tcontinue;\n\t\t\t\t\t}\n\t\t\t\t\tif (block.thinking.trim().length === 0) continue;\n\t\t\t\t\t// If thinking signature is missing/empty (e.g., from aborted stream),\n\t\t\t\t\t// convert to plain text block without <thinking> tags to avoid API rejection\n\t\t\t\t\t// and prevent Claude from mimicking the tags in responses\n\t\t\t\t\tif (!block.thinkingSignature || block.thinkingSignature.trim().length === 0) {\n\t\t\t\t\t\tblocks.push({\n\t\t\t\t\t\t\ttype: \"text\",\n\t\t\t\t\t\t\ttext: sanitizeSurrogates(block.thinking),\n\t\t\t\t\t\t});\n\t\t\t\t\t} else {\n\t\t\t\t\t\tblocks.push({\n\t\t\t\t\t\t\ttype: \"thinking\",\n\t\t\t\t\t\t\tthinking: sanitizeSurrogates(block.thinking),\n\t\t\t\t\t\t\tsignature: block.thinkingSignature,\n\t\t\t\t\t\t});\n\t\t\t\t\t}\n\t\t\t\t} else if (block.type === \"toolCall\") {\n\t\t\t\t\tblocks.push({\n\t\t\t\t\t\ttype: \"tool_use\",\n\t\t\t\t\t\t// SCRAMJET-DIVERGENCE: Sanitize tool-call IDs unconditionally on the outgoing\n\t\t\t\t\t\t// request. `transform-messages.ts` only normalizes IDs for cross-model replay\n\t\t\t\t\t\t// (`!isSameModel`), so a legacy session carrying a provider-invalid ID (e.g. one\n\t\t\t\t\t\t// embedding a raw model id like `gpt-5.5`) replayed to the SAME model would be sent\n\t\t\t\t\t\t// verbatim and rejected by Anthropic's `^[a-zA-Z0-9_-]+$` constraint. Applying the\n\t\t\t\t\t\t// idempotent `normalizeToolCallId` here (and at the matching tool_result sites below)\n\t\t\t\t\t\t// is harmless on already-clean IDs and preserves call/result correlation because the\n\t\t\t\t\t\t// same deterministic function maps both halves to the same value.\n\t\t\t\t\t\tid: normalizeToolCallId(block.id),\n\t\t\t\t\t\tname: isOAuthToken ? toClaudeCodeName(block.name) : block.name,\n\t\t\t\t\t\tinput: block.arguments ?? {},\n\t\t\t\t\t});\n\t\t\t\t}\n\t\t\t}\n\t\t\tif (blocks.length === 0) continue;\n\t\t\tparams.push({\n\t\t\t\trole: \"assistant\",\n\t\t\t\tcontent: blocks,\n\t\t\t});\n\t\t} else if (msg.role === \"toolResult\") {\n\t\t\t// Collect all consecutive toolResult messages, needed for z.ai Anthropic endpoint\n\t\t\tconst toolResults: ContentBlockParam[] = [];\n\n\t\t\t// Add the current tool result\n\t\t\ttoolResults.push({\n\t\t\t\ttype: \"tool_result\",\n\t\t\t\t// SCRAMJET-DIVERGENCE: sanitize to keep correlation with the tool_use id above.\n\t\t\t\ttool_use_id: normalizeToolCallId(msg.toolCallId),\n\t\t\t\tcontent: convertContentBlocks(msg.content),\n\t\t\t\tis_error: msg.isError,\n\t\t\t});\n\n\t\t\t// Look ahead for consecutive toolResult messages\n\t\t\tlet j = i + 1;\n\t\t\twhile (j < transformedMessages.length && transformedMessages[j].role === \"toolResult\") {\n\t\t\t\tconst nextMsg = transformedMessages[j] as ToolResultMessage; // We know it's a toolResult\n\t\t\t\ttoolResults.push({\n\t\t\t\t\ttype: \"tool_result\",\n\t\t\t\t\t// SCRAMJET-DIVERGENCE: sanitize to keep correlation with the tool_use id above.\n\t\t\t\t\ttool_use_id: normalizeToolCallId(nextMsg.toolCallId),\n\t\t\t\t\tcontent: convertContentBlocks(nextMsg.content),\n\t\t\t\t\tis_error: nextMsg.isError,\n\t\t\t\t});\n\t\t\t\tj++;\n\t\t\t}\n\n\t\t\t// Skip the messages we've already processed\n\t\t\ti = j - 1;\n\n\t\t\t// Add a single user message with all tool results\n\t\t\tparams.push({\n\t\t\t\trole: \"user\",\n\t\t\t\tcontent: toolResults,\n\t\t\t});\n\t\t}\n\t}\n\n\t// Add cache_control to the last user message to cache conversation history\n\tif (cacheControl && params.length > 0) {\n\t\tconst lastMessage = params[params.length - 1];\n\t\tif (lastMessage.role === \"user\") {\n\t\t\tif (Array.isArray(lastMessage.content)) {\n\t\t\t\tconst lastBlock = lastMessage.content[lastMessage.content.length - 1];\n\t\t\t\tif (\n\t\t\t\t\tlastBlock &&\n\t\t\t\t\t(lastBlock.type === \"text\" || lastBlock.type === \"image\" || lastBlock.type === \"tool_result\")\n\t\t\t\t) {\n\t\t\t\t\t(lastBlock as any).cache_control = cacheControl;\n\t\t\t\t}\n\t\t\t} else if (typeof lastMessage.content === \"string\") {\n\t\t\t\tlastMessage.content = [\n\t\t\t\t\t{\n\t\t\t\t\t\ttype: \"text\",\n\t\t\t\t\t\ttext: lastMessage.content,\n\t\t\t\t\t\tcache_control: cacheControl,\n\t\t\t\t\t},\n\t\t\t\t] as any;\n\t\t\t}\n\t\t}\n\t}\n\n\treturn params;\n}\n\nfunction shouldUseFineGrainedToolStreamingBeta(model: Model<\"anthropic-messages\">, context: Context): boolean {\n\treturn !!context.tools?.length && !getAnthropicCompat(model).supportsEagerToolInputStreaming;\n}\n\nfunction convertTools(\n\ttools: Tool[],\n\tisOAuthToken: boolean,\n\tsupportsEagerToolInputStreaming: boolean,\n\tcacheControl?: CacheControlEphemeral,\n): Anthropic.Messages.Tool[] {\n\tif (!tools) return [];\n\n\treturn tools.map((tool, index) => {\n\t\tconst schema = tool.parameters as { properties?: unknown; required?: string[] };\n\n\t\treturn {\n\t\t\tname: isOAuthToken ? toClaudeCodeName(tool.name) : tool.name,\n\t\t\tdescription: tool.description,\n\t\t\t...(supportsEagerToolInputStreaming ? { eager_input_streaming: true } : {}),\n\t\t\tinput_schema: {\n\t\t\t\ttype: \"object\",\n\t\t\t\tproperties: schema.properties ?? {},\n\t\t\t\trequired: schema.required ?? [],\n\t\t\t},\n\t\t\t...(cacheControl && index === tools.length - 1 ? { cache_control: cacheControl } : {}),\n\t\t};\n\t});\n}\n\nfunction mapStopReason(reason: Anthropic.Messages.StopReason | string): StopReason {\n\tswitch (reason) {\n\t\tcase \"end_turn\":\n\t\t\treturn \"stop\";\n\t\tcase \"max_tokens\":\n\t\t\treturn \"length\";\n\t\tcase \"tool_use\":\n\t\t\treturn \"toolUse\";\n\t\tcase \"refusal\":\n\t\t\treturn \"error\";\n\t\tcase \"pause_turn\": // Stop is good enough -> resubmit\n\t\t\treturn \"stop\";\n\t\tcase \"stop_sequence\":\n\t\t\treturn \"stop\"; // We don't supply stop sequences, so this should never happen\n\t\tcase \"sensitive\": // Content flagged by safety filters (not yet in SDK types)\n\t\t\treturn \"error\";\n\t\tdefault:\n\t\t\t// Handle unknown stop reasons gracefully (API may add new values)\n\t\t\tthrow new Error(`Unhandled stop reason: ${reason}`);\n\t}\n}\n"]}

package/dist/providers/anthropic.js

@@ -5,6 +5,7 @@ import { AssistantMessageEventStream } from "../utils/event-stream.js";
 import { appendBuiltinFailure, failureFromProviderError, invokeProviderCallback, RequestFailureError, } from "../utils/failure-evidence.js";
 import { headersToRecord } from "../utils/headers.js";
 import { parseJsonWithRepair, parseStreamingJson } from "../utils/json-parse.js";
+import { createProviderAbortScope } from "../utils/provider-abort-scope.js";
 import { sanitizeSurrogates } from "../utils/sanitize-unicode.js";
 import { ADAPTIVE_THINKING_PATTERNS, NATIVE_XHIGH_EFFORT_PATTERNS, normalizeForPatternMatch, } from "./anthropic-model-patterns.js";
 import { resolveCloudflareBaseUrl } from "./cloudflare.js";
@@ -343,6 +344,7 @@ export const streamAnthropic = (model, context, options) => {
             timestamp: Date.now(),
         };
         let requestStarted = false;
+        const abortScope = createProviderAbortScope(options?.signal);
         try {
             let client;
             let isOAuth;
@@ -375,7 +377,7 @@ export const streamAnthropic = (model, context, options) => {
                 params = nextParams;
             }
             const requestOptions = {
-                ...(options?.signal ? { signal: options.signal } : {}),
+                signal: abortScope.signal,
                 ...(options?.timeoutMs !== undefined ? { timeout: options.timeoutMs } : {}),
                 ...(options?.maxRetries !== undefined ? { maxRetries: options.maxRetries } : {}),
             };
@@ -385,7 +387,7 @@ export const streamAnthropic = (model, context, options) => {
             await invokeProviderCallback("onResponse", () => options?.onResponse?.({ status: response.status, headers: headersToRecord(response.headers) }, model));
             stream.push({ type: "start", partial: output });
             const blocks = output.content;
-            for await (const event of iterateAnthropicEvents(response, options?.signal)) {
+            for await (const event of iterateAnthropicEvents(response, abortScope.signal)) {
                 if (event.type === "message_start") {
                     output.responseId = event.message.id;
                     // Capture initial token usage from message_start event
@@ -587,6 +589,10 @@ export const streamAnthropic = (model, context, options) => {
             stream.push({ type: "error", reason: output.stopReason, error: output });
             stream.end();
         }
+        finally {
+            // SCRAMJET-DIVERGENCE: finish cleanup synchronously before terminal consumers resume (#587).
+            abortScope.dispose();
+        }
     })();
     return stream;
 };