@jecruz/merc-ai 0.1.3 → 0.1.6

package/dist/models.js

@@ -32,7 +32,10 @@ export function calculateCost(model, usage) {
     usage.cost.input = (cost.input / 1000000) * usage.input;
     usage.cost.output = (cost.output / 1000000) * usage.output;
     usage.cost.cacheRead = (cost.cacheRead / 1000000) * usage.cacheRead;
-    usage.cost.cacheWrite = (cost.cacheWrite / 1000000) * usage.cacheWrite;
+    // Anthropic charges 2x base input for 1h cache writes.
+    const longWrite = usage.cacheWrite1h ?? 0;
+    const shortWrite = usage.cacheWrite - longWrite;
+    usage.cost.cacheWrite = (cost.cacheWrite * shortWrite + cost.input * 2 * longWrite) / 1000000;
     usage.cost.total = usage.cost.input + usage.cost.output + usage.cost.cacheRead + usage.cost.cacheWrite;
     return usage.cost;
 }

package/dist/models.js.map

@@ -1 +1 @@
-{"version":3,"file":"models.js","sourceRoot":"","sources":["../src/models.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,MAAM,EAAE,MAAM,uBAAuB,CAAC;AAG/C,MAAM,aAAa,GAAyC,IAAI,GAAG,EAAE,CAAC;AAEtE,iDAAiD;AACjD,KAAK,MAAM,CAAC,QAAQ,EAAE,MAAM,CAAC,IAAI,MAAM,CAAC,OAAO,CAAC,MAAM,CAAC,EAAE,CAAC;IACzD,MAAM,cAAc,GAAG,IAAI,GAAG,EAAsB,CAAC;IACrD,KAAK,MAAM,CAAC,EAAE,EAAE,KAAK,CAAC,IAAI,MAAM,CAAC,OAAO,CAAC,MAAM,CAAC,EAAE,CAAC;QAClD,cAAc,CAAC,GAAG,CAAC,EAAE,EAAE,KAAmB,CAAC,CAAC;IAC7C,CAAC;IACD,aAAa,CAAC,GAAG,CAAC,QAAQ,EAAE,cAAc,CAAC,CAAC;AAC7C,CAAC;AAwBD,MAAM,UAAU,QAAQ,CACvB,QAAmB,EACnB,OAAiB,EACsB;IACvC,MAAM,cAAc,GAAG,aAAa,CAAC,GAAG,CAAC,QAAQ,CAAC,CAAC;IACnD,OAAO,cAAc,EAAE,GAAG,CAAC,OAAiB,CAAyC,CAAC;AAAA,CACtF;AAED,MAAM,UAAU,YAAY,GAAoB;IAC/C,OAAO,KAAK,CAAC,IAAI,CAAC,aAAa,CAAC,IAAI,EAAE,CAAoB,CAAC;AAAA,CAC3D;AAED,MAAM,UAAU,SAAS,CACxB,QAAmB,EAC8C;IACjE,MAAM,MAAM,GAAG,aAAa,CAAC,GAAG,CAAC,QAAQ,CAAC,CAAC;IAC3C,OAAO,MAAM,CAAC,CAAC,CAAE,KAAK,CAAC,IAAI,CAAC,MAAM,CAAC,MAAM,EAAE,CAAoE,CAAC,CAAC,CAAC,EAAE,CAAC;AAAA,CACrH;AAED,MAAM,UAAU,aAAa,CAAmB,KAAkB,EAAE,KAAY,EAAiB;IAChG,MAAM,WAAW,GAAG,KAAK,CAAC,KAAK,GAAG,KAAK,CAAC,SAAS,GAAG,KAAK,CAAC,UAAU,CAAC;IACrE,IAAI,IAAI,GAAwB,KAAK,CAAC,IAAI,CAAC;IAC3C,IAAI,gBAAgB,GAAG,CAAC,CAAC,CAAC;IAC1B,KAAK,MAAM,IAAI,IAAI,KAAK,CAAC,IAAI,CAAC,KAAK,IAAI,EAAE,EAAE,CAAC;QAC3C,IAAI,WAAW,GAAG,IAAI,CAAC,gBAAgB,IAAI,IAAI,CAAC,gBAAgB,GAAG,gBAAgB,EAAE,CAAC;YACrF,IAAI,GAAG,IAAI,CAAC;YACZ,gBAAgB,GAAG,IAAI,CAAC,gBAAgB,CAAC;QAC1C,CAAC;IACF,CAAC;IAED,KAAK,CAAC,IAAI,CAAC,KAAK,GAAG,CAAC,IAAI,CAAC,KAAK,GAAG,OAAO,CAAC,GAAG,KAAK,CAAC,KAAK,CAAC;IACxD,KAAK,CAAC,IAAI,CAAC,MAAM,GAAG,CAAC,IAAI,CAAC,MAAM,GAAG,OAAO,CAAC,GAAG,KAAK,CAAC,MAAM,CAAC;IAC3D,KAAK,CAAC,IAAI,CAAC,SAAS,GAAG,CAAC,IAAI,CAAC,SAAS,GAAG,OAAO,CAAC,GAAG,KAAK,CAAC,SAAS,CAAC;IACpE,KAAK,CAAC,IAAI,CAAC,UAAU,GAAG,CAAC,IAAI,CAAC,UAAU,GAAG,OAAO,CAAC,GAAG,KAAK,CAAC,UAAU,CAAC;IACvE,KAAK,CAAC,IAAI,CAAC,KAAK,GAAG,KAAK,CAAC,IAAI,CAAC,KAAK,GAAG,KAAK,CAAC,IAAI,CAAC,MAAM,GAAG,KAAK,CAAC,IAAI,CAAC,SAAS,GAAG,KAAK,CAAC,IAAI,CAAC,UAAU,CAAC;IACvG,OAAO,KAAK,CAAC,IAAI,CAAC;AAAA,CAClB;AAED,MAAM,wBAAwB,GAAyB,CAAC,KAAK,EAAE,SAAS,EAAE,KAAK,EAAE,QAAQ,EAAE,MAAM,EAAE,OAAO,EAAE,KAAK,CAAC,CAAC;AAEnH,MAAM,UAAU,0BAA0B,CAAmB,KAAkB,EAAwB;IACtG,IAAI,CAAC,KAAK,CAAC,SAAS;QAAE,OAAO,CAAC,KAAK,CAAC,CAAC;IAErC,OAAO,wBAAwB,CAAC,MAAM,CAAC,CAAC,KAAK,EAAE,EAAE,CAAC;QACjD,MAAM,MAAM,GAAG,KAAK,CAAC,gBAAgB,EAAE,CAAC,KAAK,CAAC,CAAC;QAC/C,IAAI,MAAM,KAAK,IAAI;YAAE,OAAO,KAAK,CAAC;QAClC,IAAI,KAAK,KAAK,OAAO,IAAI,KAAK,KAAK,KAAK;YAAE,OAAO,MAAM,KAAK,SAAS,CAAC;QACtE,OAAO,IAAI,CAAC;IAAA,CACZ,CAAC,CAAC;AAAA,CACH;AAED,MAAM,UAAU,kBAAkB,CACjC,KAAkB,EAClB,KAAyB,EACJ;IACrB,MAAM,eAAe,GAAG,0BAA0B,CAAC,KAAK,CAAC,CAAC;IAC1D,IAAI,eAAe,CAAC,QAAQ,CAAC,KAAK,CAAC;QAAE,OAAO,KAAK,CAAC;IAElD,MAAM,cAAc,GAAG,wBAAwB,CAAC,OAAO,CAAC,KAAK,CAAC,CAAC;IAC/D,IAAI,cAAc,KAAK,CAAC,CAAC;QAAE,OAAO,eAAe,CAAC,CAAC,CAAC,IAAI,KAAK,CAAC;IAE9D,KAAK,IAAI,CAAC,GAAG,cAAc,EAAE,CAAC,GAAG,wBAAwB,CAAC,MAAM,EAAE,CAAC,EAAE,EAAE,CAAC;QACvE,MAAM,SAAS,GAAG,wBAAwB,CAAC,CAAC,CAAC,CAAC;QAC9C,IAAI,eAAe,CAAC,QAAQ,CAAC,SAAS,CAAC;YAAE,OAAO,SAAS,CAAC;IAC3D,CAAC;IACD,KAAK,IAAI,CAAC,GAAG,cAAc,GAAG,CAAC,EAAE,CAAC,IAAI,CAAC,EAAE,CAAC,EAAE,EAAE,CAAC;QAC9C,MAAM,SAAS,GAAG,wBAAwB,CAAC,CAAC,CAAC,CAAC;QAC9C,IAAI,eAAe,CAAC,QAAQ,CAAC,SAAS,CAAC;YAAE,OAAO,SAAS,CAAC;IAC3D,CAAC;IACD,OAAO,eAAe,CAAC,CAAC,CAAC,IAAI,KAAK,CAAC;AAAA,CACnC;AAED;;;GAGG;AACH,MAAM,UAAU,cAAc,CAC7B,CAAiC,EACjC,CAAiC,EACvB;IACV,IAAI,CAAC,CAAC,IAAI,CAAC,CAAC;QAAE,OAAO,KAAK,CAAC;IAC3B,OAAO,CAAC,CAAC,EAAE,KAAK,CAAC,CAAC,EAAE,IAAI,CAAC,CAAC,QAAQ,KAAK,CAAC,CAAC,QAAQ,CAAC;AAAA,CAClD","sourcesContent":["import { MODELS } from \"./models.generated.js\";\nimport type { Api, KnownProvider, Model, ModelThinkingLevel, Usage } from \"./types.js\";\n\nconst modelRegistry: Map<string, Map<string, Model<Api>>> = new Map();\n\n// Initialize registry from MODELS on module load\nfor (const [provider, models] of Object.entries(MODELS)) {\n\tconst providerModels = new Map<string, Model<Api>>();\n\tfor (const [id, model] of Object.entries(models)) {\n\t\tproviderModels.set(id, model as Model<Api>);\n\t}\n\tmodelRegistry.set(provider, providerModels);\n}\n\ntype ModelApi<\n\tTProvider extends KnownProvider,\n\tTModelId extends keyof (typeof MODELS)[TProvider],\n> = (typeof MODELS)[TProvider][TModelId] extends { api: infer TApi } ? (TApi extends Api ? TApi : never) : never;\n\ntype ProviderModelApi<TProvider extends KnownProvider> =\n\t(typeof MODELS)[TProvider][keyof (typeof MODELS)[TProvider]] extends {\n\t\tapi: infer TApi;\n\t}\n\t\t? TApi extends Api\n\t\t\t? TApi\n\t\t\t: never\n\t\t: never;\n\nexport function getModel<TProvider extends KnownProvider, TModelId extends keyof (typeof MODELS)[TProvider]>(\n\tprovider: TProvider,\n\tmodelId: TModelId,\n): Model<ModelApi<TProvider, TModelId>>;\nexport function getModel<TProvider extends KnownProvider>(\n\tprovider: TProvider,\n\tmodelId: string,\n): Model<ProviderModelApi<TProvider>> | undefined;\nexport function getModel<TProvider extends KnownProvider, TModelId extends keyof (typeof MODELS)[TProvider]>(\n\tprovider: TProvider,\n\tmodelId: TModelId,\n): Model<ModelApi<TProvider, TModelId>> {\n\tconst providerModels = modelRegistry.get(provider);\n\treturn providerModels?.get(modelId as string) as Model<ModelApi<TProvider, TModelId>>;\n}\n\nexport function getProviders(): KnownProvider[] {\n\treturn Array.from(modelRegistry.keys()) as KnownProvider[];\n}\n\nexport function getModels<TProvider extends KnownProvider>(\n\tprovider: TProvider,\n): Model<ModelApi<TProvider, keyof (typeof MODELS)[TProvider]>>[] {\n\tconst models = modelRegistry.get(provider);\n\treturn models ? (Array.from(models.values()) as Model<ModelApi<TProvider, keyof (typeof MODELS)[TProvider]>>[]) : [];\n}\n\nexport function calculateCost<TApi extends Api>(model: Model<TApi>, usage: Usage): Usage[\"cost\"] {\n\tconst inputTokens = usage.input + usage.cacheRead + usage.cacheWrite;\n\tlet cost: Model<TApi>[\"cost\"] = model.cost;\n\tlet matchedThreshold = -1;\n\tfor (const tier of model.cost.tiers ?? []) {\n\t\tif (inputTokens > tier.inputTokensAbove && tier.inputTokensAbove > matchedThreshold) {\n\t\t\tcost = tier;\n\t\t\tmatchedThreshold = tier.inputTokensAbove;\n\t\t}\n\t}\n\n\tusage.cost.input = (cost.input / 1000000) * usage.input;\n\tusage.cost.output = (cost.output / 1000000) * usage.output;\n\tusage.cost.cacheRead = (cost.cacheRead / 1000000) * usage.cacheRead;\n\tusage.cost.cacheWrite = (cost.cacheWrite / 1000000) * usage.cacheWrite;\n\tusage.cost.total = usage.cost.input + usage.cost.output + usage.cost.cacheRead + usage.cost.cacheWrite;\n\treturn usage.cost;\n}\n\nconst EXTENDED_THINKING_LEVELS: ModelThinkingLevel[] = [\"off\", \"minimal\", \"low\", \"medium\", \"high\", \"xhigh\", \"max\"];\n\nexport function getSupportedThinkingLevels<TApi extends Api>(model: Model<TApi>): ModelThinkingLevel[] {\n\tif (!model.reasoning) return [\"off\"];\n\n\treturn EXTENDED_THINKING_LEVELS.filter((level) => {\n\t\tconst mapped = model.thinkingLevelMap?.[level];\n\t\tif (mapped === null) return false;\n\t\tif (level === \"xhigh\" || level === \"max\") return mapped !== undefined;\n\t\treturn true;\n\t});\n}\n\nexport function clampThinkingLevel<TApi extends Api>(\n\tmodel: Model<TApi>,\n\tlevel: ModelThinkingLevel,\n): ModelThinkingLevel {\n\tconst availableLevels = getSupportedThinkingLevels(model);\n\tif (availableLevels.includes(level)) return level;\n\n\tconst requestedIndex = EXTENDED_THINKING_LEVELS.indexOf(level);\n\tif (requestedIndex === -1) return availableLevels[0] ?? \"off\";\n\n\tfor (let i = requestedIndex; i < EXTENDED_THINKING_LEVELS.length; i++) {\n\t\tconst candidate = EXTENDED_THINKING_LEVELS[i];\n\t\tif (availableLevels.includes(candidate)) return candidate;\n\t}\n\tfor (let i = requestedIndex - 1; i >= 0; i--) {\n\t\tconst candidate = EXTENDED_THINKING_LEVELS[i];\n\t\tif (availableLevels.includes(candidate)) return candidate;\n\t}\n\treturn availableLevels[0] ?? \"off\";\n}\n\n/**\n * Check if two models are equal by comparing both their id and provider.\n * Returns false if either model is null or undefined.\n */\nexport function modelsAreEqual<TApi extends Api>(\n\ta: Model<TApi> | null | undefined,\n\tb: Model<TApi> | null | undefined,\n): boolean {\n\tif (!a || !b) return false;\n\treturn a.id === b.id && a.provider === b.provider;\n}\n"]}
+{"version":3,"file":"models.js","sourceRoot":"","sources":["../src/models.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,MAAM,EAAE,MAAM,uBAAuB,CAAC;AAG/C,MAAM,aAAa,GAAyC,IAAI,GAAG,EAAE,CAAC;AAEtE,iDAAiD;AACjD,KAAK,MAAM,CAAC,QAAQ,EAAE,MAAM,CAAC,IAAI,MAAM,CAAC,OAAO,CAAC,MAAM,CAAC,EAAE,CAAC;IACzD,MAAM,cAAc,GAAG,IAAI,GAAG,EAAsB,CAAC;IACrD,KAAK,MAAM,CAAC,EAAE,EAAE,KAAK,CAAC,IAAI,MAAM,CAAC,OAAO,CAAC,MAAM,CAAC,EAAE,CAAC;QAClD,cAAc,CAAC,GAAG,CAAC,EAAE,EAAE,KAAmB,CAAC,CAAC;IAC7C,CAAC;IACD,aAAa,CAAC,GAAG,CAAC,QAAQ,EAAE,cAAc,CAAC,CAAC;AAC7C,CAAC;AAwBD,MAAM,UAAU,QAAQ,CACvB,QAAmB,EACnB,OAAiB,EACsB;IACvC,MAAM,cAAc,GAAG,aAAa,CAAC,GAAG,CAAC,QAAQ,CAAC,CAAC;IACnD,OAAO,cAAc,EAAE,GAAG,CAAC,OAAiB,CAAyC,CAAC;AAAA,CACtF;AAED,MAAM,UAAU,YAAY,GAAoB;IAC/C,OAAO,KAAK,CAAC,IAAI,CAAC,aAAa,CAAC,IAAI,EAAE,CAAoB,CAAC;AAAA,CAC3D;AAED,MAAM,UAAU,SAAS,CACxB,QAAmB,EAC8C;IACjE,MAAM,MAAM,GAAG,aAAa,CAAC,GAAG,CAAC,QAAQ,CAAC,CAAC;IAC3C,OAAO,MAAM,CAAC,CAAC,CAAE,KAAK,CAAC,IAAI,CAAC,MAAM,CAAC,MAAM,EAAE,CAAoE,CAAC,CAAC,CAAC,EAAE,CAAC;AAAA,CACrH;AAED,MAAM,UAAU,aAAa,CAAmB,KAAkB,EAAE,KAAY,EAAiB;IAChG,MAAM,WAAW,GAAG,KAAK,CAAC,KAAK,GAAG,KAAK,CAAC,SAAS,GAAG,KAAK,CAAC,UAAU,CAAC;IACrE,IAAI,IAAI,GAAwB,KAAK,CAAC,IAAI,CAAC;IAC3C,IAAI,gBAAgB,GAAG,CAAC,CAAC,CAAC;IAC1B,KAAK,MAAM,IAAI,IAAI,KAAK,CAAC,IAAI,CAAC,KAAK,IAAI,EAAE,EAAE,CAAC;QAC3C,IAAI,WAAW,GAAG,IAAI,CAAC,gBAAgB,IAAI,IAAI,CAAC,gBAAgB,GAAG,gBAAgB,EAAE,CAAC;YACrF,IAAI,GAAG,IAAI,CAAC;YACZ,gBAAgB,GAAG,IAAI,CAAC,gBAAgB,CAAC;QAC1C,CAAC;IACF,CAAC;IAED,KAAK,CAAC,IAAI,CAAC,KAAK,GAAG,CAAC,IAAI,CAAC,KAAK,GAAG,OAAO,CAAC,GAAG,KAAK,CAAC,KAAK,CAAC;IACxD,KAAK,CAAC,IAAI,CAAC,MAAM,GAAG,CAAC,IAAI,CAAC,MAAM,GAAG,OAAO,CAAC,GAAG,KAAK,CAAC,MAAM,CAAC;IAC3D,KAAK,CAAC,IAAI,CAAC,SAAS,GAAG,CAAC,IAAI,CAAC,SAAS,GAAG,OAAO,CAAC,GAAG,KAAK,CAAC,SAAS,CAAC;IACpE,uDAAuD;IACvD,MAAM,SAAS,GAAG,KAAK,CAAC,YAAY,IAAI,CAAC,CAAC;IAC1C,MAAM,UAAU,GAAG,KAAK,CAAC,UAAU,GAAG,SAAS,CAAC;IAChD,KAAK,CAAC,IAAI,CAAC,UAAU,GAAG,CAAC,IAAI,CAAC,UAAU,GAAG,UAAU,GAAG,IAAI,CAAC,KAAK,GAAG,CAAC,GAAG,SAAS,CAAC,GAAG,OAAO,CAAC;IAC9F,KAAK,CAAC,IAAI,CAAC,KAAK,GAAG,KAAK,CAAC,IAAI,CAAC,KAAK,GAAG,KAAK,CAAC,IAAI,CAAC,MAAM,GAAG,KAAK,CAAC,IAAI,CAAC,SAAS,GAAG,KAAK,CAAC,IAAI,CAAC,UAAU,CAAC;IACvG,OAAO,KAAK,CAAC,IAAI,CAAC;AAAA,CAClB;AAED,MAAM,wBAAwB,GAAyB,CAAC,KAAK,EAAE,SAAS,EAAE,KAAK,EAAE,QAAQ,EAAE,MAAM,EAAE,OAAO,EAAE,KAAK,CAAC,CAAC;AAEnH,MAAM,UAAU,0BAA0B,CAAmB,KAAkB,EAAwB;IACtG,IAAI,CAAC,KAAK,CAAC,SAAS;QAAE,OAAO,CAAC,KAAK,CAAC,CAAC;IAErC,OAAO,wBAAwB,CAAC,MAAM,CAAC,CAAC,KAAK,EAAE,EAAE,CAAC;QACjD,MAAM,MAAM,GAAG,KAAK,CAAC,gBAAgB,EAAE,CAAC,KAAK,CAAC,CAAC;QAC/C,IAAI,MAAM,KAAK,IAAI;YAAE,OAAO,KAAK,CAAC;QAClC,IAAI,KAAK,KAAK,OAAO,IAAI,KAAK,KAAK,KAAK;YAAE,OAAO,MAAM,KAAK,SAAS,CAAC;QACtE,OAAO,IAAI,CAAC;IAAA,CACZ,CAAC,CAAC;AAAA,CACH;AAED,MAAM,UAAU,kBAAkB,CACjC,KAAkB,EAClB,KAAyB,EACJ;IACrB,MAAM,eAAe,GAAG,0BAA0B,CAAC,KAAK,CAAC,CAAC;IAC1D,IAAI,eAAe,CAAC,QAAQ,CAAC,KAAK,CAAC;QAAE,OAAO,KAAK,CAAC;IAElD,MAAM,cAAc,GAAG,wBAAwB,CAAC,OAAO,CAAC,KAAK,CAAC,CAAC;IAC/D,IAAI,cAAc,KAAK,CAAC,CAAC;QAAE,OAAO,eAAe,CAAC,CAAC,CAAC,IAAI,KAAK,CAAC;IAE9D,KAAK,IAAI,CAAC,GAAG,cAAc,EAAE,CAAC,GAAG,wBAAwB,CAAC,MAAM,EAAE,CAAC,EAAE,EAAE,CAAC;QACvE,MAAM,SAAS,GAAG,wBAAwB,CAAC,CAAC,CAAC,CAAC;QAC9C,IAAI,eAAe,CAAC,QAAQ,CAAC,SAAS,CAAC;YAAE,OAAO,SAAS,CAAC;IAC3D,CAAC;IACD,KAAK,IAAI,CAAC,GAAG,cAAc,GAAG,CAAC,EAAE,CAAC,IAAI,CAAC,EAAE,CAAC,EAAE,EAAE,CAAC;QAC9C,MAAM,SAAS,GAAG,wBAAwB,CAAC,CAAC,CAAC,CAAC;QAC9C,IAAI,eAAe,CAAC,QAAQ,CAAC,SAAS,CAAC;YAAE,OAAO,SAAS,CAAC;IAC3D,CAAC;IACD,OAAO,eAAe,CAAC,CAAC,CAAC,IAAI,KAAK,CAAC;AAAA,CACnC;AAED;;;GAGG;AACH,MAAM,UAAU,cAAc,CAC7B,CAAiC,EACjC,CAAiC,EACvB;IACV,IAAI,CAAC,CAAC,IAAI,CAAC,CAAC;QAAE,OAAO,KAAK,CAAC;IAC3B,OAAO,CAAC,CAAC,EAAE,KAAK,CAAC,CAAC,EAAE,IAAI,CAAC,CAAC,QAAQ,KAAK,CAAC,CAAC,QAAQ,CAAC;AAAA,CAClD","sourcesContent":["import { MODELS } from \"./models.generated.js\";\nimport type { Api, KnownProvider, Model, ModelThinkingLevel, Usage } from \"./types.js\";\n\nconst modelRegistry: Map<string, Map<string, Model<Api>>> = new Map();\n\n// Initialize registry from MODELS on module load\nfor (const [provider, models] of Object.entries(MODELS)) {\n\tconst providerModels = new Map<string, Model<Api>>();\n\tfor (const [id, model] of Object.entries(models)) {\n\t\tproviderModels.set(id, model as Model<Api>);\n\t}\n\tmodelRegistry.set(provider, providerModels);\n}\n\ntype ModelApi<\n\tTProvider extends KnownProvider,\n\tTModelId extends keyof (typeof MODELS)[TProvider],\n> = (typeof MODELS)[TProvider][TModelId] extends { api: infer TApi } ? (TApi extends Api ? TApi : never) : never;\n\ntype ProviderModelApi<TProvider extends KnownProvider> =\n\t(typeof MODELS)[TProvider][keyof (typeof MODELS)[TProvider]] extends {\n\t\tapi: infer TApi;\n\t}\n\t\t? TApi extends Api\n\t\t\t? TApi\n\t\t\t: never\n\t\t: never;\n\nexport function getModel<TProvider extends KnownProvider, TModelId extends keyof (typeof MODELS)[TProvider]>(\n\tprovider: TProvider,\n\tmodelId: TModelId,\n): Model<ModelApi<TProvider, TModelId>>;\nexport function getModel<TProvider extends KnownProvider>(\n\tprovider: TProvider,\n\tmodelId: string,\n): Model<ProviderModelApi<TProvider>> | undefined;\nexport function getModel<TProvider extends KnownProvider, TModelId extends keyof (typeof MODELS)[TProvider]>(\n\tprovider: TProvider,\n\tmodelId: TModelId,\n): Model<ModelApi<TProvider, TModelId>> {\n\tconst providerModels = modelRegistry.get(provider);\n\treturn providerModels?.get(modelId as string) as Model<ModelApi<TProvider, TModelId>>;\n}\n\nexport function getProviders(): KnownProvider[] {\n\treturn Array.from(modelRegistry.keys()) as KnownProvider[];\n}\n\nexport function getModels<TProvider extends KnownProvider>(\n\tprovider: TProvider,\n): Model<ModelApi<TProvider, keyof (typeof MODELS)[TProvider]>>[] {\n\tconst models = modelRegistry.get(provider);\n\treturn models ? (Array.from(models.values()) as Model<ModelApi<TProvider, keyof (typeof MODELS)[TProvider]>>[]) : [];\n}\n\nexport function calculateCost<TApi extends Api>(model: Model<TApi>, usage: Usage): Usage[\"cost\"] {\n\tconst inputTokens = usage.input + usage.cacheRead + usage.cacheWrite;\n\tlet cost: Model<TApi>[\"cost\"] = model.cost;\n\tlet matchedThreshold = -1;\n\tfor (const tier of model.cost.tiers ?? []) {\n\t\tif (inputTokens > tier.inputTokensAbove && tier.inputTokensAbove > matchedThreshold) {\n\t\t\tcost = tier;\n\t\t\tmatchedThreshold = tier.inputTokensAbove;\n\t\t}\n\t}\n\n\tusage.cost.input = (cost.input / 1000000) * usage.input;\n\tusage.cost.output = (cost.output / 1000000) * usage.output;\n\tusage.cost.cacheRead = (cost.cacheRead / 1000000) * usage.cacheRead;\n\t// Anthropic charges 2x base input for 1h cache writes.\n\tconst longWrite = usage.cacheWrite1h ?? 0;\n\tconst shortWrite = usage.cacheWrite - longWrite;\n\tusage.cost.cacheWrite = (cost.cacheWrite * shortWrite + cost.input * 2 * longWrite) / 1000000;\n\tusage.cost.total = usage.cost.input + usage.cost.output + usage.cost.cacheRead + usage.cost.cacheWrite;\n\treturn usage.cost;\n}\n\nconst EXTENDED_THINKING_LEVELS: ModelThinkingLevel[] = [\"off\", \"minimal\", \"low\", \"medium\", \"high\", \"xhigh\", \"max\"];\n\nexport function getSupportedThinkingLevels<TApi extends Api>(model: Model<TApi>): ModelThinkingLevel[] {\n\tif (!model.reasoning) return [\"off\"];\n\n\treturn EXTENDED_THINKING_LEVELS.filter((level) => {\n\t\tconst mapped = model.thinkingLevelMap?.[level];\n\t\tif (mapped === null) return false;\n\t\tif (level === \"xhigh\" || level === \"max\") return mapped !== undefined;\n\t\treturn true;\n\t});\n}\n\nexport function clampThinkingLevel<TApi extends Api>(\n\tmodel: Model<TApi>,\n\tlevel: ModelThinkingLevel,\n): ModelThinkingLevel {\n\tconst availableLevels = getSupportedThinkingLevels(model);\n\tif (availableLevels.includes(level)) return level;\n\n\tconst requestedIndex = EXTENDED_THINKING_LEVELS.indexOf(level);\n\tif (requestedIndex === -1) return availableLevels[0] ?? \"off\";\n\n\tfor (let i = requestedIndex; i < EXTENDED_THINKING_LEVELS.length; i++) {\n\t\tconst candidate = EXTENDED_THINKING_LEVELS[i];\n\t\tif (availableLevels.includes(candidate)) return candidate;\n\t}\n\tfor (let i = requestedIndex - 1; i >= 0; i--) {\n\t\tconst candidate = EXTENDED_THINKING_LEVELS[i];\n\t\tif (availableLevels.includes(candidate)) return candidate;\n\t}\n\treturn availableLevels[0] ?? \"off\";\n}\n\n/**\n * Check if two models are equal by comparing both their id and provider.\n * Returns false if either model is null or undefined.\n */\nexport function modelsAreEqual<TApi extends Api>(\n\ta: Model<TApi> | null | undefined,\n\tb: Model<TApi> | null | undefined,\n): boolean {\n\tif (!a || !b) return false;\n\treturn a.id === b.id && a.provider === b.provider;\n}\n"]}

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

@@ -1 +1 @@
-{"version":3,"file":"amazon-bedrock.d.ts","sourceRoot":"","sources":["../../src/providers/amazon-bedrock.ts"],"names":[],"mappings":"AAwBA,OAAO,KAAK,EAOX,mBAAmB,EAEnB,cAAc,EACd,aAAa,EAEb,eAAe,EAEf,aAAa,EAIb,MAAM,aAAa,CAAC;AASrB,MAAM,MAAM,sBAAsB,GAAG,YAAY,GAAG,SAAS,CAAC;AAE9D,MAAM,WAAW,cAAe,SAAQ,aAAa;IACpD,MAAM,CAAC,EAAE,MAAM,CAAC;IAChB,OAAO,CAAC,EAAE,MAAM,CAAC;IACjB,UAAU,CAAC,EAAE,MAAM,GAAG,KAAK,GAAG,MAAM,GAAG;QAAE,IAAI,EAAE,MAAM,CAAC;QAAC,IAAI,EAAE,MAAM,CAAA;KAAE,CAAC;IAEtE,SAAS,CAAC,EAAE,aAAa,CAAC;IAE1B,eAAe,CAAC,EAAE,eAAe,CAAC;IAElC,mBAAmB,CAAC,EAAE,OAAO,CAAC;IAC9B;;;;;;;;;OASG;IACH,eAAe,CAAC,EAAE,sBAAsB,CAAC;IACzC;;;sGAGkG;IAClG,eAAe,CAAC,EAAE,MAAM,CAAC,MAAM,EAAE,MAAM,CAAC,CAAC;IACzC;;;;yGAIqG;IACrG,WAAW,CAAC,EAAE,MAAM,CAAC;CACrB;AAQD,eAAO,MAAM,aAAa,EAAE,cAAc,CAAC,yBAAyB,EAAE,cAAc,CA6MnF,CAAC;AA8CF,eAAO,MAAM,mBAAmB,EAAE,cAAc,CAAC,yBAAyB,EAAE,mBAAmB,CA0C9F,CAAC","sourcesContent":["import {\n\tBedrockRuntimeClient,\n\ttype BedrockRuntimeClientConfig,\n\tBedrockRuntimeServiceException,\n\tStopReason as BedrockStopReason,\n\ttype Tool as BedrockTool,\n\tCachePointType,\n\tCacheTTL,\n\ttype ContentBlock,\n\ttype ContentBlockDeltaEvent,\n\ttype ContentBlockStartEvent,\n\ttype ContentBlockStopEvent,\n\tConversationRole,\n\tConverseStreamCommand,\n\ttype ConverseStreamMetadataEvent,\n\tImageFormat,\n\ttype Message,\n\ttype SystemContentBlock,\n\ttype ToolChoice,\n\ttype ToolConfiguration,\n\tToolResultStatus,\n} from \"@aws-sdk/client-bedrock-runtime\";\nimport type { DocumentType } from \"@smithy/types\";\nimport { calculateCost } from \"../models.js\";\nimport type {\n\tApi,\n\tAssistantMessage,\n\tCacheRetention,\n\tContext,\n\tModel,\n\tProviderEnv,\n\tSimpleStreamOptions,\n\tStopReason,\n\tStreamFunction,\n\tStreamOptions,\n\tTextContent,\n\tThinkingBudgets,\n\tThinkingContent,\n\tThinkingLevel,\n\tTool,\n\tToolCall,\n\tToolResultMessage,\n} from \"../types.js\";\nimport { normalizeProviderError } from \"../utils/error-body.js\";\nimport { AssistantMessageEventStream } from \"../utils/event-stream.js\";\nimport { parseStreamingJson } from \"../utils/json-parse.js\";\nimport { getProviderEnvValue } from \"../utils/provider-env.js\";\nimport { sanitizeSurrogates } from \"../utils/sanitize-unicode.js\";\nimport { adjustMaxTokensForThinking, buildBaseOptions, clampReasoning } from \"./simple-options.js\";\nimport { transformMessages } from \"./transform-messages.js\";\n\nexport type BedrockThinkingDisplay = \"summarized\" | \"omitted\";\n\nexport interface BedrockOptions extends StreamOptions {\n\tregion?: string;\n\tprofile?: string;\n\ttoolChoice?: \"auto\" | \"any\" | \"none\" | { type: \"tool\"; name: string };\n\t/* See https://docs.aws.amazon.com/bedrock/latest/userguide/inference-reasoning.html for supported models. */\n\treasoning?: ThinkingLevel;\n\t/* Custom token budgets per thinking level. Overrides default budgets. */\n\tthinkingBudgets?: ThinkingBudgets;\n\t/* Only supported by Claude 4.x models, see https://docs.aws.amazon.com/bedrock/latest/userguide/claude-messages-extended-thinking.html#claude-messages-extended-thinking-tool-use-interleaved */\n\tinterleavedThinking?: boolean;\n\t/**\n\t * Controls how Claude's thinking content is returned in responses.\n\t * - \"summarized\": Thinking blocks contain summarized thinking text (default here).\n\t * - \"omitted\": Thinking content is redacted but the signature still travels back\n\t *   for multi-turn continuity, reducing time-to-first-text-token.\n\t *\n\t * Note: Anthropic's API default for Claude Opus 4.7 and Mythos Preview is\n\t * \"omitted\". We default to \"summarized\" here to keep behavior consistent with\n\t * older Claude 4 models. Only applies to Claude models on Bedrock.\n\t */\n\tthinkingDisplay?: BedrockThinkingDisplay;\n\t/** Key-value pairs attached to the inference request for cost allocation tagging.\n\t * Keys: max 64 chars, no `aws:` prefix. Values: max 256 chars. Max 50 pairs.\n\t * Tags appear in AWS Cost Explorer split cost allocation data.\n\t * @see https://docs.aws.amazon.com/bedrock/latest/APIReference/API_runtime_ConverseStream.html */\n\trequestMetadata?: Record<string, string>;\n\t/** Bearer token for Bedrock API key authentication.\n\t * When set, bypasses SigV4 signing and sends Authorization: Bearer <token> instead.\n\t * Requires `bedrock:CallWithBearerToken` IAM permission on the token's identity.\n\t * Set via AWS_BEARER_TOKEN_BEDROCK env var or pass directly.\n\t * @see https://docs.aws.amazon.com/service-authorization/latest/reference/list_amazonbedrock.html */\n\tbearerToken?: string;\n}\n\ntype Block = (TextContent | ThinkingContent | ToolCall) & { index?: number; partialJson?: string };\n\nfunction isAmbientAuthMarker(value: string | undefined): value is \"<authenticated>\" {\n\treturn value === \"<authenticated>\";\n}\n\nexport const streamBedrock: StreamFunction<\"bedrock-converse-stream\", BedrockOptions> = (\n\tmodel: Model<\"bedrock-converse-stream\">,\n\tcontext: Context,\n\toptions: BedrockOptions = {},\n): AssistantMessageEventStream => {\n\tconst stream = new AssistantMessageEventStream();\n\n\t(async () => {\n\t\tconst output: AssistantMessage = {\n\t\t\trole: \"assistant\",\n\t\t\tcontent: [],\n\t\t\tapi: \"bedrock-converse-stream\" as Api,\n\t\t\tprovider: model.provider,\n\t\t\tmodel: model.id,\n\t\t\tusage: {\n\t\t\t\tinput: 0,\n\t\t\t\toutput: 0,\n\t\t\t\tcacheRead: 0,\n\t\t\t\tcacheWrite: 0,\n\t\t\t\ttotalTokens: 0,\n\t\t\t\tcost: { input: 0, output: 0, cacheRead: 0, cacheWrite: 0, total: 0 },\n\t\t\t},\n\t\t\tstopReason: \"stop\",\n\t\t\ttimestamp: Date.now(),\n\t\t};\n\n\t\tconst blocks = output.content as Block[];\n\n\t\tconst config: BedrockRuntimeClientConfig = {\n\t\t\tprofile: options.profile || getProviderEnvValue(\"AWS_PROFILE\", options.env),\n\t\t};\n\t\tconst configuredRegion = getConfiguredBedrockRegion(options);\n\t\tconst hasConfiguredProfile = hasConfiguredBedrockProfile(options.env);\n\t\tconst endpointRegion = getStandardBedrockEndpointRegion(model.baseUrl);\n\t\tconst useExplicitEndpoint = shouldUseExplicitBedrockEndpoint(\n\t\t\tmodel.baseUrl,\n\t\t\tconfiguredRegion,\n\t\t\thasConfiguredProfile,\n\t\t);\n\n\t\t// Only pin standard AWS Bedrock runtime endpoints when no region/profile is configured.\n\t\t// This preserves custom endpoints (VPC/proxy) from #3402 without forcing built-in\n\t\t// catalog defaults such as us-east-1 to override AWS_REGION/AWS_PROFILE.\n\t\tif (useExplicitEndpoint) {\n\t\t\tconfig.endpoint = model.baseUrl;\n\t\t}\n\n\t\t// Resolve bearer token for Bedrock API key auth.\n\t\tconst bearerToken = [\n\t\t\toptions.bearerToken,\n\t\t\toptions.apiKey,\n\t\t\tgetProviderEnvValue(\"AWS_BEARER_TOKEN_BEDROCK\", options.env),\n\t\t].find((value): value is string => !!value && !isAmbientAuthMarker(value));\n\t\tconst useBearerToken =\n\t\t\tbearerToken !== undefined && getProviderEnvValue(\"AWS_BEDROCK_SKIP_AUTH\", options.env) !== \"1\";\n\n\t\t// in Node.js/Bun environment only\n\t\tif (typeof process !== \"undefined\" && (process.versions?.node || process.versions?.bun)) {\n\t\t\t// Region resolution: ARN-embedded > explicit option > env vars > SDK default chain.\n\t\t\t// When the model ID is an inference profile ARN, extract the region from it.\n\t\t\t// This avoids conflicts with AWS_REGION set for other services.\n\t\t\tconst arnRegionMatch = model.id.match(/^arn:aws(?:-[a-z0-9-]+)?:bedrock:([a-z0-9-]+):/);\n\t\t\tif (arnRegionMatch) {\n\t\t\t\tconfig.region = arnRegionMatch[1];\n\t\t\t} else if (configuredRegion) {\n\t\t\t\tconfig.region = configuredRegion;\n\t\t\t} else if (endpointRegion && useExplicitEndpoint) {\n\t\t\t\tconfig.region = endpointRegion;\n\t\t\t} else if (!hasConfiguredProfile) {\n\t\t\t\tconfig.region = \"us-east-1\";\n\t\t\t}\n\n\t\t\t// Support proxies that don't need authentication\n\t\t\tif (getProviderEnvValue(\"AWS_BEDROCK_SKIP_AUTH\", options.env) === \"1\") {\n\t\t\t\tconfig.credentials = {\n\t\t\t\t\taccessKeyId: \"dummy-access-key\",\n\t\t\t\t\tsecretAccessKey: \"dummy-secret-key\",\n\t\t\t\t};\n\t\t\t}\n\n\t\t\tif (\n\t\t\t\tgetProviderEnvValue(\"HTTP_PROXY\", options.env) ||\n\t\t\t\tgetProviderEnvValue(\"HTTPS_PROXY\", options.env) ||\n\t\t\t\tgetProviderEnvValue(\"NO_PROXY\", options.env) ||\n\t\t\t\tgetProviderEnvValue(\"http_proxy\", options.env) ||\n\t\t\t\tgetProviderEnvValue(\"https_proxy\", options.env) ||\n\t\t\t\tgetProviderEnvValue(\"no_proxy\", options.env)\n\t\t\t) {\n\t\t\t\tconst nodeHttpHandler = await import(\"@smithy/node-http-handler\");\n\t\t\t\tconst proxyAgent = await import(\"proxy-agent\");\n\n\t\t\t\tconst agent = new proxyAgent.ProxyAgent();\n\n\t\t\t\t// Bedrock runtime uses NodeHttp2Handler by default since v3.798.0, which is based\n\t\t\t\t// on `http2` module and has no support for http agent.\n\t\t\t\t// Use NodeHttpHandler to support http agent.\n\t\t\t\tconfig.requestHandler = new nodeHttpHandler.NodeHttpHandler({\n\t\t\t\t\thttpAgent: agent,\n\t\t\t\t\thttpsAgent: agent,\n\t\t\t\t});\n\t\t\t} else if (getProviderEnvValue(\"AWS_BEDROCK_FORCE_HTTP1\", options.env) === \"1\") {\n\t\t\t\t// Some custom endpoints require HTTP/1.1 instead of HTTP/2\n\t\t\t\tconst nodeHttpHandler = await import(\"@smithy/node-http-handler\");\n\t\t\t\tconfig.requestHandler = new nodeHttpHandler.NodeHttpHandler();\n\t\t\t}\n\t\t} else {\n\t\t\t// Non-Node environment (browser): fall back to us-east-1 since\n\t\t\t// there's no config file resolution available.\n\t\t\tconfig.region =\n\t\t\t\tconfiguredRegion || (endpointRegion && useExplicitEndpoint ? endpointRegion : undefined) || \"us-east-1\";\n\t\t}\n\n\t\tif (useBearerToken) {\n\t\t\tconfig.token = { token: bearerToken };\n\t\t\tconfig.authSchemePreference = [\"httpBearerAuth\"];\n\t\t}\n\n\t\ttry {\n\t\t\tconst client = new BedrockRuntimeClient(config);\n\t\t\tconst cacheRetention = resolveCacheRetention(options.cacheRetention, options.env);\n\t\t\tconst inferenceMaxTokens = options.maxTokens ?? (isAnthropicClaudeModel(model) ? model.maxTokens : undefined);\n\t\t\tlet commandInput = {\n\t\t\t\tmodelId: model.id,\n\t\t\t\tmessages: convertMessages(context, model, cacheRetention, options.env),\n\t\t\t\tsystem: buildSystemPrompt(context.systemPrompt, model, cacheRetention, options.env),\n\t\t\t\tinferenceConfig: {\n\t\t\t\t\t...(inferenceMaxTokens !== undefined && { maxTokens: inferenceMaxTokens }),\n\t\t\t\t\t...(options.temperature !== undefined && { temperature: options.temperature }),\n\t\t\t\t},\n\t\t\t\ttoolConfig: convertToolConfig(context.tools, options.toolChoice),\n\t\t\t\tadditionalModelRequestFields: buildAdditionalModelRequestFields(model, options),\n\t\t\t\t...(options.requestMetadata !== undefined && { requestMetadata: options.requestMetadata }),\n\t\t\t};\n\t\t\tconst nextCommandInput = await options?.onPayload?.(commandInput, model);\n\t\t\tif (nextCommandInput !== undefined) {\n\t\t\t\tcommandInput = nextCommandInput as typeof commandInput;\n\t\t\t}\n\t\t\tconst command = new ConverseStreamCommand(commandInput);\n\n\t\t\tconst response = await client.send(command, { abortSignal: options.signal });\n\t\t\tif (response.$metadata.httpStatusCode !== undefined) {\n\t\t\t\tconst responseHeaders: Record<string, string> = {};\n\t\t\t\tif (response.$metadata.requestId) {\n\t\t\t\t\tresponseHeaders[\"x-amzn-requestid\"] = response.$metadata.requestId;\n\t\t\t\t}\n\t\t\t\tawait options?.onResponse?.({ status: response.$metadata.httpStatusCode, headers: responseHeaders }, model);\n\t\t\t}\n\n\t\t\tfor await (const item of response.stream!) {\n\t\t\t\tif (item.messageStart) {\n\t\t\t\t\tif (item.messageStart.role !== ConversationRole.ASSISTANT) {\n\t\t\t\t\t\tthrow new Error(\"Unexpected assistant message start but got user message start instead\");\n\t\t\t\t\t}\n\t\t\t\t\tstream.push({ type: \"start\", partial: output });\n\t\t\t\t} else if (item.contentBlockStart) {\n\t\t\t\t\thandleContentBlockStart(item.contentBlockStart, blocks, output, stream);\n\t\t\t\t} else if (item.contentBlockDelta) {\n\t\t\t\t\thandleContentBlockDelta(item.contentBlockDelta, blocks, output, stream);\n\t\t\t\t} else if (item.contentBlockStop) {\n\t\t\t\t\thandleContentBlockStop(item.contentBlockStop, blocks, output, stream);\n\t\t\t\t} else if (item.messageStop) {\n\t\t\t\t\tconst { stopReason, errorMessage } = mapStopReason(item.messageStop.stopReason);\n\t\t\t\t\toutput.stopReason = stopReason;\n\t\t\t\t\tif (errorMessage) {\n\t\t\t\t\t\toutput.errorMessage = errorMessage;\n\t\t\t\t\t}\n\t\t\t\t} else if (item.metadata) {\n\t\t\t\t\thandleMetadata(item.metadata, model, output);\n\t\t\t\t} else if (item.internalServerException) {\n\t\t\t\t\tthrow item.internalServerException;\n\t\t\t\t} else if (item.modelStreamErrorException) {\n\t\t\t\t\tthrow item.modelStreamErrorException;\n\t\t\t\t} else if (item.validationException) {\n\t\t\t\t\tthrow item.validationException;\n\t\t\t\t} else if (item.throttlingException) {\n\t\t\t\t\tthrow item.throttlingException;\n\t\t\t\t} else if (item.serviceUnavailableException) {\n\t\t\t\t\tthrow item.serviceUnavailableException;\n\t\t\t\t}\n\t\t\t}\n\n\t\t\tif (options.signal?.aborted) {\n\t\t\t\tthrow new Error(\"Request was aborted\");\n\t\t\t}\n\n\t\t\tif (output.stopReason === \"error\" || output.stopReason === \"aborted\") {\n\t\t\t\tthrow new Error(output.errorMessage || \"An unknown error occurred\");\n\t\t\t}\n\n\t\t\tstream.push({ type: \"done\", reason: output.stopReason, message: output });\n\t\t\tstream.end();\n\t\t} catch (error) {\n\t\t\tfor (const block of output.content) {\n\t\t\t\tdelete (block as Block).index;\n\t\t\t\t// partialJson is only a streaming scratch buffer; never persist it.\n\t\t\t\tdelete (block as Block).partialJson;\n\t\t\t}\n\t\t\toutput.stopReason = options.signal?.aborted ? \"aborted\" : \"error\";\n\t\t\toutput.errorMessage = formatBedrockError(error);\n\t\t\tstream.push({ type: \"error\", reason: output.stopReason, error: output });\n\t\t\tstream.end();\n\t\t}\n\t})();\n\n\treturn stream;\n};\n\n/**\n * Human-readable prefixes for Bedrock SDK exception names.\n * The downstream retry logic in agent-session matches patterns like\n * `server.?error` and `service.?unavailable`, so we preserve the legacy\n * prefix format rather than using the raw SDK exception name.\n */\nconst BEDROCK_ERROR_PREFIXES: Record<string, string> = {\n\tInternalServerException: \"Internal server error\",\n\tModelStreamErrorException: \"Model stream error\",\n\tValidationException: \"Validation error\",\n\tThrottlingException: \"Throttling error\",\n\tServiceUnavailableException: \"Service unavailable\",\n};\n\n/**\n * Some models reject the account/profile's configured Bedrock data retention mode\n * (e.g. \"data retention mode 'default' is not available for this model\"). Point\n * users at the AWS docs explaining how to configure a supported mode.\n */\nconst BEDROCK_DATA_RETENTION_DOCS_URL = \"https://docs.aws.amazon.com/bedrock/latest/userguide/data-retention.html\";\n\n/**\n * Format a Bedrock error with a human-readable prefix.\n * AWS SDK exceptions (both from `client.send()` and from stream event items)\n * extend BedrockRuntimeServiceException. We map the `.name` to a stable\n * human-readable prefix so downstream consumers (retry logic, context-overflow\n * detection) can distinguish error categories via simple string matching.\n */\nfunction formatBedrockError(error: unknown): string {\n\tconst normalized = normalizeProviderError(error);\n\tconst message =\n\t\t!normalized.messageCarriesBody && normalized.status !== undefined && normalized.body !== undefined\n\t\t\t? `${normalized.status}: ${normalized.body}`\n\t\t\t: normalized.message;\n\tconst dataRetentionHint = /data retention mode/i.test(message)\n\t\t? ` See ${BEDROCK_DATA_RETENTION_DOCS_URL} for supported data retention modes.`\n\t\t: \"\";\n\tif (error instanceof BedrockRuntimeServiceException) {\n\t\tconst prefix = BEDROCK_ERROR_PREFIXES[error.name] ?? error.name;\n\t\treturn `${prefix}: ${message}${dataRetentionHint}`;\n\t}\n\treturn `${message}${dataRetentionHint}`;\n}\n\nexport const streamSimpleBedrock: StreamFunction<\"bedrock-converse-stream\", SimpleStreamOptions> = (\n\tmodel: Model<\"bedrock-converse-stream\">,\n\tcontext: Context,\n\toptions?: SimpleStreamOptions,\n): AssistantMessageEventStream => {\n\tconst base = buildBaseOptions(model, options, undefined);\n\tif (!options?.reasoning) {\n\t\treturn streamBedrock(model, context, { ...base, reasoning: undefined } satisfies BedrockOptions);\n\t}\n\n\tif (isAnthropicClaudeModel(model)) {\n\t\tif (supportsAdaptiveThinking(model.id, model.name)) {\n\t\t\treturn streamBedrock(model, context, {\n\t\t\t\t...base,\n\t\t\t\treasoning: options.reasoning,\n\t\t\t\tthinkingBudgets: options.thinkingBudgets,\n\t\t\t} satisfies BedrockOptions);\n\t\t}\n\n\t\tconst adjusted = adjustMaxTokensForThinking(\n\t\t\tbase.maxTokens || 0,\n\t\t\tmodel.maxTokens,\n\t\t\toptions.reasoning,\n\t\t\toptions.thinkingBudgets,\n\t\t);\n\n\t\treturn streamBedrock(model, context, {\n\t\t\t...base,\n\t\t\tmaxTokens: adjusted.maxTokens,\n\t\t\treasoning: options.reasoning,\n\t\t\tthinkingBudgets: {\n\t\t\t\t...(options.thinkingBudgets || {}),\n\t\t\t\t[clampReasoning(options.reasoning)!]: adjusted.thinkingBudget,\n\t\t\t},\n\t\t} satisfies BedrockOptions);\n\t}\n\n\treturn streamBedrock(model, context, {\n\t\t...base,\n\t\treasoning: options.reasoning,\n\t\tthinkingBudgets: options.thinkingBudgets,\n\t} satisfies BedrockOptions);\n};\n\nfunction handleContentBlockStart(\n\tevent: ContentBlockStartEvent,\n\tblocks: Block[],\n\toutput: AssistantMessage,\n\tstream: AssistantMessageEventStream,\n): void {\n\tconst index = event.contentBlockIndex!;\n\tconst start = event.start;\n\n\tif (start?.toolUse) {\n\t\tconst block: Block = {\n\t\t\ttype: \"toolCall\",\n\t\t\tid: start.toolUse.toolUseId || \"\",\n\t\t\tname: start.toolUse.name || \"\",\n\t\t\targuments: {},\n\t\t\tpartialJson: \"\",\n\t\t\tindex,\n\t\t};\n\t\toutput.content.push(block);\n\t\tstream.push({ type: \"toolcall_start\", contentIndex: blocks.length - 1, partial: output });\n\t}\n}\n\nfunction handleContentBlockDelta(\n\tevent: ContentBlockDeltaEvent,\n\tblocks: Block[],\n\toutput: AssistantMessage,\n\tstream: AssistantMessageEventStream,\n): void {\n\tconst contentBlockIndex = event.contentBlockIndex!;\n\tconst delta = event.delta;\n\tlet index = blocks.findIndex((b) => b.index === contentBlockIndex);\n\tlet block = blocks[index];\n\n\tif (delta?.text !== undefined) {\n\t\t// If no text block exists yet, create one, as `handleContentBlockStart` is not sent for text blocks\n\t\tif (!block) {\n\t\t\tconst newBlock: Block = { type: \"text\", text: \"\", index: contentBlockIndex };\n\t\t\toutput.content.push(newBlock);\n\t\t\tindex = blocks.length - 1;\n\t\t\tblock = blocks[index];\n\t\t\tstream.push({ type: \"text_start\", contentIndex: index, partial: output });\n\t\t}\n\t\tif (block.type === \"text\") {\n\t\t\tblock.text += delta.text;\n\t\t\tstream.push({ type: \"text_delta\", contentIndex: index, delta: delta.text, partial: output });\n\t\t}\n\t} else if (delta?.toolUse && block?.type === \"toolCall\") {\n\t\tblock.partialJson = (block.partialJson || \"\") + (delta.toolUse.input || \"\");\n\t\tblock.arguments = parseStreamingJson(block.partialJson);\n\t\tstream.push({ type: \"toolcall_delta\", contentIndex: index, delta: delta.toolUse.input || \"\", partial: output });\n\t} else if (delta?.reasoningContent) {\n\t\tlet thinkingBlock = block;\n\t\tlet thinkingIndex = index;\n\n\t\tif (!thinkingBlock) {\n\t\t\tconst newBlock: Block = { type: \"thinking\", thinking: \"\", thinkingSignature: \"\", index: contentBlockIndex };\n\t\t\toutput.content.push(newBlock);\n\t\t\tthinkingIndex = blocks.length - 1;\n\t\t\tthinkingBlock = blocks[thinkingIndex];\n\t\t\tstream.push({ type: \"thinking_start\", contentIndex: thinkingIndex, partial: output });\n\t\t}\n\n\t\tif (thinkingBlock?.type === \"thinking\") {\n\t\t\tif (delta.reasoningContent.text) {\n\t\t\t\tthinkingBlock.thinking += delta.reasoningContent.text;\n\t\t\t\tstream.push({\n\t\t\t\t\ttype: \"thinking_delta\",\n\t\t\t\t\tcontentIndex: thinkingIndex,\n\t\t\t\t\tdelta: delta.reasoningContent.text,\n\t\t\t\t\tpartial: output,\n\t\t\t\t});\n\t\t\t}\n\t\t\tif (delta.reasoningContent.signature) {\n\t\t\t\tthinkingBlock.thinkingSignature =\n\t\t\t\t\t(thinkingBlock.thinkingSignature || \"\") + delta.reasoningContent.signature;\n\t\t\t}\n\t\t}\n\t}\n}\n\nfunction handleMetadata(\n\tevent: ConverseStreamMetadataEvent,\n\tmodel: Model<\"bedrock-converse-stream\">,\n\toutput: AssistantMessage,\n): void {\n\tif (event.usage) {\n\t\toutput.usage.input = event.usage.inputTokens || 0;\n\t\toutput.usage.output = event.usage.outputTokens || 0;\n\t\toutput.usage.cacheRead = event.usage.cacheReadInputTokens || 0;\n\t\toutput.usage.cacheWrite = event.usage.cacheWriteInputTokens || 0;\n\t\toutput.usage.totalTokens = event.usage.totalTokens || output.usage.input + output.usage.output;\n\t\tcalculateCost(model, output.usage);\n\t}\n}\n\nfunction handleContentBlockStop(\n\tevent: ContentBlockStopEvent,\n\tblocks: Block[],\n\toutput: AssistantMessage,\n\tstream: AssistantMessageEventStream,\n): void {\n\tconst index = blocks.findIndex((b) => b.index === event.contentBlockIndex);\n\tconst block = blocks[index];\n\tif (!block) return;\n\tdelete (block as Block).index;\n\n\tswitch (block.type) {\n\t\tcase \"text\":\n\t\t\tstream.push({ type: \"text_end\", contentIndex: index, content: block.text, partial: output });\n\t\t\tbreak;\n\t\tcase \"thinking\":\n\t\t\tstream.push({ type: \"thinking_end\", contentIndex: index, content: block.thinking, partial: output });\n\t\t\tbreak;\n\t\tcase \"toolCall\":\n\t\t\tblock.arguments = parseStreamingJson(block.partialJson);\n\t\t\t// Finalize in-place and strip the scratch buffer so replay only\n\t\t\t// carries parsed arguments.\n\t\t\tdelete (block as Block).partialJson;\n\t\t\tstream.push({ type: \"toolcall_end\", contentIndex: index, toolCall: block, partial: output });\n\t\t\tbreak;\n\t}\n}\n\n/**\n * Check if the model supports adaptive thinking (Opus 4.6+, Sonnet 4.6).\n * Checks both model ID and model name to support application inference profiles\n * whose ARNs don't contain the model name.\n */\nfunction getModelMatchCandidates(modelId: string, modelName?: string): string[] {\n\tconst values = modelName ? [modelId, modelName] : [modelId];\n\treturn values.flatMap((value) => {\n\t\tconst lower = value.toLowerCase();\n\t\treturn [lower, lower.replace(/[\\s_.:]+/g, \"-\")];\n\t});\n}\n\nfunction supportsAdaptiveThinking(modelId: string, modelName?: string): boolean {\n\tconst candidates = getModelMatchCandidates(modelId, modelName);\n\treturn candidates.some(\n\t\t(s) =>\n\t\t\ts.includes(\"opus-4-6\") ||\n\t\t\ts.includes(\"opus-4-7\") ||\n\t\t\ts.includes(\"opus-4-8\") ||\n\t\t\ts.includes(\"sonnet-4-6\") ||\n\t\t\ts.includes(\"sonnet-5\") ||\n\t\t\ts.includes(\"fable-5\"),\n\t);\n}\n\nfunction supportsNativeXhighEffort(model: Model<\"bedrock-converse-stream\">): boolean {\n\tconst candidates = getModelMatchCandidates(model.id, model.name);\n\treturn candidates.some(\n\t\t(s) => s.includes(\"opus-4-7\") || s.includes(\"opus-4-8\") || s.includes(\"sonnet-5\") || s.includes(\"fable-5\"),\n\t);\n}\n\nfunction mapThinkingLevelToEffort(\n\tmodel: Model<\"bedrock-converse-stream\">,\n\tlevel: SimpleStreamOptions[\"reasoning\"],\n): \"low\" | \"medium\" | \"high\" | \"xhigh\" | \"max\" {\n\tif (level === \"xhigh\" && supportsNativeXhighEffort(model)) return \"xhigh\";\n\tif (level === \"max\" && supportsAdaptiveThinking(model.id, model.name)) return \"max\";\n\n\tconst mapped = level ? model.thinkingLevelMap?.[level] : undefined;\n\tif (typeof mapped === \"string\") return mapped as \"low\" | \"medium\" | \"high\" | \"xhigh\" | \"max\";\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\tdefault:\n\t\t\treturn \"high\";\n\t}\n}\n\n/**\n * Resolve cache retention preference.\n * Defaults to \"short\" and uses PI_CACHE_RETENTION for backward compatibility.\n */\nfunction resolveCacheRetention(cacheRetention?: CacheRetention, env?: ProviderEnv): CacheRetention {\n\tif (cacheRetention) {\n\t\treturn cacheRetention;\n\t}\n\tif (getProviderEnvValue(\"PI_CACHE_RETENTION\", env) === \"long\") {\n\t\treturn \"long\";\n\t}\n\treturn \"short\";\n}\n\n/**\n * Check if the model is an Anthropic Claude model on Bedrock.\n * Checks both model ID and model name to support application inference profiles\n * whose ARNs don't contain the model name.\n */\nfunction isAnthropicClaudeModel(model: Model<\"bedrock-converse-stream\">): boolean {\n\tconst id = model.id.toLowerCase();\n\tconst name = model.name?.toLowerCase() ?? \"\";\n\treturn (\n\t\tid.includes(\"anthropic.claude\") ||\n\t\tid.includes(\"anthropic/claude\") ||\n\t\tname.includes(\"anthropic.claude\") ||\n\t\tname.includes(\"anthropic/claude\") ||\n\t\tname.includes(\"claude\")\n\t);\n}\n\n/**\n * Check if the model supports prompt caching.\n * Supported: Claude 3.5 Haiku, Claude 3.7 Sonnet, Claude 4.x, and Claude 5 models\n *\n * For base models and system-defined inference profiles the model ID / ARN\n * contains the model name, so we can decide locally.\n *\n * For application inference profiles (whose ARNs don't contain the model name),\n * also checks model.name which is user-controlled via models.json or registerProvider.\n * As a last resort, set AWS_BEDROCK_FORCE_CACHE=1 to enable cache points.\n * Amazon Nova models have automatic caching and don't need explicit cache points.\n */\nfunction supportsPromptCaching(model: Model<\"bedrock-converse-stream\">, env?: ProviderEnv): boolean {\n\tconst candidates = getModelMatchCandidates(model.id, model.name);\n\n\tconst hasClaudeRef = candidates.some((s) => s.includes(\"claude\"));\n\tif (!hasClaudeRef) {\n\t\t// Application inference profiles don't contain the model name in the ARN.\n\t\t// Allow users to force cache points via environment variable.\n\t\tif (getProviderEnvValue(\"AWS_BEDROCK_FORCE_CACHE\", env) === \"1\") return true;\n\t\treturn false;\n\t}\n\t// Claude 4.x models (opus-4, sonnet-4, haiku-4)\n\tif (candidates.some((s) => s.includes(\"-4-\"))) return true;\n\t// Claude 3.7 Sonnet\n\tif (candidates.some((s) => s.includes(\"claude-3-7-sonnet\"))) return true;\n\t// Claude 3.5 Haiku\n\tif (candidates.some((s) => s.includes(\"claude-3-5-haiku\"))) return true;\n\t// Claude 5 models (Fable 5, Sonnet 5)\n\tif (candidates.some((s) => s.includes(\"fable-5\") || s.includes(\"sonnet-5\"))) return true;\n\treturn false;\n}\n\n/**\n * Check if the model supports thinking signatures in reasoningContent.\n * Only Anthropic Claude models support the signature field.\n * Other models (OpenAI, Qwen, Minimax, Moonshot, etc.) reject it with:\n * \"This model doesn't support the reasoningContent.reasoningText.signature field\"\n *\n * Checks both model ID and model name to support application inference profiles.\n */\nfunction supportsThinkingSignature(model: Model<\"bedrock-converse-stream\">): boolean {\n\treturn isAnthropicClaudeModel(model);\n}\n\nfunction buildSystemPrompt(\n\tsystemPrompt: string | undefined,\n\tmodel: Model<\"bedrock-converse-stream\">,\n\tcacheRetention: CacheRetention,\n\tenv?: ProviderEnv,\n): SystemContentBlock[] | undefined {\n\tif (!systemPrompt) return undefined;\n\n\tconst blocks: SystemContentBlock[] = [{ text: sanitizeSurrogates(systemPrompt) }];\n\n\t// Add cache point for supported Claude models when caching is enabled\n\tif (cacheRetention !== \"none\" && supportsPromptCaching(model, env)) {\n\t\tblocks.push({\n\t\t\tcachePoint: { type: CachePointType.DEFAULT, ...(cacheRetention === \"long\" ? { ttl: CacheTTL.ONE_HOUR } : {}) },\n\t\t});\n\t}\n\n\treturn blocks;\n}\n\nfunction normalizeToolCallId(id: string): string {\n\tconst sanitized = id.replace(/[^a-zA-Z0-9_-]/g, \"_\");\n\treturn sanitized.length > 64 ? sanitized.slice(0, 64) : sanitized;\n}\n\nfunction convertMessages(\n\tcontext: Context,\n\tmodel: Model<\"bedrock-converse-stream\">,\n\tcacheRetention: CacheRetention,\n\tenv?: ProviderEnv,\n): Message[] {\n\tconst result: Message[] = [];\n\tconst transformedMessages = transformMessages(context.messages, model, normalizeToolCallId);\n\n\tfor (let i = 0; i < transformedMessages.length; i++) {\n\t\tconst m = transformedMessages[i];\n\n\t\tswitch (m.role) {\n\t\t\tcase \"user\": {\n\t\t\t\tconst content: ContentBlock[] = [];\n\t\t\t\tif (typeof m.content === \"string\") {\n\t\t\t\t\tcontent.push({ text: sanitizeSurrogates(m.content) });\n\t\t\t\t} else {\n\t\t\t\t\tfor (const c of m.content) {\n\t\t\t\t\t\tswitch (c.type) {\n\t\t\t\t\t\t\tcase \"text\":\n\t\t\t\t\t\t\t\tcontent.push({ text: sanitizeSurrogates(c.text) });\n\t\t\t\t\t\t\t\tbreak;\n\t\t\t\t\t\t\tcase \"image\":\n\t\t\t\t\t\t\t\tcontent.push({ image: createImageBlock(c.mimeType, c.data) });\n\t\t\t\t\t\t\t\tbreak;\n\t\t\t\t\t\t\tdefault:\n\t\t\t\t\t\t\t\tcontinue;\n\t\t\t\t\t\t}\n\t\t\t\t\t}\n\t\t\t\t}\n\t\t\t\tif (content.length === 0) continue;\n\t\t\t\tresult.push({\n\t\t\t\t\trole: ConversationRole.USER,\n\t\t\t\t\tcontent,\n\t\t\t\t});\n\t\t\t\tbreak;\n\t\t\t}\n\t\t\tcase \"assistant\": {\n\t\t\t\t// Skip assistant messages with empty content (e.g., from aborted requests)\n\t\t\t\t// Bedrock rejects messages with empty content arrays\n\t\t\t\tif (m.content.length === 0) {\n\t\t\t\t\tcontinue;\n\t\t\t\t}\n\t\t\t\tconst contentBlocks: ContentBlock[] = [];\n\t\t\t\tfor (const c of m.content) {\n\t\t\t\t\tswitch (c.type) {\n\t\t\t\t\t\tcase \"text\":\n\t\t\t\t\t\t\t// Skip empty text blocks\n\t\t\t\t\t\t\tif (c.text.trim().length === 0) continue;\n\t\t\t\t\t\t\tcontentBlocks.push({ text: sanitizeSurrogates(c.text) });\n\t\t\t\t\t\t\tbreak;\n\t\t\t\t\t\tcase \"toolCall\":\n\t\t\t\t\t\t\tcontentBlocks.push({\n\t\t\t\t\t\t\t\ttoolUse: { toolUseId: c.id, name: c.name, input: c.arguments },\n\t\t\t\t\t\t\t});\n\t\t\t\t\t\t\tbreak;\n\t\t\t\t\t\tcase \"thinking\":\n\t\t\t\t\t\t\t// Skip empty thinking blocks\n\t\t\t\t\t\t\tif (c.thinking.trim().length === 0) continue;\n\t\t\t\t\t\t\t// Only Anthropic models support the signature field in reasoningText.\n\t\t\t\t\t\t\t// For other models, we omit the signature to avoid errors like:\n\t\t\t\t\t\t\t// \"This model doesn't support the reasoningContent.reasoningText.signature field\"\n\t\t\t\t\t\t\tif (supportsThinkingSignature(model)) {\n\t\t\t\t\t\t\t\t// Signatures arrive after thinking deltas. If a partial or externally\n\t\t\t\t\t\t\t\t// persisted message lacks a signature, Bedrock rejects the replayed\n\t\t\t\t\t\t\t\t// reasoning block. Fall back to plain text, matching Anthropic.\n\t\t\t\t\t\t\t\tif (!c.thinkingSignature || c.thinkingSignature.trim().length === 0) {\n\t\t\t\t\t\t\t\t\tcontentBlocks.push({ text: sanitizeSurrogates(c.thinking) });\n\t\t\t\t\t\t\t\t} else {\n\t\t\t\t\t\t\t\t\tcontentBlocks.push({\n\t\t\t\t\t\t\t\t\t\treasoningContent: {\n\t\t\t\t\t\t\t\t\t\t\treasoningText: {\n\t\t\t\t\t\t\t\t\t\t\t\ttext: sanitizeSurrogates(c.thinking),\n\t\t\t\t\t\t\t\t\t\t\t\tsignature: c.thinkingSignature,\n\t\t\t\t\t\t\t\t\t\t\t},\n\t\t\t\t\t\t\t\t\t\t},\n\t\t\t\t\t\t\t\t\t});\n\t\t\t\t\t\t\t\t}\n\t\t\t\t\t\t\t} else {\n\t\t\t\t\t\t\t\tcontentBlocks.push({\n\t\t\t\t\t\t\t\t\treasoningContent: {\n\t\t\t\t\t\t\t\t\t\treasoningText: { text: sanitizeSurrogates(c.thinking) },\n\t\t\t\t\t\t\t\t\t},\n\t\t\t\t\t\t\t\t});\n\t\t\t\t\t\t\t}\n\t\t\t\t\t\t\tbreak;\n\t\t\t\t\t\tdefault:\n\t\t\t\t\t\t\tcontinue;\n\t\t\t\t\t}\n\t\t\t\t}\n\t\t\t\t// Skip if all content blocks were filtered out\n\t\t\t\tif (contentBlocks.length === 0) {\n\t\t\t\t\tcontinue;\n\t\t\t\t}\n\t\t\t\tresult.push({\n\t\t\t\t\trole: ConversationRole.ASSISTANT,\n\t\t\t\t\tcontent: contentBlocks,\n\t\t\t\t});\n\t\t\t\tbreak;\n\t\t\t}\n\t\t\tcase \"toolResult\": {\n\t\t\t\t// Collect all consecutive toolResult messages into a single user message\n\t\t\t\t// Bedrock requires all tool results to be in one message\n\t\t\t\tconst toolResults: ContentBlock.ToolResultMember[] = [];\n\n\t\t\t\t// Add current tool result with all content blocks combined\n\t\t\t\ttoolResults.push({\n\t\t\t\t\ttoolResult: {\n\t\t\t\t\t\ttoolUseId: m.toolCallId,\n\t\t\t\t\t\tcontent: m.content.map((c) =>\n\t\t\t\t\t\t\tc.type === \"image\"\n\t\t\t\t\t\t\t\t? { image: createImageBlock(c.mimeType, c.data) }\n\t\t\t\t\t\t\t\t: { text: sanitizeSurrogates(c.text) },\n\t\t\t\t\t\t),\n\t\t\t\t\t\tstatus: m.isError ? ToolResultStatus.ERROR : ToolResultStatus.SUCCESS,\n\t\t\t\t\t},\n\t\t\t\t});\n\n\t\t\t\t// Look ahead for consecutive toolResult messages\n\t\t\t\tlet j = i + 1;\n\t\t\t\twhile (j < transformedMessages.length && transformedMessages[j].role === \"toolResult\") {\n\t\t\t\t\tconst nextMsg = transformedMessages[j] as ToolResultMessage;\n\t\t\t\t\ttoolResults.push({\n\t\t\t\t\t\ttoolResult: {\n\t\t\t\t\t\t\ttoolUseId: nextMsg.toolCallId,\n\t\t\t\t\t\t\tcontent: nextMsg.content.map((c) =>\n\t\t\t\t\t\t\t\tc.type === \"image\"\n\t\t\t\t\t\t\t\t\t? { image: createImageBlock(c.mimeType, c.data) }\n\t\t\t\t\t\t\t\t\t: { text: sanitizeSurrogates(c.text) },\n\t\t\t\t\t\t\t),\n\t\t\t\t\t\t\tstatus: nextMsg.isError ? ToolResultStatus.ERROR : ToolResultStatus.SUCCESS,\n\t\t\t\t\t\t},\n\t\t\t\t\t});\n\t\t\t\t\tj++;\n\t\t\t\t}\n\n\t\t\t\t// Skip the messages we've already processed\n\t\t\t\ti = j - 1;\n\n\t\t\t\tresult.push({\n\t\t\t\t\trole: ConversationRole.USER,\n\t\t\t\t\tcontent: toolResults,\n\t\t\t\t});\n\t\t\t\tbreak;\n\t\t\t}\n\t\t\tdefault:\n\t\t\t\tcontinue;\n\t\t}\n\t}\n\n\t// Add cache point to the last user message for supported Claude models when caching is enabled\n\tif (cacheRetention !== \"none\" && supportsPromptCaching(model, env) && result.length > 0) {\n\t\tconst lastMessage = result[result.length - 1];\n\t\tif (lastMessage.role === ConversationRole.USER && lastMessage.content) {\n\t\t\t(lastMessage.content as ContentBlock[]).push({\n\t\t\t\tcachePoint: {\n\t\t\t\t\ttype: CachePointType.DEFAULT,\n\t\t\t\t\t...(cacheRetention === \"long\" ? { ttl: CacheTTL.ONE_HOUR } : {}),\n\t\t\t\t},\n\t\t\t});\n\t\t}\n\t}\n\n\treturn result;\n}\n\nfunction convertToolConfig(\n\ttools: Tool[] | undefined,\n\ttoolChoice: BedrockOptions[\"toolChoice\"],\n): ToolConfiguration | undefined {\n\tif (!tools?.length || toolChoice === \"none\") return undefined;\n\n\tconst bedrockTools: BedrockTool[] = tools.map((tool) => ({\n\t\ttoolSpec: {\n\t\t\tname: tool.name,\n\t\t\tdescription: tool.description,\n\t\t\tinputSchema: { json: tool.parameters as unknown as DocumentType },\n\t\t},\n\t}));\n\n\tlet bedrockToolChoice: ToolChoice | undefined;\n\tswitch (toolChoice) {\n\t\tcase \"auto\":\n\t\t\tbedrockToolChoice = { auto: {} };\n\t\t\tbreak;\n\t\tcase \"any\":\n\t\t\tbedrockToolChoice = { any: {} };\n\t\t\tbreak;\n\t\tdefault:\n\t\t\tif (toolChoice?.type === \"tool\") {\n\t\t\t\tbedrockToolChoice = { tool: { name: toolChoice.name } };\n\t\t\t}\n\t}\n\n\treturn { tools: bedrockTools, toolChoice: bedrockToolChoice };\n}\n\nfunction mapStopReason(reason: string | undefined): { stopReason: StopReason; errorMessage?: string } {\n\tswitch (reason) {\n\t\tcase BedrockStopReason.END_TURN:\n\t\tcase BedrockStopReason.STOP_SEQUENCE:\n\t\t\treturn { stopReason: \"stop\" };\n\t\tcase BedrockStopReason.MAX_TOKENS:\n\t\tcase BedrockStopReason.MODEL_CONTEXT_WINDOW_EXCEEDED:\n\t\t\treturn { stopReason: \"length\" };\n\t\tcase BedrockStopReason.TOOL_USE:\n\t\t\treturn { stopReason: \"toolUse\" };\n\t\tdefault:\n\t\t\treturn reason ? { stopReason: \"error\", errorMessage: reason } : { stopReason: \"error\" };\n\t}\n}\n\nfunction getConfiguredBedrockRegion(options: BedrockOptions): string | undefined {\n\tif (typeof process === \"undefined\") {\n\t\treturn options.region;\n\t}\n\n\treturn (\n\t\toptions.region ||\n\t\tgetProviderEnvValue(\"AWS_REGION\", options.env) ||\n\t\tgetProviderEnvValue(\"AWS_DEFAULT_REGION\", options.env) ||\n\t\tundefined\n\t);\n}\n\nfunction hasConfiguredBedrockProfile(env?: ProviderEnv): boolean {\n\tif (typeof process === \"undefined\") {\n\t\treturn false;\n\t}\n\n\treturn Boolean(getProviderEnvValue(\"AWS_PROFILE\", env));\n}\n\nfunction getStandardBedrockEndpointRegion(baseUrl: string | undefined): string | undefined {\n\tif (!baseUrl) {\n\t\treturn undefined;\n\t}\n\n\ttry {\n\t\tconst { hostname } = new URL(baseUrl);\n\t\tconst match = hostname.toLowerCase().match(/^bedrock-runtime(?:-fips)?\\.([a-z0-9-]+)\\.amazonaws\\.com(?:\\.cn)?$/);\n\t\treturn match?.[1];\n\t} catch {\n\t\treturn undefined;\n\t}\n}\n\nfunction shouldUseExplicitBedrockEndpoint(\n\tbaseUrl: string,\n\tconfiguredRegion: string | undefined,\n\thasConfiguredProfile: boolean,\n): boolean {\n\tconst endpointRegion = getStandardBedrockEndpointRegion(baseUrl);\n\tif (!endpointRegion) {\n\t\treturn true;\n\t}\n\n\treturn !configuredRegion && !hasConfiguredProfile;\n}\n\nfunction isGovCloudBedrockTarget(model: Model<\"bedrock-converse-stream\">, options: BedrockOptions): boolean {\n\tconst region = getConfiguredBedrockRegion(options);\n\tif (region?.toLowerCase().startsWith(\"us-gov-\")) {\n\t\treturn true;\n\t}\n\n\tconst modelId = model.id.toLowerCase();\n\treturn modelId.startsWith(\"us-gov.\") || modelId.startsWith(\"arn:aws-us-gov:\");\n}\n\nfunction buildAdditionalModelRequestFields(\n\tmodel: Model<\"bedrock-converse-stream\">,\n\toptions: BedrockOptions,\n): Record<string, any> | undefined {\n\tif (!options.reasoning || !model.reasoning) {\n\t\treturn undefined;\n\t}\n\n\tif (isAnthropicClaudeModel(model)) {\n\t\t// GovCloud Bedrock currently rejects the Claude thinking.display field.\n\t\t// Omit it there until the GovCloud Converse schema catches up.\n\t\tconst display = isGovCloudBedrockTarget(model, options) ? undefined : (options.thinkingDisplay ?? \"summarized\");\n\t\tconst result: Record<string, any> = supportsAdaptiveThinking(model.id, model.name)\n\t\t\t? {\n\t\t\t\t\tthinking: { type: \"adaptive\", ...(display !== undefined ? { display } : {}) },\n\t\t\t\t\toutput_config: { effort: mapThinkingLevelToEffort(model, options.reasoning) },\n\t\t\t\t}\n\t\t\t: (() => {\n\t\t\t\t\tconst defaultBudgets: Record<ThinkingLevel, number> = {\n\t\t\t\t\t\tminimal: 1024,\n\t\t\t\t\t\tlow: 2048,\n\t\t\t\t\t\tmedium: 8192,\n\t\t\t\t\t\thigh: 16384,\n\t\t\t\t\t\txhigh: 16384, // Claude doesn't support xhigh, clamp to high\n\t\t\t\t\t\tmax: 16384, // Non-adaptive Claude models do not support max effort\n\t\t\t\t\t};\n\n\t\t\t\t\t// Custom budgets override defaults; extended effort levels use the high budget.\n\t\t\t\t\tconst level = options.reasoning === \"xhigh\" || options.reasoning === \"max\" ? \"high\" : options.reasoning;\n\t\t\t\t\tconst budget = options.thinkingBudgets?.[level] ?? defaultBudgets[level];\n\n\t\t\t\t\treturn {\n\t\t\t\t\t\tthinking: {\n\t\t\t\t\t\t\ttype: \"enabled\",\n\t\t\t\t\t\t\tbudget_tokens: budget,\n\t\t\t\t\t\t\t...(display !== undefined ? { display } : {}),\n\t\t\t\t\t\t},\n\t\t\t\t\t};\n\t\t\t\t})();\n\n\t\tif (!supportsAdaptiveThinking(model.id, model.name) && (options.interleavedThinking ?? true)) {\n\t\t\tresult.anthropic_beta = [\"interleaved-thinking-2025-05-14\"];\n\t\t}\n\n\t\treturn result;\n\t}\n\n\treturn undefined;\n}\n\nfunction createImageBlock(mimeType: string, data: string) {\n\tlet format: ImageFormat;\n\tswitch (mimeType) {\n\t\tcase \"image/jpeg\":\n\t\tcase \"image/jpg\":\n\t\t\tformat = ImageFormat.JPEG;\n\t\t\tbreak;\n\t\tcase \"image/png\":\n\t\t\tformat = ImageFormat.PNG;\n\t\t\tbreak;\n\t\tcase \"image/gif\":\n\t\t\tformat = ImageFormat.GIF;\n\t\t\tbreak;\n\t\tcase \"image/webp\":\n\t\t\tformat = ImageFormat.WEBP;\n\t\t\tbreak;\n\t\tdefault:\n\t\t\tthrow new Error(`Unknown image type: ${mimeType}`);\n\t}\n\n\tconst binaryString = atob(data);\n\tconst bytes = new Uint8Array(binaryString.length);\n\tfor (let i = 0; i < binaryString.length; i++) {\n\t\tbytes[i] = binaryString.charCodeAt(i);\n\t}\n\n\treturn { source: { bytes }, format };\n}\n"]}
+{"version":3,"file":"amazon-bedrock.d.ts","sourceRoot":"","sources":["../../src/providers/amazon-bedrock.ts"],"names":[],"mappings":"AAwBA,OAAO,KAAK,EAQX,mBAAmB,EAEnB,cAAc,EACd,aAAa,EAEb,eAAe,EAEf,aAAa,EAIb,MAAM,aAAa,CAAC;AAerB,MAAM,MAAM,sBAAsB,GAAG,YAAY,GAAG,SAAS,CAAC;AAE9D,MAAM,WAAW,cAAe,SAAQ,aAAa;IACpD,MAAM,CAAC,EAAE,MAAM,CAAC;IAChB,OAAO,CAAC,EAAE,MAAM,CAAC;IACjB,UAAU,CAAC,EAAE,MAAM,GAAG,KAAK,GAAG,MAAM,GAAG;QAAE,IAAI,EAAE,MAAM,CAAC;QAAC,IAAI,EAAE,MAAM,CAAA;KAAE,CAAC;IAEtE,SAAS,CAAC,EAAE,aAAa,CAAC;IAE1B,eAAe,CAAC,EAAE,eAAe,CAAC;IAElC,mBAAmB,CAAC,EAAE,OAAO,CAAC;IAC9B;;;;;;;;;OASG;IACH,eAAe,CAAC,EAAE,sBAAsB,CAAC;IACzC;;;sGAGkG;IAClG,eAAe,CAAC,EAAE,MAAM,CAAC,MAAM,EAAE,MAAM,CAAC,CAAC;IACzC;;;;yGAIqG;IACrG,WAAW,CAAC,EAAE,MAAM,CAAC;CACrB;AAgBD,eAAO,MAAM,aAAa,EAAE,cAAc,CAAC,yBAAyB,EAAE,cAAc,CAqNnF,CAAC;AAiFF,eAAO,MAAM,mBAAmB,EAAE,cAAc,CAAC,yBAAyB,EAAE,mBAAmB,CA8C9F,CAAC","sourcesContent":["import {\n\tBedrockRuntimeClient,\n\ttype BedrockRuntimeClientConfig,\n\tBedrockRuntimeServiceException,\n\tStopReason as BedrockStopReason,\n\ttype Tool as BedrockTool,\n\tCachePointType,\n\tCacheTTL,\n\ttype ContentBlock,\n\ttype ContentBlockDeltaEvent,\n\ttype ContentBlockStartEvent,\n\ttype ContentBlockStopEvent,\n\tConversationRole,\n\tConverseStreamCommand,\n\ttype ConverseStreamMetadataEvent,\n\tImageFormat,\n\ttype Message,\n\ttype SystemContentBlock,\n\ttype ToolChoice,\n\ttype ToolConfiguration,\n\tToolResultStatus,\n} from \"@aws-sdk/client-bedrock-runtime\";\nimport type { DeserializeMiddleware, DocumentType, HttpResponse, MetadataBearer } from \"@smithy/types\";\nimport { calculateCost } from \"../models.js\";\nimport type {\n\tApi,\n\tAssistantMessage,\n\tCacheRetention,\n\tContext,\n\tModel,\n\tProviderEnv,\n\tProviderResponse,\n\tSimpleStreamOptions,\n\tStopReason,\n\tStreamFunction,\n\tStreamOptions,\n\tTextContent,\n\tThinkingBudgets,\n\tThinkingContent,\n\tThinkingLevel,\n\tTool,\n\tToolCall,\n\tToolResultMessage,\n} from \"../types.js\";\nimport { normalizeProviderError } from \"../utils/error-body.js\";\nimport { AssistantMessageEventStream } from \"../utils/event-stream.js\";\nimport { finalizeToolCallArguments, parseStreamingJson } from \"../utils/json-parse.js\";\nimport { getProviderEnvValue } from \"../utils/provider-env.js\";\nimport { sanitizeSurrogates } from \"../utils/sanitize-unicode.js\";\nimport { createProviderToolCallIdNormalizer } from \"../utils/tool-call-id.js\";\nimport {\n\tadjustMaxTokensForThinking,\n\tbuildBaseOptions,\n\tclampMaxTokensToContext,\n\tclampReasoning,\n} from \"./simple-options.js\";\nimport { transformMessages } from \"./transform-messages.js\";\n\nexport type BedrockThinkingDisplay = \"summarized\" | \"omitted\";\n\nexport interface BedrockOptions extends StreamOptions {\n\tregion?: string;\n\tprofile?: string;\n\ttoolChoice?: \"auto\" | \"any\" | \"none\" | { type: \"tool\"; name: string };\n\t/* See https://docs.aws.amazon.com/bedrock/latest/userguide/inference-reasoning.html for supported models. */\n\treasoning?: ThinkingLevel;\n\t/* Custom token budgets per thinking level. Overrides default budgets. */\n\tthinkingBudgets?: ThinkingBudgets;\n\t/* Only supported by Claude 4.x models, see https://docs.aws.amazon.com/bedrock/latest/userguide/claude-messages-extended-thinking.html#claude-messages-extended-thinking-tool-use-interleaved */\n\tinterleavedThinking?: boolean;\n\t/**\n\t * Controls how Claude's thinking content is returned in responses.\n\t * - \"summarized\": Thinking blocks contain summarized thinking text (default here).\n\t * - \"omitted\": Thinking content is redacted but the signature still travels back\n\t *   for multi-turn continuity, reducing time-to-first-text-token.\n\t *\n\t * Note: Anthropic's API default for Claude Opus 4.7 and Mythos Preview is\n\t * \"omitted\". We default to \"summarized\" here to keep behavior consistent with\n\t * older Claude 4 models. Only applies to Claude models on Bedrock.\n\t */\n\tthinkingDisplay?: BedrockThinkingDisplay;\n\t/** Key-value pairs attached to the inference request for cost allocation tagging.\n\t * Keys: max 64 chars, no `aws:` prefix. Values: max 256 chars. Max 50 pairs.\n\t * Tags appear in AWS Cost Explorer split cost allocation data.\n\t * @see https://docs.aws.amazon.com/bedrock/latest/APIReference/API_runtime_ConverseStream.html */\n\trequestMetadata?: Record<string, string>;\n\t/** Bearer token for Bedrock API key authentication.\n\t * When set, bypasses SigV4 signing and sends Authorization: Bearer <token> instead.\n\t * Requires `bedrock:CallWithBearerToken` IAM permission on the token's identity.\n\t * Set via AWS_BEARER_TOKEN_BEDROCK env var or pass directly.\n\t * @see https://docs.aws.amazon.com/service-authorization/latest/reference/list_amazonbedrock.html */\n\tbearerToken?: string;\n}\n\ntype Block = (TextContent | ThinkingContent | ToolCall) & {\n\tindex?: number;\n\tpartialJson?: string;\n\t/** Scratch buffer for encrypted reasoning deltas, joined into `thinkingSignature`. */\n\tredactedChunks?: Uint8Array[];\n};\n\n/** Matches the placeholder the Anthropic API path uses for redacted thinking. */\nconst REDACTED_THINKING_PLACEHOLDER = \"[Reasoning redacted]\";\n\nfunction isAmbientAuthMarker(value: string | undefined): value is \"<authenticated>\" {\n\treturn value === \"<authenticated>\";\n}\n\nexport const streamBedrock: StreamFunction<\"bedrock-converse-stream\", BedrockOptions> = (\n\tmodel: Model<\"bedrock-converse-stream\">,\n\tcontext: Context,\n\toptions: BedrockOptions = {},\n): AssistantMessageEventStream => {\n\tconst stream = new AssistantMessageEventStream();\n\n\t(async () => {\n\t\tconst output: AssistantMessage = {\n\t\t\trole: \"assistant\",\n\t\t\tcontent: [],\n\t\t\tapi: \"bedrock-converse-stream\" as Api,\n\t\t\tprovider: model.provider,\n\t\t\tmodel: model.id,\n\t\t\tusage: {\n\t\t\t\tinput: 0,\n\t\t\t\toutput: 0,\n\t\t\t\tcacheRead: 0,\n\t\t\t\tcacheWrite: 0,\n\t\t\t\ttotalTokens: 0,\n\t\t\t\tcost: { input: 0, output: 0, cacheRead: 0, cacheWrite: 0, total: 0 },\n\t\t\t},\n\t\t\tstopReason: \"stop\",\n\t\t\ttimestamp: Date.now(),\n\t\t};\n\n\t\tconst blocks = output.content as Block[];\n\n\t\tconst config: BedrockRuntimeClientConfig = {\n\t\t\tprofile: options.profile || getProviderEnvValue(\"AWS_PROFILE\", options.env),\n\t\t};\n\t\tconst configuredRegion = getConfiguredBedrockRegion(options);\n\t\tconst hasConfiguredProfile = hasConfiguredBedrockProfile(options.env);\n\t\tconst endpointRegion = getStandardBedrockEndpointRegion(model.baseUrl);\n\t\tconst useExplicitEndpoint = shouldUseExplicitBedrockEndpoint(\n\t\t\tmodel.baseUrl,\n\t\t\tconfiguredRegion,\n\t\t\thasConfiguredProfile,\n\t\t);\n\n\t\t// Only pin standard AWS Bedrock runtime endpoints when no region/profile is configured.\n\t\t// This preserves custom endpoints (VPC/proxy) from #3402 without forcing built-in\n\t\t// catalog defaults such as us-east-1 to override AWS_REGION/AWS_PROFILE.\n\t\tif (useExplicitEndpoint) {\n\t\t\tconfig.endpoint = model.baseUrl;\n\t\t}\n\n\t\t// Resolve bearer token for Bedrock API key auth.\n\t\tconst bearerToken = [\n\t\t\toptions.bearerToken,\n\t\t\toptions.apiKey,\n\t\t\tgetProviderEnvValue(\"AWS_BEARER_TOKEN_BEDROCK\", options.env),\n\t\t].find((value): value is string => !!value && !isAmbientAuthMarker(value));\n\t\tconst useBearerToken =\n\t\t\tbearerToken !== undefined && getProviderEnvValue(\"AWS_BEDROCK_SKIP_AUTH\", options.env) !== \"1\";\n\n\t\t// in Node.js/Bun environment only\n\t\tif (typeof process !== \"undefined\" && (process.versions?.node || process.versions?.bun)) {\n\t\t\t// Region resolution: ARN-embedded > explicit option > env vars > SDK default chain.\n\t\t\t// When the model ID is an inference profile ARN, extract the region from it.\n\t\t\t// This avoids conflicts with AWS_REGION set for other services.\n\t\t\tconst arnRegionMatch = model.id.match(/^arn:aws(?:-[a-z0-9-]+)?:bedrock:([a-z0-9-]+):/);\n\t\t\tif (arnRegionMatch) {\n\t\t\t\tconfig.region = arnRegionMatch[1];\n\t\t\t} else if (configuredRegion) {\n\t\t\t\tconfig.region = configuredRegion;\n\t\t\t} else if (endpointRegion && useExplicitEndpoint) {\n\t\t\t\tconfig.region = endpointRegion;\n\t\t\t} else if (!hasConfiguredProfile) {\n\t\t\t\tconfig.region = \"us-east-1\";\n\t\t\t}\n\n\t\t\t// Support proxies that don't need authentication\n\t\t\tif (getProviderEnvValue(\"AWS_BEDROCK_SKIP_AUTH\", options.env) === \"1\") {\n\t\t\t\tconfig.credentials = {\n\t\t\t\t\taccessKeyId: \"dummy-access-key\",\n\t\t\t\t\tsecretAccessKey: \"dummy-secret-key\",\n\t\t\t\t};\n\t\t\t}\n\n\t\t\tif (\n\t\t\t\tgetProviderEnvValue(\"HTTP_PROXY\", options.env) ||\n\t\t\t\tgetProviderEnvValue(\"HTTPS_PROXY\", options.env) ||\n\t\t\t\tgetProviderEnvValue(\"NO_PROXY\", options.env) ||\n\t\t\t\tgetProviderEnvValue(\"http_proxy\", options.env) ||\n\t\t\t\tgetProviderEnvValue(\"https_proxy\", options.env) ||\n\t\t\t\tgetProviderEnvValue(\"no_proxy\", options.env)\n\t\t\t) {\n\t\t\t\tconst nodeHttpHandler = await import(\"@smithy/node-http-handler\");\n\t\t\t\tconst proxyAgent = await import(\"proxy-agent\");\n\n\t\t\t\tconst agent = new proxyAgent.ProxyAgent();\n\n\t\t\t\t// Bedrock runtime uses NodeHttp2Handler by default since v3.798.0, which is based\n\t\t\t\t// on `http2` module and has no support for http agent.\n\t\t\t\t// Use NodeHttpHandler to support http agent.\n\t\t\t\tconfig.requestHandler = new nodeHttpHandler.NodeHttpHandler({\n\t\t\t\t\thttpAgent: agent,\n\t\t\t\t\thttpsAgent: agent,\n\t\t\t\t});\n\t\t\t} else if (getProviderEnvValue(\"AWS_BEDROCK_FORCE_HTTP1\", options.env) === \"1\") {\n\t\t\t\t// Some custom endpoints require HTTP/1.1 instead of HTTP/2\n\t\t\t\tconst nodeHttpHandler = await import(\"@smithy/node-http-handler\");\n\t\t\t\tconfig.requestHandler = new nodeHttpHandler.NodeHttpHandler();\n\t\t\t}\n\t\t} else {\n\t\t\t// Non-Node environment (browser): fall back to us-east-1 since\n\t\t\t// there's no config file resolution available.\n\t\t\tconfig.region =\n\t\t\t\tconfiguredRegion || (endpointRegion && useExplicitEndpoint ? endpointRegion : undefined) || \"us-east-1\";\n\t\t}\n\n\t\tif (useBearerToken) {\n\t\t\tconfig.token = { token: bearerToken };\n\t\t\tconfig.authSchemePreference = [\"httpBearerAuth\"];\n\t\t}\n\n\t\ttry {\n\t\t\tconst client = new BedrockRuntimeClient(config);\n\t\t\tlet observedRawResponse = false;\n\t\t\tif (options.onResponse) {\n\t\t\t\taddResponseHeadersMiddleware(client, options.onResponse, model, () => {\n\t\t\t\t\tobservedRawResponse = true;\n\t\t\t\t});\n\t\t\t}\n\t\t\tconst cacheRetention = resolveCacheRetention(options.cacheRetention, options.env);\n\t\t\tconst inferenceMaxTokens = options.maxTokens ?? (isAnthropicClaudeModel(model) ? model.maxTokens : undefined);\n\t\t\tlet commandInput = {\n\t\t\t\tmodelId: model.id,\n\t\t\t\tmessages: convertMessages(context, model, cacheRetention, options.env),\n\t\t\t\tsystem: buildSystemPrompt(context.systemPrompt, model, cacheRetention, options.env),\n\t\t\t\tinferenceConfig: {\n\t\t\t\t\t...(inferenceMaxTokens !== undefined && { maxTokens: inferenceMaxTokens }),\n\t\t\t\t\t...(options.temperature !== undefined && { temperature: options.temperature }),\n\t\t\t\t},\n\t\t\t\ttoolConfig: convertToolConfig(context.tools, options.toolChoice),\n\t\t\t\tadditionalModelRequestFields: buildAdditionalModelRequestFields(model, options),\n\t\t\t\t...(options.requestMetadata !== undefined && { requestMetadata: options.requestMetadata }),\n\t\t\t};\n\t\t\tconst nextCommandInput = await options?.onPayload?.(commandInput, model);\n\t\t\tif (nextCommandInput !== undefined) {\n\t\t\t\tcommandInput = nextCommandInput as typeof commandInput;\n\t\t\t}\n\t\t\tconst command = new ConverseStreamCommand(commandInput);\n\n\t\t\tconst response = await client.send(command, { abortSignal: options.signal });\n\t\t\tif (!observedRawResponse && response.$metadata.httpStatusCode !== undefined) {\n\t\t\t\tconst responseHeaders: Record<string, string> = {};\n\t\t\t\tif (response.$metadata.requestId) {\n\t\t\t\t\tresponseHeaders[\"x-amzn-requestid\"] = response.$metadata.requestId;\n\t\t\t\t}\n\t\t\t\tawait options?.onResponse?.({ status: response.$metadata.httpStatusCode, headers: responseHeaders }, model);\n\t\t\t}\n\n\t\t\tfor await (const item of response.stream!) {\n\t\t\t\tif (item.messageStart) {\n\t\t\t\t\tif (item.messageStart.role !== ConversationRole.ASSISTANT) {\n\t\t\t\t\t\tthrow new Error(\"Unexpected assistant message start but got user message start instead\");\n\t\t\t\t\t}\n\t\t\t\t\tstream.push({ type: \"start\", partial: output });\n\t\t\t\t} else if (item.contentBlockStart) {\n\t\t\t\t\thandleContentBlockStart(item.contentBlockStart, blocks, output, stream);\n\t\t\t\t} else if (item.contentBlockDelta) {\n\t\t\t\t\thandleContentBlockDelta(item.contentBlockDelta, blocks, output, stream);\n\t\t\t\t} else if (item.contentBlockStop) {\n\t\t\t\t\thandleContentBlockStop(item.contentBlockStop, blocks, output, stream);\n\t\t\t\t} else if (item.messageStop) {\n\t\t\t\t\toutput.rawStopReason = item.messageStop.stopReason;\n\t\t\t\t\tconst { stopReason, errorMessage } = mapStopReason(item.messageStop.stopReason);\n\t\t\t\t\toutput.stopReason = stopReason;\n\t\t\t\t\tif (errorMessage) {\n\t\t\t\t\t\toutput.errorMessage = errorMessage;\n\t\t\t\t\t}\n\t\t\t\t} else if (item.metadata) {\n\t\t\t\t\thandleMetadata(item.metadata, model, output);\n\t\t\t\t} else if (item.internalServerException) {\n\t\t\t\t\tthrow item.internalServerException;\n\t\t\t\t} else if (item.modelStreamErrorException) {\n\t\t\t\t\tthrow item.modelStreamErrorException;\n\t\t\t\t} else if (item.validationException) {\n\t\t\t\t\tthrow item.validationException;\n\t\t\t\t} else if (item.throttlingException) {\n\t\t\t\t\tthrow item.throttlingException;\n\t\t\t\t} else if (item.serviceUnavailableException) {\n\t\t\t\t\tthrow item.serviceUnavailableException;\n\t\t\t\t}\n\t\t\t}\n\n\t\t\tif (options.signal?.aborted) {\n\t\t\t\tthrow new Error(\"Request was aborted\");\n\t\t\t}\n\n\t\t\tif (output.stopReason === \"error\" || output.stopReason === \"aborted\") {\n\t\t\t\tthrow new Error(output.errorMessage || \"An unknown error occurred\");\n\t\t\t}\n\n\t\t\t// A stream can settle without stopping every block, so finalize here too.\n\t\t\tfor (const block of output.content) finalizeStreamingBlock(block as Block);\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\tfinalizeStreamingBlock(block as Block);\n\t\t\t}\n\t\t\toutput.stopReason = options.signal?.aborted ? \"aborted\" : \"error\";\n\t\t\toutput.errorMetadata = normalizeProviderError(error).errorMetadata;\n\t\t\toutput.errorMessage = formatBedrockError(error);\n\t\t\tstream.push({ type: \"error\", reason: output.stopReason, error: output });\n\t\t\tstream.end();\n\t\t}\n\t})();\n\n\treturn stream;\n};\n\n/**\n * Human-readable prefixes for Bedrock SDK exception names.\n * The downstream retry logic in agent-session matches patterns like\n * `server.?error` and `service.?unavailable`, so we preserve the legacy\n * prefix format rather than using the raw SDK exception name.\n */\nconst BEDROCK_ERROR_PREFIXES: Record<string, string> = {\n\tInternalServerException: \"Internal server error\",\n\tModelStreamErrorException: \"Model stream error\",\n\tValidationException: \"Validation error\",\n\tThrottlingException: \"Throttling error\",\n\tServiceUnavailableException: \"Service unavailable\",\n};\n\n/**\n * Some models reject the account/profile's configured Bedrock data retention mode\n * (e.g. \"data retention mode 'default' is not available for this model\"). Point\n * users at the AWS docs explaining how to configure a supported mode.\n */\nconst BEDROCK_DATA_RETENTION_DOCS_URL = \"https://docs.aws.amazon.com/bedrock/latest/userguide/data-retention.html\";\n\n/**\n * Format a Bedrock error with a human-readable prefix.\n * AWS SDK exceptions (both from `client.send()` and from stream event items)\n * extend BedrockRuntimeServiceException. We map the `.name` to a stable\n * human-readable prefix so downstream consumers (retry logic, context-overflow\n * detection) can distinguish error categories via simple string matching.\n */\nfunction formatBedrockError(error: unknown): string {\n\tconst normalized = normalizeProviderError(error);\n\tconst message =\n\t\t!normalized.messageCarriesBody && normalized.status !== undefined && normalized.body !== undefined\n\t\t\t? `${normalized.status}: ${normalized.body}`\n\t\t\t: normalized.message;\n\tconst dataRetentionHint = /data retention mode/i.test(message)\n\t\t? ` See ${BEDROCK_DATA_RETENTION_DOCS_URL} for supported data retention modes.`\n\t\t: \"\";\n\tif (error instanceof BedrockRuntimeServiceException) {\n\t\tconst prefix = BEDROCK_ERROR_PREFIXES[error.name] ?? error.name;\n\t\treturn `${prefix}: ${message}${dataRetentionHint}`;\n\t}\n\treturn `${message}${dataRetentionHint}`;\n}\n\nfunction isSmithyHttpResponse(response: unknown): response is HttpResponse {\n\tif (!response || typeof response !== \"object\") return false;\n\tconst candidate = response as Partial<HttpResponse>;\n\treturn typeof candidate.statusCode === \"number\" && !!candidate.headers && typeof candidate.headers === \"object\";\n}\n\nfunction toProviderResponse(response: unknown): ProviderResponse | undefined {\n\tif (!isSmithyHttpResponse(response)) return undefined;\n\treturn { status: response.statusCode, headers: { ...response.headers } };\n}\n\n/**\n * Bedrock's modeled `$metadata` only preserves selected HTTP metadata (for example\n * requestId), so custom gateway headers are otherwise lost before callers see\n * `onResponse`. Capture the raw Smithy HTTP response at the deserialize step,\n * after the SDK receives the response but before the event stream is consumed.\n */\nfunction addResponseHeadersMiddleware(\n\tclient: BedrockRuntimeClient,\n\tonResponse: NonNullable<BedrockOptions[\"onResponse\"]>,\n\tmodel: Model<\"bedrock-converse-stream\">,\n\tonObserved: () => void,\n): void {\n\tconst middleware: DeserializeMiddleware<object, MetadataBearer> = (next) => async (args) => {\n\t\tconst result = await next(args);\n\t\tconst providerResponse = toProviderResponse(result.response);\n\t\tif (providerResponse) {\n\t\t\tonObserved();\n\t\t\tawait onResponse(providerResponse, model);\n\t\t}\n\t\treturn result;\n\t};\n\tclient.middlewareStack.add(middleware, { step: \"deserialize\", name: \"merc-ai-response-headers\" });\n}\n\nexport const streamSimpleBedrock: StreamFunction<\"bedrock-converse-stream\", SimpleStreamOptions> = (\n\tmodel: Model<\"bedrock-converse-stream\">,\n\tcontext: Context,\n\toptions?: SimpleStreamOptions,\n): AssistantMessageEventStream => {\n\tconst base = buildBaseOptions(model, context, options, undefined);\n\tif (!options?.reasoning) {\n\t\treturn streamBedrock(model, context, { ...base, reasoning: undefined } satisfies BedrockOptions);\n\t}\n\n\tif (isAnthropicClaudeModel(model)) {\n\t\tif (supportsAdaptiveThinking(model.id, model.name)) {\n\t\t\treturn streamBedrock(model, context, {\n\t\t\t\t...base,\n\t\t\t\treasoning: options.reasoning,\n\t\t\t\tthinkingBudgets: options.thinkingBudgets,\n\t\t\t} satisfies BedrockOptions);\n\t\t}\n\n\t\t// Undefined means the caller did not request an output cap; let the helper use the model cap.\n\t\t// Do not coerce to 0 here, or the thinking budget would become the entire maxTokens value.\n\t\tconst adjusted = adjustMaxTokensForThinking(\n\t\t\tbase.maxTokens,\n\t\t\tmodel.maxTokens,\n\t\t\toptions.reasoning,\n\t\t\toptions.thinkingBudgets,\n\t\t);\n\n\t\tconst maxTokens = clampMaxTokensToContext(model, context, adjusted.maxTokens);\n\n\t\treturn streamBedrock(model, context, {\n\t\t\t...base,\n\t\t\tmaxTokens,\n\t\t\treasoning: options.reasoning,\n\t\t\tthinkingBudgets: {\n\t\t\t\t...(options.thinkingBudgets || {}),\n\t\t\t\t[clampReasoning(options.reasoning)!]: Math.min(adjusted.thinkingBudget, Math.max(0, maxTokens - 1024)),\n\t\t\t},\n\t\t} satisfies BedrockOptions);\n\t}\n\n\treturn streamBedrock(model, context, {\n\t\t...base,\n\t\treasoning: options.reasoning,\n\t\tthinkingBudgets: options.thinkingBudgets,\n\t} satisfies BedrockOptions);\n};\n\nfunction handleContentBlockStart(\n\tevent: ContentBlockStartEvent,\n\tblocks: Block[],\n\toutput: AssistantMessage,\n\tstream: AssistantMessageEventStream,\n): void {\n\tconst index = event.contentBlockIndex!;\n\tconst start = event.start;\n\n\tif (start?.toolUse) {\n\t\tconst block: Block = {\n\t\t\ttype: \"toolCall\",\n\t\t\tid: start.toolUse.toolUseId || \"\",\n\t\t\tname: start.toolUse.name || \"\",\n\t\t\targuments: {},\n\t\t\tpartialJson: \"\",\n\t\t\tindex,\n\t\t};\n\t\toutput.content.push(block);\n\t\tstream.push({ type: \"toolcall_start\", contentIndex: blocks.length - 1, partial: output });\n\t}\n}\n\nfunction handleContentBlockDelta(\n\tevent: ContentBlockDeltaEvent,\n\tblocks: Block[],\n\toutput: AssistantMessage,\n\tstream: AssistantMessageEventStream,\n): void {\n\tconst contentBlockIndex = event.contentBlockIndex!;\n\tconst delta = event.delta;\n\tlet index = blocks.findIndex((b) => b.index === contentBlockIndex);\n\tlet block = blocks[index];\n\n\tif (delta?.text !== undefined) {\n\t\t// If no text block exists yet, create one, as `handleContentBlockStart` is not sent for text blocks\n\t\tif (!block) {\n\t\t\tconst newBlock: Block = { type: \"text\", text: \"\", index: contentBlockIndex };\n\t\t\toutput.content.push(newBlock);\n\t\t\tindex = blocks.length - 1;\n\t\t\tblock = blocks[index];\n\t\t\tstream.push({ type: \"text_start\", contentIndex: index, partial: output });\n\t\t}\n\t\tif (block.type === \"text\") {\n\t\t\tblock.text += delta.text;\n\t\t\tstream.push({ type: \"text_delta\", contentIndex: index, delta: delta.text, partial: output });\n\t\t}\n\t} else if (delta?.toolUse && block?.type === \"toolCall\") {\n\t\tblock.partialJson = (block.partialJson || \"\") + (delta.toolUse.input || \"\");\n\t\tblock.arguments = parseStreamingJson(block.partialJson);\n\t\tstream.push({ type: \"toolcall_delta\", contentIndex: index, delta: delta.toolUse.input || \"\", partial: output });\n\t} else if (delta?.reasoningContent) {\n\t\tlet thinkingBlock = block;\n\t\tlet thinkingIndex = index;\n\n\t\tif (!thinkingBlock) {\n\t\t\tconst newBlock: Block = { type: \"thinking\", thinking: \"\", thinkingSignature: \"\", index: contentBlockIndex };\n\t\t\toutput.content.push(newBlock);\n\t\t\tthinkingIndex = blocks.length - 1;\n\t\t\tthinkingBlock = blocks[thinkingIndex];\n\t\t\tstream.push({ type: \"thinking_start\", contentIndex: thinkingIndex, partial: output });\n\t\t}\n\n\t\tif (thinkingBlock?.type === \"thinking\") {\n\t\t\tif (delta.reasoningContent.text) {\n\t\t\t\tthinkingBlock.thinking += delta.reasoningContent.text;\n\t\t\t\tstream.push({\n\t\t\t\t\ttype: \"thinking_delta\",\n\t\t\t\t\tcontentIndex: thinkingIndex,\n\t\t\t\t\tdelta: delta.reasoningContent.text,\n\t\t\t\t\tpartial: output,\n\t\t\t\t});\n\t\t\t}\n\t\t\t// `thinkingSignature` holds either an Anthropic signature or an opaque redacted\n\t\t\t// payload, never both: mixing them would corrupt whichever arrived first.\n\t\t\tif (delta.reasoningContent.signature && !thinkingBlock.redacted) {\n\t\t\t\tthinkingBlock.thinkingSignature =\n\t\t\t\t\t(thinkingBlock.thinkingSignature || \"\") + delta.reasoningContent.signature;\n\t\t\t}\n\t\t\tif (delta.reasoningContent.redactedContent?.length) {\n\t\t\t\t// Encrypted reasoning from non-Anthropic models on Bedrock (e.g. OpenAI GPT-5.6).\n\t\t\t\t// The payload is opaque, so keep it verbatim in `thinkingSignature` the way the\n\t\t\t\t// Anthropic path stores redacted thinking, and replay it on the next turn.\n\t\t\t\tif (!thinkingBlock.redacted) {\n\t\t\t\t\tthinkingBlock.redacted = true;\n\t\t\t\t\tthinkingBlock.thinkingSignature = \"\";\n\t\t\t\t\tthinkingBlock.thinking += REDACTED_THINKING_PLACEHOLDER;\n\t\t\t\t\tstream.push({\n\t\t\t\t\t\ttype: \"thinking_delta\",\n\t\t\t\t\t\tcontentIndex: thinkingIndex,\n\t\t\t\t\t\tdelta: REDACTED_THINKING_PLACEHOLDER,\n\t\t\t\t\t\tpartial: output,\n\t\t\t\t\t});\n\t\t\t\t}\n\t\t\t\tthinkingBlock.redactedChunks ??= [];\n\t\t\t\tthinkingBlock.redactedChunks.push(delta.reasoningContent.redactedContent);\n\t\t\t}\n\t\t}\n\t}\n}\n\n/**\n * Encodes buffered encrypted reasoning into `thinkingSignature` and drops the scratch\n * buffer, which must never reach a persisted message: `Uint8Array` serializes to an\n * index-keyed object roughly ten times the size of the base64 payload.\n */\nfunction flushRedactedContent(block: Block): void {\n\tif (block.type !== \"thinking\" || !block.redactedChunks) return;\n\tblock.thinkingSignature = bytesToBase64(block.redactedChunks);\n\tdelete block.redactedChunks;\n}\n\n/**\n * Strips every streaming scratch field. Runs from the terminal paths as well as\n * `contentBlockStop`, because a stream can settle without stopping each block.\n */\nfunction finalizeStreamingBlock(block: Block): void {\n\tif (block.type === \"toolCall\" && block.partialJson !== undefined) {\n\t\tfinalizeToolCallArguments(block, block.partialJson);\n\t}\n\tdelete block.index;\n\t// partialJson is only a streaming scratch buffer; never persist it.\n\tdelete block.partialJson;\n\tflushRedactedContent(block);\n}\n\nfunction handleMetadata(\n\tevent: ConverseStreamMetadataEvent,\n\tmodel: Model<\"bedrock-converse-stream\">,\n\toutput: AssistantMessage,\n): void {\n\tif (event.usage) {\n\t\toutput.usage.input = event.usage.inputTokens || 0;\n\t\toutput.usage.output = event.usage.outputTokens || 0;\n\t\toutput.usage.cacheRead = event.usage.cacheReadInputTokens || 0;\n\t\toutput.usage.cacheWrite = event.usage.cacheWriteInputTokens || 0;\n\t\toutput.usage.cacheWrite1h = event.usage.cacheDetails?.reduce(\n\t\t\t(total, detail) => total + (detail.ttl === CacheTTL.ONE_HOUR ? (detail.inputTokens ?? 0) : 0),\n\t\t\t0,\n\t\t);\n\t\toutput.usage.totalTokens = event.usage.totalTokens || output.usage.input + output.usage.output;\n\t\tcalculateCost(model, output.usage);\n\t}\n}\n\nfunction handleContentBlockStop(\n\tevent: ContentBlockStopEvent,\n\tblocks: Block[],\n\toutput: AssistantMessage,\n\tstream: AssistantMessageEventStream,\n): void {\n\tconst index = blocks.findIndex((b) => b.index === event.contentBlockIndex);\n\tconst block = blocks[index];\n\tif (!block) return;\n\tdelete (block as Block).index;\n\n\tswitch (block.type) {\n\t\tcase \"text\":\n\t\t\tstream.push({ type: \"text_end\", contentIndex: index, content: block.text, partial: output });\n\t\t\tbreak;\n\t\tcase \"thinking\":\n\t\t\tflushRedactedContent(block);\n\t\t\tstream.push({ type: \"thinking_end\", contentIndex: index, content: block.thinking, partial: output });\n\t\t\tbreak;\n\t\tcase \"toolCall\":\n\t\t\tfinalizeToolCallArguments(block, block.partialJson);\n\t\t\t// Finalize in-place and strip the scratch buffer so replay only\n\t\t\t// carries parsed arguments.\n\t\t\tdelete (block as Block).partialJson;\n\t\t\tstream.push({ type: \"toolcall_end\", contentIndex: index, toolCall: block, partial: output });\n\t\t\tbreak;\n\t}\n}\n\n/**\n * Check if the model supports adaptive thinking (Opus 4.6+, Sonnet 4.6).\n * Checks both model ID and model name to support application inference profiles\n * whose ARNs don't contain the model name.\n */\nfunction getModelMatchCandidates(modelId: string, modelName?: string): string[] {\n\tconst values = modelName ? [modelId, modelName] : [modelId];\n\treturn values.flatMap((value) => {\n\t\tconst lower = value.toLowerCase();\n\t\treturn [lower, lower.replace(/[\\s_.:]+/g, \"-\")];\n\t});\n}\n\nfunction supportsAdaptiveThinking(modelId: string, modelName?: string): boolean {\n\tconst candidates = getModelMatchCandidates(modelId, modelName);\n\treturn candidates.some(\n\t\t(s) =>\n\t\t\ts.includes(\"opus-4-6\") ||\n\t\t\ts.includes(\"opus-4-7\") ||\n\t\t\ts.includes(\"opus-4-8\") ||\n\t\t\ts.includes(\"sonnet-4-6\") ||\n\t\t\ts.includes(\"sonnet-5\") ||\n\t\t\ts.includes(\"fable-5\"),\n\t);\n}\n\nfunction supportsNativeXhighEffort(model: Model<\"bedrock-converse-stream\">): boolean {\n\tconst candidates = getModelMatchCandidates(model.id, model.name);\n\treturn candidates.some(\n\t\t(s) => s.includes(\"opus-4-7\") || s.includes(\"opus-4-8\") || s.includes(\"sonnet-5\") || s.includes(\"fable-5\"),\n\t);\n}\n\nfunction mapThinkingLevelToEffort(\n\tmodel: Model<\"bedrock-converse-stream\">,\n\tlevel: SimpleStreamOptions[\"reasoning\"],\n): \"low\" | \"medium\" | \"high\" | \"xhigh\" | \"max\" {\n\tif (level === \"xhigh\" && supportsNativeXhighEffort(model)) return \"xhigh\";\n\tif (level === \"max\" && supportsAdaptiveThinking(model.id, model.name)) return \"max\";\n\n\tconst mapped = level ? model.thinkingLevelMap?.[level] : undefined;\n\tif (typeof mapped === \"string\") return mapped as \"low\" | \"medium\" | \"high\" | \"xhigh\" | \"max\";\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\tdefault:\n\t\t\treturn \"high\";\n\t}\n}\n\n/**\n * Resolve cache retention preference.\n * Defaults to \"short\" and uses PI_CACHE_RETENTION for backward compatibility.\n */\nfunction resolveCacheRetention(cacheRetention?: CacheRetention, env?: ProviderEnv): CacheRetention {\n\tif (cacheRetention) {\n\t\treturn cacheRetention;\n\t}\n\tif (getProviderEnvValue(\"PI_CACHE_RETENTION\", env) === \"long\") {\n\t\treturn \"long\";\n\t}\n\treturn \"short\";\n}\n\n/**\n * Check if the model is an Anthropic Claude model on Bedrock.\n * Checks both model ID and model name to support application inference profiles\n * whose ARNs don't contain the model name.\n */\nfunction isAnthropicClaudeModel(model: Model<\"bedrock-converse-stream\">): boolean {\n\tconst id = model.id.toLowerCase();\n\tconst name = model.name?.toLowerCase() ?? \"\";\n\treturn (\n\t\tid.includes(\"anthropic.claude\") ||\n\t\tid.includes(\"anthropic/claude\") ||\n\t\tname.includes(\"anthropic.claude\") ||\n\t\tname.includes(\"anthropic/claude\") ||\n\t\tname.includes(\"claude\")\n\t);\n}\n\n/**\n * Check if the model supports prompt caching.\n * Supported: Claude 3.5 Haiku, Claude 3.7 Sonnet, Claude 4.x, and Claude 5 models\n *\n * For base models and system-defined inference profiles the model ID / ARN\n * contains the model name, so we can decide locally.\n *\n * For application inference profiles (whose ARNs don't contain the model name),\n * also checks model.name which is user-controlled via models.json or registerProvider.\n * As a last resort, set AWS_BEDROCK_FORCE_CACHE=1 to enable cache points.\n * Amazon Nova models have automatic caching and don't need explicit cache points.\n */\nfunction supportsPromptCaching(model: Model<\"bedrock-converse-stream\">, env?: ProviderEnv): boolean {\n\tconst candidates = getModelMatchCandidates(model.id, model.name);\n\n\tconst hasClaudeRef = candidates.some((s) => s.includes(\"claude\"));\n\tif (!hasClaudeRef) {\n\t\t// Application inference profiles don't contain the model name in the ARN.\n\t\t// Allow users to force cache points via environment variable.\n\t\tif (getProviderEnvValue(\"AWS_BEDROCK_FORCE_CACHE\", env) === \"1\") return true;\n\t\treturn false;\n\t}\n\t// Claude 4.x models (opus-4, sonnet-4, haiku-4)\n\tif (candidates.some((s) => s.includes(\"-4-\"))) return true;\n\t// Claude 3.7 Sonnet\n\tif (candidates.some((s) => s.includes(\"claude-3-7-sonnet\"))) return true;\n\t// Claude 3.5 Haiku\n\tif (candidates.some((s) => s.includes(\"claude-3-5-haiku\"))) return true;\n\t// Claude 5 models (Fable 5, Sonnet 5)\n\tif (candidates.some((s) => s.includes(\"fable-5\") || s.includes(\"sonnet-5\"))) return true;\n\treturn false;\n}\n\n/**\n * Check if the model supports thinking signatures in reasoningContent.\n * Only Anthropic Claude models support the signature field.\n * Other models (OpenAI, Qwen, Minimax, Moonshot, etc.) reject it with:\n * \"This model doesn't support the reasoningContent.reasoningText.signature field\"\n *\n * Checks both model ID and model name to support application inference profiles.\n */\nfunction supportsThinkingSignature(model: Model<\"bedrock-converse-stream\">): boolean {\n\treturn isAnthropicClaudeModel(model);\n}\n\nfunction buildSystemPrompt(\n\tsystemPrompt: string | undefined,\n\tmodel: Model<\"bedrock-converse-stream\">,\n\tcacheRetention: CacheRetention,\n\tenv?: ProviderEnv,\n): SystemContentBlock[] | undefined {\n\tif (!systemPrompt) return undefined;\n\n\tconst blocks: SystemContentBlock[] = [{ text: sanitizeSurrogates(systemPrompt) }];\n\n\t// Add cache point for supported Claude models when caching is enabled\n\tif (cacheRetention !== \"none\" && supportsPromptCaching(model, env)) {\n\t\tblocks.push({\n\t\t\tcachePoint: { type: CachePointType.DEFAULT, ...(cacheRetention === \"long\" ? { ttl: CacheTTL.ONE_HOUR } : {}) },\n\t\t});\n\t}\n\n\treturn blocks;\n}\n\n/** Bedrock rejects empty property names in replayed tool input; drop them without mutating the stored arguments. */\nfunction sanitizeBedrockDocument(value: DocumentType): DocumentType {\n\tif (Array.isArray(value)) {\n\t\treturn value.map(sanitizeBedrockDocument);\n\t}\n\tif (value !== null && typeof value === \"object\") {\n\t\treturn Object.fromEntries(\n\t\t\tObject.entries(value)\n\t\t\t\t.filter(([key]) => key.length > 0)\n\t\t\t\t.map(([key, nestedValue]) => [key, sanitizeBedrockDocument(nestedValue)]),\n\t\t);\n\t}\n\treturn value;\n}\n\nfunction convertMessages(\n\tcontext: Context,\n\tmodel: Model<\"bedrock-converse-stream\">,\n\tcacheRetention: CacheRetention,\n\tenv?: ProviderEnv,\n): Message[] {\n\tconst result: Message[] = [];\n\tconst transformedMessages = transformMessages(context.messages, model, createProviderToolCallIdNormalizer(model));\n\n\tfor (let i = 0; i < transformedMessages.length; i++) {\n\t\tconst m = transformedMessages[i];\n\n\t\tswitch (m.role) {\n\t\t\tcase \"user\": {\n\t\t\t\tconst content: ContentBlock[] = [];\n\t\t\t\tif (typeof m.content === \"string\") {\n\t\t\t\t\tcontent.push({ text: sanitizeSurrogates(m.content) });\n\t\t\t\t} else {\n\t\t\t\t\tfor (const c of m.content) {\n\t\t\t\t\t\tswitch (c.type) {\n\t\t\t\t\t\t\tcase \"text\":\n\t\t\t\t\t\t\t\tcontent.push({ text: sanitizeSurrogates(c.text) });\n\t\t\t\t\t\t\t\tbreak;\n\t\t\t\t\t\t\tcase \"image\":\n\t\t\t\t\t\t\t\tcontent.push({ image: createImageBlock(c.mimeType, c.data) });\n\t\t\t\t\t\t\t\tbreak;\n\t\t\t\t\t\t\tdefault:\n\t\t\t\t\t\t\t\tcontinue;\n\t\t\t\t\t\t}\n\t\t\t\t\t}\n\t\t\t\t}\n\t\t\t\tif (content.length === 0) continue;\n\t\t\t\tresult.push({\n\t\t\t\t\trole: ConversationRole.USER,\n\t\t\t\t\tcontent,\n\t\t\t\t});\n\t\t\t\tbreak;\n\t\t\t}\n\t\t\tcase \"assistant\": {\n\t\t\t\t// Skip assistant messages with empty content (e.g., from aborted requests)\n\t\t\t\t// Bedrock rejects messages with empty content arrays\n\t\t\t\tif (m.content.length === 0) {\n\t\t\t\t\tcontinue;\n\t\t\t\t}\n\t\t\t\tconst contentBlocks: ContentBlock[] = [];\n\t\t\t\tfor (const c of m.content) {\n\t\t\t\t\tswitch (c.type) {\n\t\t\t\t\t\tcase \"text\":\n\t\t\t\t\t\t\t// Skip empty text blocks\n\t\t\t\t\t\t\tif (c.text.trim().length === 0) continue;\n\t\t\t\t\t\t\tcontentBlocks.push({ text: sanitizeSurrogates(c.text) });\n\t\t\t\t\t\t\tbreak;\n\t\t\t\t\t\tcase \"toolCall\":\n\t\t\t\t\t\t\tcontentBlocks.push({\n\t\t\t\t\t\t\t\ttoolUse: { toolUseId: c.id, name: c.name, input: sanitizeBedrockDocument(c.arguments) },\n\t\t\t\t\t\t\t});\n\t\t\t\t\t\t\tbreak;\n\t\t\t\t\t\tcase \"thinking\":\n\t\t\t\t\t\t\t// Encrypted reasoning is opaque: replay the stored payload as the\n\t\t\t\t\t\t\t// `redactedContent` member instead of lowering it to reasoning text.\n\t\t\t\t\t\t\tif (c.redacted) {\n\t\t\t\t\t\t\t\tconst redactedContent = decodeRedactedContent(c.thinkingSignature);\n\t\t\t\t\t\t\t\tif (redactedContent?.length) {\n\t\t\t\t\t\t\t\t\tcontentBlocks.push({ reasoningContent: { redactedContent } });\n\t\t\t\t\t\t\t\t}\n\t\t\t\t\t\t\t\tcontinue;\n\t\t\t\t\t\t\t}\n\t\t\t\t\t\t\t// Skip empty thinking blocks\n\t\t\t\t\t\t\tif (c.thinking.trim().length === 0) continue;\n\t\t\t\t\t\t\t// Only Anthropic models support the signature field in reasoningText.\n\t\t\t\t\t\t\t// For other models, we omit the signature to avoid errors like:\n\t\t\t\t\t\t\t// \"This model doesn't support the reasoningContent.reasoningText.signature field\"\n\t\t\t\t\t\t\tif (supportsThinkingSignature(model)) {\n\t\t\t\t\t\t\t\t// Signatures arrive after thinking deltas. If a partial or externally\n\t\t\t\t\t\t\t\t// persisted message lacks a signature, Bedrock rejects the replayed\n\t\t\t\t\t\t\t\t// reasoning block. Fall back to plain text, matching Anthropic.\n\t\t\t\t\t\t\t\tif (!c.thinkingSignature || c.thinkingSignature.trim().length === 0) {\n\t\t\t\t\t\t\t\t\tcontentBlocks.push({ text: sanitizeSurrogates(c.thinking) });\n\t\t\t\t\t\t\t\t} else {\n\t\t\t\t\t\t\t\t\tcontentBlocks.push({\n\t\t\t\t\t\t\t\t\t\treasoningContent: {\n\t\t\t\t\t\t\t\t\t\t\treasoningText: {\n\t\t\t\t\t\t\t\t\t\t\t\ttext: sanitizeSurrogates(c.thinking),\n\t\t\t\t\t\t\t\t\t\t\t\tsignature: c.thinkingSignature,\n\t\t\t\t\t\t\t\t\t\t\t},\n\t\t\t\t\t\t\t\t\t\t},\n\t\t\t\t\t\t\t\t\t});\n\t\t\t\t\t\t\t\t}\n\t\t\t\t\t\t\t} else {\n\t\t\t\t\t\t\t\tcontentBlocks.push({\n\t\t\t\t\t\t\t\t\treasoningContent: {\n\t\t\t\t\t\t\t\t\t\treasoningText: { text: sanitizeSurrogates(c.thinking) },\n\t\t\t\t\t\t\t\t\t},\n\t\t\t\t\t\t\t\t});\n\t\t\t\t\t\t\t}\n\t\t\t\t\t\t\tbreak;\n\t\t\t\t\t\tdefault:\n\t\t\t\t\t\t\tcontinue;\n\t\t\t\t\t}\n\t\t\t\t}\n\t\t\t\t// Skip if all content blocks were filtered out\n\t\t\t\tif (contentBlocks.length === 0) {\n\t\t\t\t\tcontinue;\n\t\t\t\t}\n\t\t\t\tresult.push({\n\t\t\t\t\trole: ConversationRole.ASSISTANT,\n\t\t\t\t\tcontent: contentBlocks,\n\t\t\t\t});\n\t\t\t\tbreak;\n\t\t\t}\n\t\t\tcase \"toolResult\": {\n\t\t\t\t// Collect all consecutive toolResult messages into a single user message\n\t\t\t\t// Bedrock requires all tool results to be in one message\n\t\t\t\tconst toolResults: ContentBlock.ToolResultMember[] = [];\n\n\t\t\t\t// Add current tool result with all content blocks combined\n\t\t\t\ttoolResults.push({\n\t\t\t\t\ttoolResult: {\n\t\t\t\t\t\ttoolUseId: m.toolCallId,\n\t\t\t\t\t\tcontent: m.content.map((c) =>\n\t\t\t\t\t\t\tc.type === \"image\"\n\t\t\t\t\t\t\t\t? { image: createImageBlock(c.mimeType, c.data) }\n\t\t\t\t\t\t\t\t: { text: sanitizeSurrogates(c.text) },\n\t\t\t\t\t\t),\n\t\t\t\t\t\tstatus: m.isError ? ToolResultStatus.ERROR : ToolResultStatus.SUCCESS,\n\t\t\t\t\t},\n\t\t\t\t});\n\n\t\t\t\t// Look ahead for consecutive toolResult messages\n\t\t\t\tlet j = i + 1;\n\t\t\t\twhile (j < transformedMessages.length && transformedMessages[j].role === \"toolResult\") {\n\t\t\t\t\tconst nextMsg = transformedMessages[j] as ToolResultMessage;\n\t\t\t\t\ttoolResults.push({\n\t\t\t\t\t\ttoolResult: {\n\t\t\t\t\t\t\ttoolUseId: nextMsg.toolCallId,\n\t\t\t\t\t\t\tcontent: nextMsg.content.map((c) =>\n\t\t\t\t\t\t\t\tc.type === \"image\"\n\t\t\t\t\t\t\t\t\t? { image: createImageBlock(c.mimeType, c.data) }\n\t\t\t\t\t\t\t\t\t: { text: sanitizeSurrogates(c.text) },\n\t\t\t\t\t\t\t),\n\t\t\t\t\t\t\tstatus: nextMsg.isError ? ToolResultStatus.ERROR : ToolResultStatus.SUCCESS,\n\t\t\t\t\t\t},\n\t\t\t\t\t});\n\t\t\t\t\tj++;\n\t\t\t\t}\n\n\t\t\t\t// Skip the messages we've already processed\n\t\t\t\ti = j - 1;\n\n\t\t\t\tresult.push({\n\t\t\t\t\trole: ConversationRole.USER,\n\t\t\t\t\tcontent: toolResults,\n\t\t\t\t});\n\t\t\t\tbreak;\n\t\t\t}\n\t\t\tdefault:\n\t\t\t\tcontinue;\n\t\t}\n\t}\n\n\t// Add cache point to the last user message for supported Claude models when caching is enabled\n\tif (cacheRetention !== \"none\" && supportsPromptCaching(model, env) && result.length > 0) {\n\t\tconst lastMessage = result[result.length - 1];\n\t\tif (lastMessage.role === ConversationRole.USER && lastMessage.content) {\n\t\t\t(lastMessage.content as ContentBlock[]).push({\n\t\t\t\tcachePoint: {\n\t\t\t\t\ttype: CachePointType.DEFAULT,\n\t\t\t\t\t...(cacheRetention === \"long\" ? { ttl: CacheTTL.ONE_HOUR } : {}),\n\t\t\t\t},\n\t\t\t});\n\t\t}\n\t}\n\n\treturn result;\n}\n\nfunction convertToolConfig(\n\ttools: Tool[] | undefined,\n\ttoolChoice: BedrockOptions[\"toolChoice\"],\n): ToolConfiguration | undefined {\n\tif (!tools?.length || toolChoice === \"none\") return undefined;\n\n\tconst bedrockTools: BedrockTool[] = tools.map((tool) => ({\n\t\ttoolSpec: {\n\t\t\tname: tool.name,\n\t\t\tdescription: tool.description,\n\t\t\tinputSchema: { json: tool.parameters as unknown as DocumentType },\n\t\t},\n\t}));\n\n\tlet bedrockToolChoice: ToolChoice | undefined;\n\tswitch (toolChoice) {\n\t\tcase \"auto\":\n\t\t\tbedrockToolChoice = { auto: {} };\n\t\t\tbreak;\n\t\tcase \"any\":\n\t\t\tbedrockToolChoice = { any: {} };\n\t\t\tbreak;\n\t\tdefault:\n\t\t\tif (toolChoice?.type === \"tool\") {\n\t\t\t\tbedrockToolChoice = { tool: { name: toolChoice.name } };\n\t\t\t}\n\t}\n\n\treturn { tools: bedrockTools, toolChoice: bedrockToolChoice };\n}\n\nfunction mapStopReason(reason: string | undefined): { stopReason: StopReason; errorMessage?: string } {\n\tswitch (reason) {\n\t\tcase BedrockStopReason.END_TURN:\n\t\tcase BedrockStopReason.STOP_SEQUENCE:\n\t\t\treturn { stopReason: \"stop\" };\n\t\tcase BedrockStopReason.MAX_TOKENS:\n\t\tcase BedrockStopReason.MODEL_CONTEXT_WINDOW_EXCEEDED:\n\t\t\treturn { stopReason: \"length\" };\n\t\tcase BedrockStopReason.TOOL_USE:\n\t\t\treturn { stopReason: \"toolUse\" };\n\t\tdefault:\n\t\t\treturn reason\n\t\t\t\t? { stopReason: \"error\", errorMessage: `Bedrock stopped with: ${reason}` }\n\t\t\t\t: { stopReason: \"error\" };\n\t}\n}\n\nfunction getConfiguredBedrockRegion(options: BedrockOptions): string | undefined {\n\tif (typeof process === \"undefined\") {\n\t\treturn options.region;\n\t}\n\n\treturn (\n\t\toptions.region ||\n\t\tgetProviderEnvValue(\"AWS_REGION\", options.env) ||\n\t\tgetProviderEnvValue(\"AWS_DEFAULT_REGION\", options.env) ||\n\t\tundefined\n\t);\n}\n\nfunction hasConfiguredBedrockProfile(env?: ProviderEnv): boolean {\n\tif (typeof process === \"undefined\") {\n\t\treturn false;\n\t}\n\n\treturn Boolean(getProviderEnvValue(\"AWS_PROFILE\", env));\n}\n\nfunction getStandardBedrockEndpointRegion(baseUrl: string | undefined): string | undefined {\n\tif (!baseUrl) {\n\t\treturn undefined;\n\t}\n\n\ttry {\n\t\tconst { hostname } = new URL(baseUrl);\n\t\tconst match = hostname.toLowerCase().match(/^bedrock-runtime(?:-fips)?\\.([a-z0-9-]+)\\.amazonaws\\.com(?:\\.cn)?$/);\n\t\treturn match?.[1];\n\t} catch {\n\t\treturn undefined;\n\t}\n}\n\nfunction shouldUseExplicitBedrockEndpoint(\n\tbaseUrl: string,\n\tconfiguredRegion: string | undefined,\n\thasConfiguredProfile: boolean,\n): boolean {\n\tconst endpointRegion = getStandardBedrockEndpointRegion(baseUrl);\n\tif (!endpointRegion) {\n\t\treturn true;\n\t}\n\n\treturn !configuredRegion && !hasConfiguredProfile;\n}\n\nfunction isGovCloudBedrockTarget(model: Model<\"bedrock-converse-stream\">, options: BedrockOptions): boolean {\n\tconst region = getConfiguredBedrockRegion(options);\n\tif (region?.toLowerCase().startsWith(\"us-gov-\")) {\n\t\treturn true;\n\t}\n\n\tconst modelId = model.id.toLowerCase();\n\treturn modelId.startsWith(\"us-gov.\") || modelId.startsWith(\"arn:aws-us-gov:\");\n}\n\nfunction buildAdditionalModelRequestFields(\n\tmodel: Model<\"bedrock-converse-stream\">,\n\toptions: BedrockOptions,\n): Record<string, any> | undefined {\n\tif (!options.reasoning || !model.reasoning) {\n\t\treturn undefined;\n\t}\n\n\tif (isAnthropicClaudeModel(model)) {\n\t\t// GovCloud Bedrock currently rejects the Claude thinking.display field.\n\t\t// Omit it there until the GovCloud Converse schema catches up.\n\t\tconst display = isGovCloudBedrockTarget(model, options) ? undefined : (options.thinkingDisplay ?? \"summarized\");\n\t\tconst result: Record<string, any> = supportsAdaptiveThinking(model.id, model.name)\n\t\t\t? {\n\t\t\t\t\tthinking: { type: \"adaptive\", ...(display !== undefined ? { display } : {}) },\n\t\t\t\t\toutput_config: { effort: mapThinkingLevelToEffort(model, options.reasoning) },\n\t\t\t\t}\n\t\t\t: (() => {\n\t\t\t\t\tconst defaultBudgets: Record<ThinkingLevel, number> = {\n\t\t\t\t\t\tminimal: 1024,\n\t\t\t\t\t\tlow: 2048,\n\t\t\t\t\t\tmedium: 8192,\n\t\t\t\t\t\thigh: 16384,\n\t\t\t\t\t\txhigh: 16384, // Claude doesn't support xhigh, clamp to high\n\t\t\t\t\t\tmax: 16384, // Non-adaptive Claude models do not support max effort\n\t\t\t\t\t};\n\n\t\t\t\t\t// Custom budgets override defaults; extended effort levels use the high budget.\n\t\t\t\t\tconst level = options.reasoning === \"xhigh\" || options.reasoning === \"max\" ? \"high\" : options.reasoning;\n\t\t\t\t\tconst budget = options.thinkingBudgets?.[level] ?? defaultBudgets[level];\n\n\t\t\t\t\treturn {\n\t\t\t\t\t\tthinking: {\n\t\t\t\t\t\t\ttype: \"enabled\",\n\t\t\t\t\t\t\tbudget_tokens: budget,\n\t\t\t\t\t\t\t...(display !== undefined ? { display } : {}),\n\t\t\t\t\t\t},\n\t\t\t\t\t};\n\t\t\t\t})();\n\n\t\tif (!supportsAdaptiveThinking(model.id, model.name) && (options.interleavedThinking ?? true)) {\n\t\t\tresult.anthropic_beta = [\"interleaved-thinking-2025-05-14\"];\n\t\t}\n\n\t\treturn result;\n\t}\n\n\tconst candidates = getModelMatchCandidates(model.id, model.name);\n\n\tif (candidates.some((s) => s.includes(\"gpt-oss\"))) {\n\t\treturn { reasoning_effort: OPENAI_GPT_OSS_EFFORT[options.reasoning] };\n\t}\n\n\tif (candidates.some((s) => s.includes(\"gpt-\"))) {\n\t\tconst mapped = model.thinkingLevelMap?.[options.reasoning];\n\t\treturn {\n\t\t\treasoning: { effort: typeof mapped === \"string\" ? mapped : OPENAI_GPT_EFFORT[options.reasoning] },\n\t\t};\n\t}\n\n\treturn undefined;\n}\n\ntype OpenAIGptEffort = \"low\" | \"medium\" | \"high\" | \"xhigh\" | \"max\";\ntype OpenAIGptOssEffort = \"low\" | \"medium\" | \"high\";\n\n/** OpenAI GPT models (GPT-5.x, GPT-6) take a nested `reasoning.effort` and reject `minimal`. */\nconst OPENAI_GPT_EFFORT: Record<ThinkingLevel, OpenAIGptEffort> = {\n\tminimal: \"low\",\n\tlow: \"low\",\n\tmedium: \"medium\",\n\thigh: \"high\",\n\txhigh: \"xhigh\",\n\tmax: \"max\",\n};\n\n/** gpt-oss takes a flat `reasoning_effort` and only accepts low, medium and high. */\nconst OPENAI_GPT_OSS_EFFORT: Record<ThinkingLevel, OpenAIGptOssEffort> = {\n\tminimal: \"low\",\n\tlow: \"low\",\n\tmedium: \"medium\",\n\thigh: \"high\",\n\txhigh: \"high\",\n\tmax: \"high\",\n};\n\nfunction createImageBlock(mimeType: string, data: string) {\n\tlet format: ImageFormat;\n\tswitch (mimeType) {\n\t\tcase \"image/jpeg\":\n\t\tcase \"image/jpg\":\n\t\t\tformat = ImageFormat.JPEG;\n\t\t\tbreak;\n\t\tcase \"image/png\":\n\t\t\tformat = ImageFormat.PNG;\n\t\t\tbreak;\n\t\tcase \"image/gif\":\n\t\t\tformat = ImageFormat.GIF;\n\t\t\tbreak;\n\t\tcase \"image/webp\":\n\t\t\tformat = ImageFormat.WEBP;\n\t\t\tbreak;\n\t\tdefault:\n\t\t\tthrow new Error(`Unknown image type: ${mimeType}`);\n\t}\n\n\treturn { source: { bytes: base64ToBytes(data) }, format };\n}\n\nfunction base64ToBytes(data: string): Uint8Array {\n\tconst binaryString = atob(data);\n\tconst bytes = new Uint8Array(binaryString.length);\n\tfor (let i = 0; i < binaryString.length; i++) {\n\t\tbytes[i] = binaryString.charCodeAt(i);\n\t}\n\treturn bytes;\n}\n\n/**\n * Decodes a stored redacted payload. The AWS SDK hands the blob over as bytes, but a\n * persisted session carries it as base64. A hand-edited or externally produced session\n * can hold a signature that is not base64; drop that block instead of failing the\n * whole request.\n */\nfunction decodeRedactedContent(signature: string | undefined): Uint8Array | undefined {\n\tif (!signature) return undefined;\n\ttry {\n\t\treturn base64ToBytes(signature);\n\t} catch {\n\t\treturn undefined;\n\t}\n}\n\nfunction bytesToBase64(chunks: Uint8Array[]): string {\n\t// Encrypted reasoning runs to tens of KB, so build the binary string in slices\n\t// rather than one concatenation per byte. The window stays under the engine's\n\t// argument-count limit for spread calls.\n\tconst WINDOW = 0x8000;\n\tlet binary = \"\";\n\tfor (const chunk of chunks) {\n\t\tfor (let i = 0; i < chunk.length; i += WINDOW) {\n\t\t\tbinary += String.fromCharCode(...chunk.subarray(i, i + WINDOW));\n\t\t}\n\t}\n\treturn btoa(binary);\n}\n"]}

package/dist/providers/amazon-bedrock.js

@@ -2,11 +2,14 @@ import { BedrockRuntimeClient, BedrockRuntimeServiceException, StopReason as Bed
 import { calculateCost } from "../models.js";
 import { normalizeProviderError } from "../utils/error-body.js";
 import { AssistantMessageEventStream } from "../utils/event-stream.js";
-import { parseStreamingJson } from "../utils/json-parse.js";
+import { finalizeToolCallArguments, parseStreamingJson } from "../utils/json-parse.js";
 import { getProviderEnvValue } from "../utils/provider-env.js";
 import { sanitizeSurrogates } from "../utils/sanitize-unicode.js";
-import { adjustMaxTokensForThinking, buildBaseOptions, clampReasoning } from "./simple-options.js";
+import { createProviderToolCallIdNormalizer } from "../utils/tool-call-id.js";
+import { adjustMaxTokensForThinking, buildBaseOptions, clampMaxTokensToContext, clampReasoning, } from "./simple-options.js";
 import { transformMessages } from "./transform-messages.js";
+/** Matches the placeholder the Anthropic API path uses for redacted thinking. */
+const REDACTED_THINKING_PLACEHOLDER = "[Reasoning redacted]";
 function isAmbientAuthMarker(value) {
     return value === "<authenticated>";
 }
@@ -111,6 +114,12 @@ export const streamBedrock = (model, context, options = {}) => {
         }
         try {
             const client = new BedrockRuntimeClient(config);
+            let observedRawResponse = false;
+            if (options.onResponse) {
+                addResponseHeadersMiddleware(client, options.onResponse, model, () => {
+                    observedRawResponse = true;
+                });
+            }
             const cacheRetention = resolveCacheRetention(options.cacheRetention, options.env);
             const inferenceMaxTokens = options.maxTokens ?? (isAnthropicClaudeModel(model) ? model.maxTokens : undefined);
             let commandInput = {
@@ -131,7 +140,7 @@ export const streamBedrock = (model, context, options = {}) => {
             }
             const command = new ConverseStreamCommand(commandInput);
             const response = await client.send(command, { abortSignal: options.signal });
-            if (response.$metadata.httpStatusCode !== undefined) {
+            if (!observedRawResponse && response.$metadata.httpStatusCode !== undefined) {
                 const responseHeaders = {};
                 if (response.$metadata.requestId) {
                     responseHeaders["x-amzn-requestid"] = response.$metadata.requestId;
@@ -155,6 +164,7 @@ export const streamBedrock = (model, context, options = {}) => {
                     handleContentBlockStop(item.contentBlockStop, blocks, output, stream);
                 }
                 else if (item.messageStop) {
+                    output.rawStopReason = item.messageStop.stopReason;
                     const { stopReason, errorMessage } = mapStopReason(item.messageStop.stopReason);
                     output.stopReason = stopReason;
                     if (errorMessage) {
@@ -186,16 +196,18 @@ export const streamBedrock = (model, context, options = {}) => {
             if (output.stopReason === "error" || output.stopReason === "aborted") {
                 throw new Error(output.errorMessage || "An unknown error occurred");
             }
+            // A stream can settle without stopping every block, so finalize here too.
+            for (const block of output.content)
+                finalizeStreamingBlock(block);
             stream.push({ type: "done", reason: output.stopReason, message: output });
             stream.end();
         }
         catch (error) {
             for (const block of output.content) {
-                delete block.index;
-                // partialJson is only a streaming scratch buffer; never persist it.
-                delete block.partialJson;
+                finalizeStreamingBlock(block);
             }
             output.stopReason = options.signal?.aborted ? "aborted" : "error";
+            output.errorMetadata = normalizeProviderError(error).errorMetadata;
             output.errorMessage = formatBedrockError(error);
             stream.push({ type: "error", reason: output.stopReason, error: output });
             stream.end();
@@ -243,8 +255,37 @@ function formatBedrockError(error) {
     }
     return `${message}${dataRetentionHint}`;
 }
+function isSmithyHttpResponse(response) {
+    if (!response || typeof response !== "object")
+        return false;
+    const candidate = response;
+    return typeof candidate.statusCode === "number" && !!candidate.headers && typeof candidate.headers === "object";
+}
+function toProviderResponse(response) {
+    if (!isSmithyHttpResponse(response))
+        return undefined;
+    return { status: response.statusCode, headers: { ...response.headers } };
+}
+/**
+ * Bedrock's modeled `$metadata` only preserves selected HTTP metadata (for example
+ * requestId), so custom gateway headers are otherwise lost before callers see
+ * `onResponse`. Capture the raw Smithy HTTP response at the deserialize step,
+ * after the SDK receives the response but before the event stream is consumed.
+ */
+function addResponseHeadersMiddleware(client, onResponse, model, onObserved) {
+    const middleware = (next) => async (args) => {
+        const result = await next(args);
+        const providerResponse = toProviderResponse(result.response);
+        if (providerResponse) {
+            onObserved();
+            await onResponse(providerResponse, model);
+        }
+        return result;
+    };
+    client.middlewareStack.add(middleware, { step: "deserialize", name: "merc-ai-response-headers" });
+}
 export const streamSimpleBedrock = (model, context, options) => {
-    const base = buildBaseOptions(model, options, undefined);
+    const base = buildBaseOptions(model, context, options, undefined);
     if (!options?.reasoning) {
         return streamBedrock(model, context, { ...base, reasoning: undefined });
     }
@@ -256,14 +297,17 @@ export const streamSimpleBedrock = (model, context, options) => {
                 thinkingBudgets: options.thinkingBudgets,
             });
         }
-        const adjusted = adjustMaxTokensForThinking(base.maxTokens || 0, model.maxTokens, options.reasoning, options.thinkingBudgets);
+        // Undefined means the caller did not request an output cap; let the helper use the model cap.
+        // Do not coerce to 0 here, or the thinking budget would become the entire maxTokens value.
+        const adjusted = adjustMaxTokensForThinking(base.maxTokens, model.maxTokens, options.reasoning, options.thinkingBudgets);
+        const maxTokens = clampMaxTokensToContext(model, context, adjusted.maxTokens);
         return streamBedrock(model, context, {
             ...base,
-            maxTokens: adjusted.maxTokens,
+            maxTokens,
             reasoning: options.reasoning,
             thinkingBudgets: {
                 ...(options.thinkingBudgets || {}),
-                [clampReasoning(options.reasoning)]: adjusted.thinkingBudget,
+                [clampReasoning(options.reasoning)]: Math.min(adjusted.thinkingBudget, Math.max(0, maxTokens - 1024)),
             },
         });
     }
@@ -333,19 +377,64 @@ function handleContentBlockDelta(event, blocks, output, stream) {
                     partial: output,
                 });
             }
-            if (delta.reasoningContent.signature) {
+            // `thinkingSignature` holds either an Anthropic signature or an opaque redacted
+            // payload, never both: mixing them would corrupt whichever arrived first.
+            if (delta.reasoningContent.signature && !thinkingBlock.redacted) {
                 thinkingBlock.thinkingSignature =
                     (thinkingBlock.thinkingSignature || "") + delta.reasoningContent.signature;
             }
+            if (delta.reasoningContent.redactedContent?.length) {
+                // Encrypted reasoning from non-Anthropic models on Bedrock (e.g. OpenAI GPT-5.6).
+                // The payload is opaque, so keep it verbatim in `thinkingSignature` the way the
+                // Anthropic path stores redacted thinking, and replay it on the next turn.
+                if (!thinkingBlock.redacted) {
+                    thinkingBlock.redacted = true;
+                    thinkingBlock.thinkingSignature = "";
+                    thinkingBlock.thinking += REDACTED_THINKING_PLACEHOLDER;
+                    stream.push({
+                        type: "thinking_delta",
+                        contentIndex: thinkingIndex,
+                        delta: REDACTED_THINKING_PLACEHOLDER,
+                        partial: output,
+                    });
+                }
+                thinkingBlock.redactedChunks ??= [];
+                thinkingBlock.redactedChunks.push(delta.reasoningContent.redactedContent);
+            }
         }
     }
 }
+/**
+ * Encodes buffered encrypted reasoning into `thinkingSignature` and drops the scratch
+ * buffer, which must never reach a persisted message: `Uint8Array` serializes to an
+ * index-keyed object roughly ten times the size of the base64 payload.
+ */
+function flushRedactedContent(block) {
+    if (block.type !== "thinking" || !block.redactedChunks)
+        return;
+    block.thinkingSignature = bytesToBase64(block.redactedChunks);
+    delete block.redactedChunks;
+}
+/**
+ * Strips every streaming scratch field. Runs from the terminal paths as well as
+ * `contentBlockStop`, because a stream can settle without stopping each block.
+ */
+function finalizeStreamingBlock(block) {
+    if (block.type === "toolCall" && block.partialJson !== undefined) {
+        finalizeToolCallArguments(block, block.partialJson);
+    }
+    delete block.index;
+    // partialJson is only a streaming scratch buffer; never persist it.
+    delete block.partialJson;
+    flushRedactedContent(block);
+}
 function handleMetadata(event, model, output) {
     if (event.usage) {
         output.usage.input = event.usage.inputTokens || 0;
         output.usage.output = event.usage.outputTokens || 0;
         output.usage.cacheRead = event.usage.cacheReadInputTokens || 0;
         output.usage.cacheWrite = event.usage.cacheWriteInputTokens || 0;
+        output.usage.cacheWrite1h = event.usage.cacheDetails?.reduce((total, detail) => total + (detail.ttl === CacheTTL.ONE_HOUR ? (detail.inputTokens ?? 0) : 0), 0);
         output.usage.totalTokens = event.usage.totalTokens || output.usage.input + output.usage.output;
         calculateCost(model, output.usage);
     }
@@ -361,10 +450,11 @@ function handleContentBlockStop(event, blocks, output, stream) {
             stream.push({ type: "text_end", contentIndex: index, content: block.text, partial: output });
             break;
         case "thinking":
+            flushRedactedContent(block);
             stream.push({ type: "thinking_end", contentIndex: index, content: block.thinking, partial: output });
             break;
         case "toolCall":
-            block.arguments = parseStreamingJson(block.partialJson);
+            finalizeToolCallArguments(block, block.partialJson);
             // Finalize in-place and strip the scratch buffer so replay only
             // carries parsed arguments.
             delete block.partialJson;
@@ -503,13 +593,21 @@ function buildSystemPrompt(systemPrompt, model, cacheRetention, env) {
     }
     return blocks;
 }
-function normalizeToolCallId(id) {
-    const sanitized = id.replace(/[^a-zA-Z0-9_-]/g, "_");
-    return sanitized.length > 64 ? sanitized.slice(0, 64) : sanitized;
+/** Bedrock rejects empty property names in replayed tool input; drop them without mutating the stored arguments. */
+function sanitizeBedrockDocument(value) {
+    if (Array.isArray(value)) {
+        return value.map(sanitizeBedrockDocument);
+    }
+    if (value !== null && typeof value === "object") {
+        return Object.fromEntries(Object.entries(value)
+            .filter(([key]) => key.length > 0)
+            .map(([key, nestedValue]) => [key, sanitizeBedrockDocument(nestedValue)]));
+    }
+    return value;
 }
 function convertMessages(context, model, cacheRetention, env) {
     const result = [];
-    const transformedMessages = transformMessages(context.messages, model, normalizeToolCallId);
+    const transformedMessages = transformMessages(context.messages, model, createProviderToolCallIdNormalizer(model));
     for (let i = 0; i < transformedMessages.length; i++) {
         const m = transformedMessages[i];
         switch (m.role) {
@@ -557,10 +655,19 @@ function convertMessages(context, model, cacheRetention, env) {
                             break;
                         case "toolCall":
                             contentBlocks.push({
-                                toolUse: { toolUseId: c.id, name: c.name, input: c.arguments },
+                                toolUse: { toolUseId: c.id, name: c.name, input: sanitizeBedrockDocument(c.arguments) },
                             });
                             break;
                         case "thinking":
+                            // Encrypted reasoning is opaque: replay the stored payload as the
+                            // `redactedContent` member instead of lowering it to reasoning text.
+                            if (c.redacted) {
+                                const redactedContent = decodeRedactedContent(c.thinkingSignature);
+                                if (redactedContent?.length) {
+                                    contentBlocks.push({ reasoningContent: { redactedContent } });
+                                }
+                                continue;
+                            }
                             // Skip empty thinking blocks
                             if (c.thinking.trim().length === 0)
                                 continue;
@@ -698,7 +805,9 @@ function mapStopReason(reason) {
         case BedrockStopReason.TOOL_USE:
             return { stopReason: "toolUse" };
         default:
-            return reason ? { stopReason: "error", errorMessage: reason } : { stopReason: "error" };
+            return reason
+                ? { stopReason: "error", errorMessage: `Bedrock stopped with: ${reason}` }
+                : { stopReason: "error" };
     }
 }
 function getConfiguredBedrockRegion(options) {
@@ -782,8 +891,36 @@ function buildAdditionalModelRequestFields(model, options) {
         }
         return result;
     }
+    const candidates = getModelMatchCandidates(model.id, model.name);
+    if (candidates.some((s) => s.includes("gpt-oss"))) {
+        return { reasoning_effort: OPENAI_GPT_OSS_EFFORT[options.reasoning] };
+    }
+    if (candidates.some((s) => s.includes("gpt-"))) {
+        const mapped = model.thinkingLevelMap?.[options.reasoning];
+        return {
+            reasoning: { effort: typeof mapped === "string" ? mapped : OPENAI_GPT_EFFORT[options.reasoning] },
+        };
+    }
     return undefined;
 }
+/** OpenAI GPT models (GPT-5.x, GPT-6) take a nested `reasoning.effort` and reject `minimal`. */
+const OPENAI_GPT_EFFORT = {
+    minimal: "low",
+    low: "low",
+    medium: "medium",
+    high: "high",
+    xhigh: "xhigh",
+    max: "max",
+};
+/** gpt-oss takes a flat `reasoning_effort` and only accepts low, medium and high. */
+const OPENAI_GPT_OSS_EFFORT = {
+    minimal: "low",
+    low: "low",
+    medium: "medium",
+    high: "high",
+    xhigh: "high",
+    max: "high",
+};
 function createImageBlock(mimeType, data) {
     let format;
     switch (mimeType) {
@@ -803,11 +940,43 @@ function createImageBlock(mimeType, data) {
         default:
             throw new Error(`Unknown image type: ${mimeType}`);
     }
+    return { source: { bytes: base64ToBytes(data) }, format };
+}
+function base64ToBytes(data) {
     const binaryString = atob(data);
     const bytes = new Uint8Array(binaryString.length);
     for (let i = 0; i < binaryString.length; i++) {
         bytes[i] = binaryString.charCodeAt(i);
     }
-    return { source: { bytes }, format };
+    return bytes;
+}
+/**
+ * Decodes a stored redacted payload. The AWS SDK hands the blob over as bytes, but a
+ * persisted session carries it as base64. A hand-edited or externally produced session
+ * can hold a signature that is not base64; drop that block instead of failing the
+ * whole request.
+ */
+function decodeRedactedContent(signature) {
+    if (!signature)
+        return undefined;
+    try {
+        return base64ToBytes(signature);
+    }
+    catch {
+        return undefined;
+    }
+}
+function bytesToBase64(chunks) {
+    // Encrypted reasoning runs to tens of KB, so build the binary string in slices
+    // rather than one concatenation per byte. The window stays under the engine's
+    // argument-count limit for spread calls.
+    const WINDOW = 0x8000;
+    let binary = "";
+    for (const chunk of chunks) {
+        for (let i = 0; i < chunk.length; i += WINDOW) {
+            binary += String.fromCharCode(...chunk.subarray(i, i + WINDOW));
+        }
+    }
+    return btoa(binary);
 }
 //# sourceMappingURL=amazon-bedrock.js.map