dhaga.docs
API referenceClasses

AnthropicLLMClient

Defined in: llm/anthropic-client/index.ts:23

Split per the 150-line rule: request-building and response-parsing live in ./sync (extract/complete) and ./batch (the Message Batches API — CLAUDE.md's "Nightly/latency-insensitive jobs: Batch API" rule). This class just owns the SDK client and satisfies both gateway interfaces so callers keep depending on LLMClient / BatchLLMClient, never on this class directly (Dependency Inversion).

Implements

Constructors

Constructor

new AnthropicLLMClient(apiKey): AnthropicLLMClient;

Defined in: llm/anthropic-client/index.ts:26

Parameters

apiKey

string

Returns

AnthropicLLMClient

Methods

extract()

extract<T>(options): Promise<LLMResult<T>>;

Defined in: llm/anthropic-client/index.ts:30

Structured extraction — output is guaranteed to match the Zod schema.

Type Parameters

T

T

Parameters

options

ExtractOptions<T>

Returns

Promise<LLMResult<T>>

Implementation of

LLMClient.extract


complete()

complete(options): Promise<LLMResult<string>>;

Defined in: llm/anthropic-client/index.ts:34

Free-text completion (drafts, search answers).

Parameters

options

CompleteOptions

Returns

Promise<LLMResult<string>>

Implementation of

LLMClient.complete


submitExtractBatch()

submitExtractBatch<T>(items): Promise<string>;

Defined in: llm/anthropic-client/index.ts:38

Submit many extract-shaped requests as one batch. Returns immediately with a batch id — does not wait for completion (a batch can take anywhere from minutes up to 24 hours).

Type Parameters

T

T

Parameters

items

BatchExtractItem<T>[]

Returns

Promise<string>

Implementation of

BatchLLMClient.submitExtractBatch


isBatchDone()

isBatchDone(batchId): Promise<boolean>;

Defined in: llm/anthropic-client/index.ts:42

True once every request in the batch has succeeded, errored, expired, or been canceled — only then is getBatchResults safe to call.

Parameters

batchId

string

Returns

Promise<boolean>

Implementation of

BatchLLMClient.isBatchDone


getBatchResults()

getBatchResults<T>(batchId, schema): Promise<BatchExtractResult<T>[]>;

Defined in: llm/anthropic-client/index.ts:46

One entry per submitted item, in no particular order — match results to requests via id. Only valid once isBatchDone() is true.

Type Parameters

T

T

Parameters

batchId

string

schema

ZodType<T>

Returns

Promise<BatchExtractResult<T>[]>

Implementation of

BatchLLMClient.getBatchResults

On this page