Interface FactExtractorOptions

interface FactExtractorOptions {
    llmInvoker: ((systemPrompt, userPrompt) => Promise<string>);
    cacheFingerprint: string;
    maxOutputTokens?: number;
}

Properties

llmInvoker cacheFingerprint maxOutputTokens?

Properties

llmInvoker

llmInvoker: ((systemPrompt, userPrompt) => Promise<string>)

Called with (systemPrompt, userPrompt). Must return the model's text response.

Type declaration

    • (systemPrompt, userPrompt): Promise<string>

    • Parameters

      • systemPrompt: string
      • userPrompt: string

      Returns Promise<string>

cacheFingerprint

cacheFingerprint: string

Fingerprint appended to the content hash for cache keying. Bump when the prompt or schema changes so prior-version extractions don't leak into new runs.

Optional maxOutputTokens

maxOutputTokens?: number

Hard cap on LLM output tokens. Default 1024.

Settings

Member Visibility

Theme

On This Page

llmInvokercacheFingerprintmaxOutputTokens