Skip to main content

Interface: ParseJsonOptions<_T>

Defined in: packages/agentos/src/nlp/ai_utilities/IUtilityAI.ts:164

Options for safe JSON parsing.

Type Parameters​

_T​

_T = any

Properties​

attemptFixWithLLM?​

optional attemptFixWithLLM: boolean

Defined in: packages/agentos/src/nlp/ai_utilities/IUtilityAI.ts:166

If true, attempts to use an LLM to fix or extract JSON if standard parsing fails.

llmModelIdForFix?​

optional llmModelIdForFix: string

Defined in: packages/agentos/src/nlp/ai_utilities/IUtilityAI.ts:168

Model ID to use for LLM-based fixing.

llmProviderIdForFix?​

optional llmProviderIdForFix: string

Defined in: packages/agentos/src/nlp/ai_utilities/IUtilityAI.ts:170

Provider ID for the LLM fixer.

maxRepairAttempts?​

optional maxRepairAttempts: number

Defined in: packages/agentos/src/nlp/ai_utilities/IUtilityAI.ts:177

Max repair attempts with LLM if schema validation fails.

targetSchema?​

optional targetSchema: JSONSchemaObject

Defined in: packages/agentos/src/nlp/ai_utilities/IUtilityAI.ts:175

Optional JSON schema to validate the parsed object against. If validation fails, the method may return null or attempt to fix again.