/**
 * Text Chunking Utility
 *
 * Splits large texts into chunks suitable for embedding generation.
 * Handles token limits, word boundaries, and overlap for context preservation.
 */
/**
 * Estimate token count for text.
 *
 * Uses a simple heuristic: ~4 characters per token for English text.
 * This is approximate but works well for our use case.
 *
 * @param text - Text to estimate tokens for
 * @returns Estimated token count
 */
export declare function estimateTokens(text: string): number;
/**
 * Split text into chunks that fit within max tokens.
 *
 * Tries to break at word boundaries when possible.
 * Adds overlap between chunks to preserve context.
 *
 * @param text - Text to chunk
 * @param maxTokens - Maximum tokens per chunk (default: 8000 for OpenRouter)
 * @param overlapTokens - Number of tokens to overlap between chunks (default: 0)
 * @returns Array of text chunks
 */
export declare function chunkText(text: string, maxTokens?: number, overlapTokens?: number): string[];
/**
 * Chunk text specifically for embedding generation.
 *
 * Uses 8000 token limit (OpenRouter's limit for text-embedding-3-small).
 *
 * @param text - Text to chunk
 * @returns Array of text chunks suitable for embeddings
 */
export declare function chunkForEmbedding(text: string): string[];
//# sourceMappingURL=chunking.d.ts.map