import type { CrawlResult } from "../types.js";
/**
 * Extract all internal (same-origin) links from an HTML page.
 * Returns a deduplicated array of absolute URLs, excluding
 * fragments, mailto, tel, and javascript: links.
 */
export declare function extractInternalLinks(pageUrl: string): Promise<readonly string[]>;
/**
 * Filter links to only same-origin URLs, removing fragments,
 * non-http schemes, and normalizing paths.
 */
export declare function filterInternalLinks(links: readonly string[], origin: string): readonly string[];
/**
 * Normalize a URL: resolve relative, strip fragments, validate same-origin.
 * Returns null if the URL should be excluded.
 */
export declare function normalizeUrl(link: string, origin: string): string | null;
/**
 * Remove duplicate URLs while preserving order.
 */
export declare function deduplicateUrls(urls: readonly string[]): readonly string[];
/**
 * Clamp maxPages within valid bounds.
 */
export declare function clampMaxPages(maxPages: number | undefined): number;
/**
 * Crawl a website from a starting URL and run lightweight audits
 * (screenshot + accessibility + performance) on each discovered page.
 *
 * - Extracts internal links from the starting page
 * - Deduplicates and limits to maxPages
 * - Audits each page independently; failures on one page don't stop others
 * - Does NOT run full Lighthouse (too slow for multi-page crawl)
 */
export declare function crawlAndReview(startUrl: string, maxPages?: number): Promise<CrawlResult>;
/**
 * Format a CrawlResult into a human-readable summary report.
 */
export declare function formatCrawlReport(result: CrawlResult): string;
//# sourceMappingURL=crawl.d.ts.map