import type { RuleResult } from "../../types.js";
/**
 * Default AI crawler user-agents to check. Ordered by prevalence.
 */
export declare const DEFAULT_AI_CRAWLERS: readonly ["GPTBot", "ChatGPT-User", "ClaudeBot", "PerplexityBot", "Bytespider", "Google-Extended", "CCBot", "Applebot-Extended"];
/**
 * Parse robots.txt into a map of user-agent -> list of Disallow patterns.
 * User-agent keys are lowercased for case-insensitive lookup.
 * A blank Disallow value is treated as "allow all" and produces an empty array
 * (matches the robots spec: "Disallow: " with no value means no restrictions).
 */
export declare function parseRobotsByUserAgent(robotsTxt: string): Map<string, string[]>;
/**
 * Parse robots.txt into a map of user-agent -> list of Allow patterns.
 * Mirrors parseRobotsByUserAgent but captures Allow directives.
 */
export declare function parseRobotsAllowByUserAgent(robotsTxt: string): Map<string, string[]>;
/** True if the Disallow list includes a root block (`/`). */
export declare function isFullyDisallowed(patterns: string[] | undefined): boolean;
/**
 * RFC 9309 block status for an agent given its disallow and allow patterns.
 * Returns:
 *   "none"    — not blocked (no root disallow, or root disallow overridden by Allow: /)
 *   "partial" — root disallow with some Allow paths that reopen part of the site (but not all)
 *   "full"    — root disallow with no overriding Allow
 */
export declare function blockStatus(disallowPatterns: string[] | undefined, allowPatterns: string[] | undefined): "none" | "partial" | "full";
export interface CrawlerAccessOptions {
    /** List of AI crawler user-agents to check. */
    crawlers?: readonly string[];
}
/**
 * Warn per blocked AI crawler; escalate to error when all configured crawlers are blocked.
 * Wildcard blocks (`User-agent: *` + `Disallow: /`) also count as blocking each named crawler
 * unless the crawler has its own more-permissive block.
 * Per RFC 9309, Allow directives override Disallow when more specific (or equal length).
 */
export declare function crawlerAccessRule(robotsTxtContent: string, options?: CrawlerAccessOptions): RuleResult[];
//# sourceMappingURL=crawler-access.d.ts.map