@spaik/mcp-server-roi/dist/services/response-adapter.d.ts

MCP server for AI ROI prediction and tracking with Monte Carlo simulations

import { z } from 'zod';
/**
 * Response Adapter Service
 *
 * Adapts responses based on AI agent preferences, format requirements,
 * and interaction patterns for optimal consumption.
 */
export declare const ResponseFormatSchema: z.ZodEnum<["full_json", "executive_only", "narrative", "structured_summary", "conversational", "markdown_report", "csv_data", "visual_ready"]>;
export declare const AgentPreferencesSchema: z.ZodObject<{
    preferred_format: z.ZodEnum<["full_json", "executive_only", "narrative", "structured_summary", "conversational", "markdown_report", "csv_data", "visual_ready"]>;
    detail_level: z.ZodEnum<["minimal", "standard", "comprehensive"]>;
    language_style: z.ZodEnum<["formal", "casual", "technical", "executive"]>;
    include_visuals: z.ZodDefault<z.ZodBoolean>;
    max_response_tokens: z.ZodOptional<z.ZodNumber>;
    focus_areas: z.ZodOptional<z.ZodArray<z.ZodString, "many">>;
    skip_sections: z.ZodOptional<z.ZodArray<z.ZodString, "many">>;
}, "strip", z.ZodTypeAny, {
    preferred_format: "narrative" | "full_json" | "executive_only" | "structured_summary" | "conversational" | "markdown_report" | "csv_data" | "visual_ready";
    detail_level: "minimal" | "standard" | "comprehensive";
    language_style: "formal" | "casual" | "technical" | "executive";
    include_visuals: boolean;
    max_response_tokens?: number | undefined;
    focus_areas?: string[] | undefined;
    skip_sections?: string[] | undefined;
}, {
    preferred_format: "narrative" | "full_json" | "executive_only" | "structured_summary" | "conversational" | "markdown_report" | "csv_data" | "visual_ready";
    detail_level: "minimal" | "standard" | "comprehensive";
    language_style: "formal" | "casual" | "technical" | "executive";
    include_visuals?: boolean | undefined;
    max_response_tokens?: number | undefined;
    focus_areas?: string[] | undefined;
    skip_sections?: string[] | undefined;
}>;
export declare const AdaptedResponseSchema: z.ZodObject<{
    format: z.ZodEnum<["full_json", "executive_only", "narrative", "structured_summary", "conversational", "markdown_report", "csv_data", "visual_ready"]>;
    content: z.ZodAny;
    metadata: z.ZodObject<{
        original_format: z.ZodString;
        adaptation_notes: z.ZodArray<z.ZodString, "many">;
        tokens_saved: z.ZodOptional<z.ZodNumber>;
        quality_score: z.ZodNumber;
    }, "strip", z.ZodTypeAny, {
        original_format: string;
        adaptation_notes: string[];
        quality_score: number;
        tokens_saved?: number | undefined;
    }, {
        original_format: string;
        adaptation_notes: string[];
        quality_score: number;
        tokens_saved?: number | undefined;
    }>;
}, "strip", z.ZodTypeAny, {
    metadata: {
        original_format: string;
        adaptation_notes: string[];
        quality_score: number;
        tokens_saved?: number | undefined;
    };
    format: "narrative" | "full_json" | "executive_only" | "structured_summary" | "conversational" | "markdown_report" | "csv_data" | "visual_ready";
    content?: any;
}, {
    metadata: {
        original_format: string;
        adaptation_notes: string[];
        quality_score: number;
        tokens_saved?: number | undefined;
    };
    format: "narrative" | "full_json" | "executive_only" | "structured_summary" | "conversational" | "markdown_report" | "csv_data" | "visual_ready";
    content?: any;
}>;
export type ResponseFormat = z.infer<typeof ResponseFormatSchema>;
export type AgentPreferences = z.infer<typeof AgentPreferencesSchema>;
export type AdaptedResponse = z.infer<typeof AdaptedResponseSchema>;
export declare class ResponseAdapter {
    private logger;
    private readonly AGENT_PROFILES;
    /**
     * Adapt response based on detected or specified preferences
     */
    adaptResponse(response: any, preferences?: AgentPreferences, context?: any): Promise<AdaptedResponse>;
    /**
     * Detect agent type and preferences from interaction patterns
     */
    detectAgentType(queryHistory: any[], currentQuery: any): Promise<{
        agent_type: string;
        confidence: number;
        detected_patterns: string[];
        recommended_preferences: AgentPreferences;
    }>;
    /**
     * Optimize response for specific use cases
     */
    optimizeForUseCase(response: any, useCase: 'decision_making' | 'exploration' | 'reporting' | 'monitoring'): Promise<any>;
    /**
     * Convert response to specific output formats
     */
    convertToFormat(response: any, targetFormat: 'json' | 'yaml' | 'xml' | 'html' | 'pdf_ready'): Promise<string>;
    private formatExecutiveOnly;
    private formatNarrative;
    private formatStructuredSummary;
    private formatConversational;
    private formatMarkdownReport;
    private formatCSVData;
    private formatVisualReady;
    private detectPreferences;
    private extractKeyMetrics;
    private extractDecisionPoints;
    private extractTopRecommendations;
    private extractTopRisks;
    private makeExecutiveLanguage;
    private applyTokenLimit;
    private generateAdaptationMetadata;
    private assessAdaptationQuality;
    private analyzeInteractionPatterns;
    private classifyAgentType;
    private calculateTypeConfidence;
    private extractFocusAreaData;
    private optimizeForDecisionMaking;
    private optimizeForExploration;
    private optimizeForReporting;
    private optimizeForMonitoring;
    private calculateHealthScore;
    private calculateChange;
    private generateAlerts;
    private convertToYAML;
    private objectToYAML;
    private convertToXML;
    private objectToXML;
    private escapeXML;
    private convertToHTML;
    private convertToPDFReady;
    private prepareROIChartData;
    private prepareTimelineData;
    private prepareValueDistribution;
    private prepareMetricsTable;
    private prepareKPIs;
}
export declare const responseAdapter: ResponseAdapter;
//# sourceMappingURL=response-adapter.d.ts.map