@framers/agentos 0.1.0

package/dist/core/llm/IPromptEngine.d.ts

@@ -0,0 +1,627 @@
+/**
+ * @fileoverview Defines the IPromptEngine interface and related types that form the core
+ * of AgentOS's adaptive and contextual prompting system. The PromptEngine is responsible
+ * for dynamically constructing prompts based on rich contextual information, persona
+ * definitions, and runtime state. It supports contextual element selection, token budgeting,
+ * content summarization, and multiple prompt template formats.
+ *
+ * This interface is central to the GMI (Generalized Mind Instance) architecture, enabling
+ * sophisticated prompt adaptation based on user skill level, mood, task complexity,
+ * conversation state, and persona-specific rules.
+ *
+ * Key responsibilities include:
+ * - Dynamic contextual element selection based on PromptExecutionContext.
+ * - Intelligent token budgeting and content truncation/summarization using an IUtilityAI helper.
+ * - Multi-modal prompt component integration (text, vision, audio data references, tools).
+ * - Template-based prompt formatting for different LLM providers and models.
+ * - Comprehensive error handling and issue reporting within the PromptEngineResult.
+ * - Performance optimization strategies such as caching.
+ *
+ * @module backend/agentos/core/llm/IPromptEngine
+ * @see {@link ../../../docs/PROMPTS.MD} for detailed architectural documentation.
+ * @see {@link ../../cognitive_substrate/personas/IPersonaDefinition.ts} for persona-driven prompting.
+ */
+import { IPersonaDefinition, ContextualPromptElement, ContextualPromptElementCriteria } from '../../cognitive_substrate/personas/IPersonaDefinition';
+import { IWorkingMemory } from '../../cognitive_substrate/memory/IWorkingMemory';
+import { ITool } from '../tools/ITool';
+import { VisionInputData, AudioInputData } from '../../cognitive_substrate/IGMI';
+import { ConversationMessage as Message } from '../conversation/ConversationMessage';
+import { ChatMessage } from './providers/IProvider';
+/**
+ * Represents different types of contextual prompt elements that can be dynamically
+ * selected and integrated into prompts based on execution context.
+ * These elements allow for fine-grained adaptation of prompts.
+ * @enum {string}
+ */
+export declare enum ContextualElementType {
+    /** Additional system-level instructions appended to base system prompt. */
+    SYSTEM_INSTRUCTION_ADDON = "system_instruction_addon",
+    /** Dynamic few-shot examples selected based on context to guide the LLM. */
+    FEW_SHOT_EXAMPLE = "few_shot_example",
+    /** Behavioral guidance or tone adjustments for the persona. */
+    BEHAVIORAL_GUIDANCE = "behavioral_guidance",
+    /** Specific instructions or constraints related to the current task. */
+    TASK_SPECIFIC_INSTRUCTION = "task_specific_instruction",
+    /** Instructions for handling errors or recovering from unexpected situations. */
+    ERROR_HANDLING_GUIDANCE = "error_handling_guidance",
+    /** Adjustments to the GMI's interaction style with the user. */
+    INTERACTION_STYLE_MODIFIER = "interaction_style_modifier",
+    /** Domain-specific knowledge, facts, or context relevant to the current query. */
+    DOMAIN_CONTEXT = "domain_context",
+    /** Ethical guidelines or safety instructions to ensure responsible AI behavior. */
+    ETHICAL_GUIDELINE = "ethical_guideline",
+    /** Specifications for the desired output format (e.g., JSON, Markdown). */
+    OUTPUT_FORMAT_SPEC = "output_format_spec",
+    /** Instructions for specific reasoning protocols (e.g., chain-of-thought, tree-of-thought). */
+    REASONING_PROTOCOL = "reasoning_protocol",
+    /** Dynamic content to be injected directly into the user part of the prompt. */
+    USER_PROMPT_AUGMENTATION = "user_prompt_augmentation",
+    /** Content to be injected into the assistant part for few-shot or role-play setup. */
+    ASSISTANT_PROMPT_AUGMENTATION = "assistant_prompt_augmentation"
+}
+/**
+ * Comprehensive execution context that drives dynamic prompt construction.
+ * This context is assembled by the GMI and contains all relevant information
+ * needed for intelligent contextual element selection and prompt adaptation.
+ * @interface PromptExecutionContext
+ */
+export interface PromptExecutionContext {
+    /** The currently active persona definition, guiding overall behavior and prompt structure. */
+    activePersona: IPersonaDefinition;
+    /** Access to the GMI's working memory for dynamic value retrieval (e.g., user preferences, GMI traits). */
+    workingMemory: IWorkingMemory;
+    /** Current mood state of the GMI, which can influence tone and element selection. */
+    currentMood?: string;
+    /** Assessed or declared user skill level (e.g., 'beginner', 'expert') relevant to the task. */
+    userSkillLevel?: 'beginner' | 'intermediate' | 'advanced' | 'expert' | string;
+    /** Hint about the current task type (e.g., 'coding', 'writing', 'data_analysis'). */
+    taskHint?: string;
+    /** Assessed complexity level of the current task (e.g., 'simple', 'complex'). */
+    taskComplexity?: 'simple' | 'moderate' | 'complex' | 'highly_complex' | string;
+    /** Target language for the response (BCP-47 code, e.g., 'en-US', 'fr-FR'). */
+    language?: string;
+    /** Signals detected in the conversation (e.g., 'user_confused', 'task_completed_successfully'). */
+    conversationSignals?: string[];
+    /** Custom context flags for specialized prompting scenarios or A/B testing. */
+    customContext?: Record<string, unknown>;
+    /** User-specific preferences that affect prompting (e.g., verbosity, preferred formats). */
+    userPreferences?: Record<string, unknown>;
+    /** Historical interaction patterns that inform adaptation (e.g., preferred response styles). */
+    interactionHistorySummary?: {
+        successfulInteractionCount: number;
+        problematicInteractionCount: number;
+        commonTopics?: string[];
+        lastInteractionTimestamp?: string;
+    };
+    /** Current session metadata (e.g., duration, number of turns). */
+    sessionMetadata?: {
+        sessionDurationSeconds: number;
+        interactionCountInSession: number;
+        userEngagementLevel?: 'low' | 'medium' | 'high';
+    };
+}
+/**
+ * Core components that form the foundation of any prompt construction.
+ * These are gathered from various sources (GMI state, user input, RAG) and then
+ * augmented with dynamically selected contextual elements.
+ * @interface PromptComponents
+ */
+export interface PromptComponents {
+    /** System-level prompts with optional priority ordering. Higher priority usually means placed earlier or given more weight. */
+    systemPrompts?: Array<{
+        content: string;
+        priority?: number;
+        source?: string;
+    }>;
+    /** Conversation history messages, typically an array of `Message` objects. */
+    conversationHistory?: Message[];
+    /** Current user input text. */
+    userInput?: string | null;
+    /** Visual inputs (images) if the target model supports vision. */
+    visionInputs?: VisionInputData[];
+    /** Audio input data references if the model or a pre-processing step handles audio. */
+    audioInput?: AudioInputData;
+    /** Available tools and their schemas, for models that support function/tool calling. */
+    tools?: ITool[];
+    /**
+     * Pre-formatted tool/function schemas that should be forwarded to the model as-is.
+     * Useful when upstream logic has already normalized the schema definitions.
+     */
+    toolSchemas?: Array<Record<string, unknown>>;
+    /** Retrieved context from a RAG system, to be incorporated into the prompt. */
+    retrievedContext?: string | Array<{
+        source: string;
+        content: string;
+        relevance?: number;
+        type?: string;
+    }>;
+    /** Task-specific data or parameters that need to be included in the prompt. */
+    taskSpecificData?: Record<string, unknown>;
+    /** Additional custom components that templates might use. */
+    customComponents?: Record<string, unknown>;
+}
+/**
+ * Information about the target AI model that affects prompt construction.
+ * This guides template selection, token limits, and capability-specific formatting.
+ * @interface ModelTargetInfo
+ */
+export interface ModelTargetInfo {
+    /** Unique identifier of the target model (e.g., "gpt-4o", "ollama/llama3"). */
+    modelId: string;
+    /** Identifier of the provider hosting the model (e.g., "openai", "ollama"). */
+    providerId: string;
+    /** Maximum context length in tokens supported by the model. */
+    maxContextTokens: number;
+    /**
+     * Optional: Optimal context length for best performance/cost-efficiency, if different from max.
+     * Prompts might be targeted to this length.
+     */
+    optimalContextTokens?: number;
+    /** A list of functional capabilities of the model (e.g., 'tool_use', 'vision_input', 'json_mode'). */
+    capabilities: string[];
+    /**
+     * The type of prompt format the model expects.
+     * 'openai_chat': Standard OpenAI chat completion format (array of messages).
+     * 'anthropic_messages': Anthropic Messages API format (messages array + optional system prompt).
+     * 'generic_completion': A single string prompt for older completion-style models.
+     * 'custom': A custom format handled by a specific template.
+     */
+    promptFormatType: 'openai_chat' | 'anthropic_messages' | 'generic_completion' | 'custom' | string;
+    /** Configuration for tool/function calling support. */
+    toolSupport: {
+        supported: boolean;
+        /** Format expected by the model for tool definitions and calls. */
+        format: 'openai_functions' | 'anthropic_tools' | 'google_tools' | 'custom' | string;
+        /** Maximum number of tools that can be defined or called in a single interaction. */
+        maxToolsPerCall?: number;
+    };
+    /** Vision input support configuration. */
+    visionSupport?: {
+        supported: boolean;
+        maxImages?: number;
+        supportedFormats?: string[];
+        maxImageResolution?: string;
+    };
+    /** Audio input support configuration (more likely for pre-processing than direct model input). */
+    audioSupport?: {
+        supported: boolean;
+        requiresTranscription?: boolean;
+    };
+    /** Model-specific hints for optimizing prompt construction or token budgeting. */
+    optimizationHints?: {
+        preferredSystemPromptLength?: number;
+        optimalHistoryMessages?: number;
+        tokenBudgetingStrategy?: 'aggressive_truncate' | 'summarize_old' | 'balanced';
+    };
+}
+/**
+ * The final formatted prompt ready for submission to an LLM provider.
+ * The structure varies based on the target model's requirements.
+ * - `ChatMessage[]`: For models like OpenAI Chat.
+ * - `string`: For older text completion models.
+ * - `object`: For models with more complex input structures (e.g., Anthropic Messages API which takes an object with `messages` and `system`).
+ * @type FormattedPrompt
+ */
+export type FormattedPrompt = ChatMessage[] | string | {
+    messages: ChatMessage[];
+    system?: string;
+    tools?: any[];
+    [key: string]: any;
+};
+/**
+ * Comprehensive result object returned by prompt construction, containing
+ * the formatted prompt, metadata, issues encountered, and optimization information.
+ * @interface PromptEngineResult
+ */
+export interface PromptEngineResult {
+    /** The final formatted prompt ready for LLM consumption. */
+    prompt: FormattedPrompt;
+    /**
+     * Formatted tool schemas compatible with the target model's API, if tools are used.
+     * The structure of `any[]` depends on `ModelTargetInfo.toolSupport.format`.
+     */
+    formattedToolSchemas?: any[];
+    /** Estimated token count of the constructed prompt before precise provider counting. */
+    estimatedTokenCount?: number;
+    /** Precise token count, if available from a tokenizer or after construction. */
+    tokenCount?: number;
+    /** Any issues (errors, warnings, info) encountered during prompt construction. */
+    issues?: Array<{
+        type: 'error' | 'warning' | 'info';
+        code: string;
+        message: string;
+        details?: unknown;
+        suggestion?: string;
+        component?: string;
+    }>;
+    /** Indicates if content was truncated or summarized to fit token limits. */
+    wasTruncatedOrSummarized: boolean;
+    /** Details about modifications made during construction (e.g., which components were truncated). */
+    modificationDetails?: {
+        originalEstimatedTokenCount?: number;
+        truncatedComponents?: string[];
+        summarizedComponents?: string[];
+        removedComponents?: string[];
+        addedContextualElementIds?: string[];
+    };
+    /** Performance metrics and metadata related to the prompt construction process. */
+    metadata: {
+        constructionTimeMs: number;
+        selectedContextualElementIds: string[];
+        templateUsed: string;
+        totalSystemPromptsApplied: number;
+        historyMessagesIncluded: number;
+        ragContextTokensUsed?: number;
+        [key: string]: unknown;
+    };
+    /** An optional cache key if the result was retrieved from or stored in a cache. */
+    cacheKey?: string;
+}
+/**
+ * Configuration options for the PromptEngine's behavior, optimization strategies,
+ * and integration with other services like IUtilityAI.
+ * @interface PromptEngineConfig
+ */
+export interface PromptEngineConfig {
+    /** Default template name (from `availableTemplates`) to use if none is specified or inferable. */
+    defaultTemplateName: string;
+    /** A record of available prompt template functions, keyed by template name. */
+    availableTemplates: Record<string, PromptTemplateFunction>;
+    /** Configuration for token counting strategies. */
+    tokenCounting: {
+        strategy: 'estimated' | 'precise' | 'hybrid';
+        estimationModel?: 'gpt-3.5-turbo' | 'gpt-4' | 'claude-3' | 'generic';
+    };
+    /** Configuration for managing conversation history within prompts. */
+    historyManagement: {
+        defaultMaxMessages: number;
+        maxTokensForHistory: number;
+        summarizationTriggerRatio: number;
+        preserveImportantMessages: boolean;
+    };
+    /** Configuration for managing retrieved context (e.g., from RAG). */
+    contextManagement: {
+        maxRAGContextTokens: number;
+        summarizationQualityTier: 'fast' | 'balanced' | 'high_quality';
+        preserveSourceAttributionInSummary: boolean;
+        minContextRelevanceThreshold?: number;
+    };
+    /** Configuration for selecting and applying contextual elements. */
+    contextualElementSelection: {
+        /** Max number of elements to apply per type. */
+        maxElementsPerType: Partial<Record<ContextualElementType, number>>;
+        defaultMaxElementsPerType: number;
+        priorityResolutionStrategy: 'highest_first' | 'weighted_random' | 'persona_preference';
+        conflictResolutionStrategy: 'skip_conflicting' | 'merge_compatible' | 'error_on_conflict';
+    };
+    /** Performance optimization settings. */
+    performance: {
+        enableCaching: boolean;
+        cacheTimeoutSeconds: number;
+        maxCacheSizeBytes?: number;
+    };
+    /** Debugging and logging settings. */
+    debugging?: {
+        logConstructionSteps?: boolean;
+        includeDebugMetadataInResult?: boolean;
+        logSelectedContextualElements?: boolean;
+    };
+    /**
+     * Optional tool schema registration manifest enabling per-persona and per-model enable/disable semantics.
+     * Structure:
+     *  - key: personaId (string)
+     *  - value: {
+     *       enabledToolIds: string[];            // Tools explicitly allowed for this persona (intersection with runtime tool list)
+     *       disabledToolIds?: string[];          // Tools explicitly disallowed (takes precedence over enabledToolIds)
+     *       modelOverrides?: {                   // Per model ID fine-grained overrides
+     *         [modelId: string]: string[];       // Exact list of tool IDs allowed for that model when persona active
+     *       };
+     *    }
+     * Resolution Order when filtering tools for prompt construction:
+     *   1. If personaId present in manifest:
+     *      a. If modelOverrides[modelId] exists => allowed set = that array (disabledToolIds still removes).
+     *      b. Else allowed base = enabledToolIds (if defined) else all runtime tools.
+     *      c. Remove any disabledToolIds from allowed set.
+     *   2. If personaId absent => all runtime tools (no filtering).
+     * Note: Unknown tool IDs in manifest are ignored gracefully.
+     */
+    toolSchemaManifest?: Record<string, {
+        enabledToolIds?: string[];
+        disabledToolIds?: string[];
+        modelOverrides?: Record<string, string[]>;
+    }>;
+}
+/**
+ * Function signature for prompt template implementations.
+ * Templates are responsible for taking all processed prompt components and
+ * formatting them into the final `FormattedPrompt` structure required by a specific
+ * LLM provider or model type.
+ * @type PromptTemplateFunction
+ * @param {PromptComponents} components - The core and augmented prompt components.
+ * @param {ModelTargetInfo} modelInfo - Information about the target AI model.
+ * @param {ContextualPromptElement[]} selectedContextualElements - Contextual elements chosen for this prompt.
+ * @param {Readonly<PromptEngineConfig>} config - A read-only view of the PromptEngine's current configuration.
+ * @param {TokenEstimator} estimateTokenCountFn - A function to estimate token counts, useful within templates.
+ * @returns {Promise<FormattedPrompt>} A promise that resolves to the final, formatted prompt.
+ */
+export type PromptTemplateFunction = (components: Readonly<PromptComponents>, modelInfo: Readonly<ModelTargetInfo>, selectedContextualElements: ReadonlyArray<ContextualPromptElement>, config: Readonly<PromptEngineConfig>, estimateTokenCountFn: (content: string, modelId?: string) => Promise<number>) => Promise<FormattedPrompt>;
+/**
+ * Function signature for estimating token counts.
+ * This is passed to templates to allow them to make token-aware decisions.
+ * @type TokenEstimator
+ */
+export type TokenEstimator = (content: string, modelId?: string) => Promise<number>;
+/**
+ * Interface for utility AI services that assist the PromptEngine with complex
+ * content processing tasks like summarization and relevance analysis, specifically
+ * tailored for prompt construction needs.
+ * This is a focused interface used internally by the PromptEngine.
+ * @interface IPromptEngineUtilityAI
+ */
+export interface IPromptEngineUtilityAI {
+    /**
+     * Summarizes conversation history to fit within token constraints, attempting to preserve key information.
+     * @param {Message[]} messages - The array of conversation messages to summarize.
+     * @param {number} targetTokenCount - The desired maximum token count for the summary.
+     * @param {Readonly<ModelTargetInfo>} modelInfo - Information about the model for which the summary is being prepared.
+     * @param {boolean} [preserveImportantMessages] - If true, attempt to identify and keep important messages verbatim.
+     * @returns {Promise<{ summaryMessages: Message[]; originalTokenCount: number; finalTokenCount: number; messagesSummarized: number }>}
+     * A summary (which might be a single system message or a condensed list of messages),
+     * and metadata about the summarization.
+     */
+    summarizeConversationHistory(messages: ReadonlyArray<Message>, targetTokenCount: number, modelInfo: Readonly<ModelTargetInfo>, preserveImportantMessages?: boolean): Promise<{
+        summaryMessages: Message[];
+        originalTokenCount: number;
+        finalTokenCount: number;
+        messagesSummarized: number;
+    }>;
+    /**
+     * Summarizes retrieved RAG context to fit token limits, ideally preserving source attribution if possible.
+     * @param {string | Array<{ source: string; content: string; relevance?: number }>} context - The RAG context to summarize.
+     * @param {number} targetTokenCount - The desired maximum token count for the summarized context.
+     * @param {Readonly<ModelTargetInfo>} modelInfo - Information about the model.
+     * @param {boolean} [preserveSourceAttribution] - If true, attempt to retain source information in the summary.
+     * @returns {Promise<{ summary: string; originalTokenCount: number; finalTokenCount: number; preservedSources?: string[] }>}
+     * The summarized text and metadata.
+     */
+    summarizeRAGContext(context: string | ReadonlyArray<{
+        source: string;
+        content: string;
+        relevance?: number;
+    }>, targetTokenCount: number, modelInfo: Readonly<ModelTargetInfo>, preserveSourceAttribution?: boolean): Promise<{
+        summary: string;
+        originalTokenCount: number;
+        finalTokenCount: number;
+        preservedSources?: string[];
+    }>;
+    /**
+     * Analyzes a piece of content for its relevance and importance within the current execution context.
+     * This can be used to prioritize which content to include or how to emphasize it.
+     * @param {string} content - The text content to analyze.
+     * @param {Readonly<PromptExecutionContext>} executionContext - The current execution context.
+     * @param {Readonly<ModelTargetInfo>} modelInfo - Information about the model.
+     * @returns {Promise<{ relevanceScore: number; importanceScore: number; keywords?: string[]; topics?: string[] }>}
+     * Scores and extracted metadata.
+     */
+    analyzeContentRelevance?(content: string, executionContext: Readonly<PromptExecutionContext>, modelInfo: Readonly<ModelTargetInfo>): Promise<{
+        relevanceScore: number;
+        importanceScore: number;
+        keywords?: string[];
+        topics?: string[];
+    }>;
+}
+/**
+ * Core interface for the PromptEngine, responsible for intelligent and adaptive
+ * prompt construction based on rich contextual information and persona definitions.
+ *
+ * The PromptEngine serves as the central orchestrator for AgentOS's sophisticated
+ * prompting system, capable of dynamically selecting contextual elements,
+ * managing token budgets, integrating multi-modal content, and optimizing
+ * prompts for different AI models and interaction patterns.
+ *
+ * @interface IPromptEngine
+ */
+export interface IPromptEngine {
+    /**
+     * Initializes the PromptEngine with its configuration and an optional utility AI service.
+     * This method must be called successfully before any prompt construction.
+     *
+     * @async
+     * @param {PromptEngineConfig} config - The comprehensive configuration for the engine,
+     * including template definitions, token counting strategies, history management rules,
+     * contextual element selection parameters, performance settings, and debugging options.
+     * @param {IPromptEngineUtilityAI} [utilityAI] - An optional utility AI service instance, conforming to
+     * `IPromptEngineUtilityAI`, used for advanced content processing tasks like summarization
+     * of conversation history or RAG context within the prompt construction pipeline.
+     * @returns {Promise<void>} A promise that resolves when the engine is fully initialized and ready.
+     * @throws {PromptEngineError} If the provided configuration is invalid or if a critical
+     * initialization step fails (e.g., loading default templates).
+     *
+     * @example
+     * const engine = new PromptEngine();
+     * await engine.initialize(myAppConfig.promptEngine, myUtilityAIService);
+     */
+    initialize(config: PromptEngineConfig, utilityAI?: IPromptEngineUtilityAI): Promise<void>;
+    /**
+     * The primary method for constructing an adaptive and contextually relevant prompt.
+     * This orchestrates the entire pipeline: contextual element evaluation and selection,
+     * augmentation of base components, token budget management (including truncation and
+     * summarization), and final formatting using a model-appropriate template.
+     *
+     * @async
+     * @param {Readonly<PromptComponents>} baseComponents - The core, static components of the prompt,
+     * such as system instructions, conversation history, and current user input. These are read-only
+     * to prevent unintended modification by the method.
+     * @param {Readonly<ModelTargetInfo>} modelTargetInfo - Detailed information about the target AI model,
+     * including its ID, provider, capabilities, token limits, and expected prompt format. This is crucial
+     * for tailoring the prompt effectively.
+     * @param {Readonly<PromptExecutionContext>} [executionContext] - Optional. The rich runtime context,
+     * including active persona, working memory, user state, and task details. This drives the
+     * dynamic selection and application of contextual prompt elements.
+     * @param {string} [templateName] - Optional. The explicit name of a prompt template to use.
+     * If not provided, the engine selects a default template based on `modelTargetInfo.promptFormatType`
+     * or the `defaultTemplateName` from its configuration.
+     * @returns {Promise<PromptEngineResult>} A promise resolving to a `PromptEngineResult` object,
+     * which contains the final formatted prompt, along with metadata about its construction,
+     * token counts, any issues encountered, and modifications made.
+     * @throws {PromptEngineError} If a non-recoverable error occurs during any stage of prompt
+     * construction (e.g., template not found, critical component missing, tokenization failure).
+     */
+    constructPrompt(baseComponents: Readonly<PromptComponents>, modelTargetInfo: Readonly<ModelTargetInfo>, executionContext?: Readonly<PromptExecutionContext>, templateName?: string): Promise<PromptEngineResult>;
+    /**
+     * Evaluates whether a given set of `ContextualPromptElementCriteria` is satisfied by the current
+     * `PromptExecutionContext`. This method is the core of the dynamic adaptation logic, determining
+     * which contextual elements are relevant and should be incorporated into the prompt.
+     * It supports checking various context aspects like mood, user skill, task hints, working memory values, etc.
+     *
+     * @async
+     * @param {Readonly<ContextualPromptElementCriteria>} criteria - The criteria defined within a `ContextualPromptElement`
+     * from the persona definition. These criteria specify the conditions under which the element applies.
+     * @param {Readonly<PromptExecutionContext>} context - The current execution context against which
+     * the criteria are evaluated.
+     * @returns {Promise<boolean>} A promise resolving to `true` if all conditions within the criteria
+     * are met (respecting logical operators like AND/OR if defined in criteria), `false` otherwise.
+     * @throws {PromptEngineError} If criteria evaluation encounters an unexpected error (e.g., accessing
+     * a non-existent working memory key specified in a query).
+     */
+    evaluateCriteria(criteria: Readonly<ContextualPromptElementCriteria>, context: Readonly<PromptExecutionContext>): Promise<boolean>;
+    /**
+     * Estimates the token count for a given piece of text, optionally using a specific model ID
+     * to inform a more precise estimation if available (e.g., using a model-specific tokenizer).
+     * This is used internally for token budgeting and can also be exposed as a utility.
+     *
+     * @async
+     * @param {string} content - The text content for which to estimate token count.
+     * @param {string} [modelId] - Optional. The ID of the target model. If provided, the engine may
+     * attempt a more precise tokenization based on this model's characteristics.
+     * @returns {Promise<number>} A promise resolving to the estimated number of tokens.
+     */
+    estimateTokenCount(content: string, modelId?: string): Promise<number>;
+    /**
+     * Registers a new prompt template function with the engine. This allows for extending
+     * the engine's capabilities to support new LLM providers, model families, or specialized
+     * prompt formatting requirements dynamically at runtime.
+     *
+     * @async
+     * @param {string} templateName - A unique name to identify the template. If a template with
+     * this name already exists, it will be overwritten (a warning may be logged).
+     * @param {PromptTemplateFunction} templateFunction - The function that implements the template logic.
+     * It receives `PromptComponents`, `ModelTargetInfo`, selected `ContextualPromptElement[]`,
+     * engine `config`, and a `tokenEstimatorFn`, and must return a `FormattedPrompt`.
+     * @returns {Promise<void>} A promise that resolves when the template is successfully registered.
+     * @throws {PromptEngineError} If the `templateName` is invalid or `templateFunction` is not a function.
+     */
+    registerTemplate(templateName: string, templateFunction: PromptTemplateFunction): Promise<void>;
+    /**
+     * Validates a given set of prompt components and model information against the engine's
+     * understanding of best practices and potential issues. Useful for development, debugging
+     * persona definitions, or providing feedback to users designing prompts.
+     *
+     * @async
+     * @param {Readonly<PromptComponents>} components - The prompt components to validate.
+     * @param {Readonly<ModelTargetInfo>} modelTargetInfo - Information about the target model.
+     * @param {Readonly<PromptExecutionContext>} [executionContext] - Optional. The context in which these
+     * components would be used, allowing for context-aware validation.
+     * @returns {Promise<{ isValid: boolean; issues: Array<{ type: 'error' | 'warning'; message: string; suggestion?: string }>; recommendations: string[]; }>}
+     * An object containing a boolean `isValid` (true if no errors), a list of `issues` found
+     * (errors or warnings with messages and suggestions), and a list of `recommendations` for improvement.
+     */
+    validatePromptConfiguration(components: Readonly<PromptComponents>, modelTargetInfo: Readonly<ModelTargetInfo>, executionContext?: Readonly<PromptExecutionContext>): Promise<{
+        isValid: boolean;
+        issues: Array<{
+            type: 'error' | 'warning';
+            code: string;
+            message: string;
+            suggestion?: string;
+            component?: string;
+        }>;
+        recommendations?: string[];
+    }>;
+    /**
+     * Clears internal caches used by the PromptEngine (e.g., for prompt construction results
+     * or token counts). This can be used to free memory or to force re-computation for
+     * debugging or after configuration changes.
+     *
+     * @async
+     * @param {string} [selectivePattern] - Optional. A pattern or key to clear only specific
+     * cache entries (e.g., "modelId:gpt-4o*"). If omitted, the entire cache is cleared.
+     * The exact format of the pattern is implementation-dependent.
+     * @returns {Promise<void>} A promise that resolves when the cache clearing operation is complete.
+     */
+    clearCache(selectivePattern?: string): Promise<void>;
+    /**
+     * Retrieves current performance and usage statistics from the PromptEngine.
+     * This data can be used for monitoring, optimization, and understanding engine behavior.
+     *
+     * @async
+     * @returns {Promise<PromptEngineStats>} A promise resolving to an object containing
+     * various statistics like total prompts constructed, average construction time,
+     * cache hit rate, error rates, and usage of contextual elements.
+     */
+    getEngineStatistics(): Promise<{
+        totalPromptsConstructed: number;
+        averageConstructionTimeMs: number;
+        cacheStats: {
+            hits: number;
+            misses: number;
+            currentSize: number;
+            maxSize?: number;
+            effectivenessRatio: number;
+        };
+        tokenCountingStats: {
+            operations: number;
+            averageAccuracy?: number;
+        };
+        contextualElementUsage: Record<string, {
+            count: number;
+            averageEvaluationTimeMs?: number;
+        }>;
+        errorRatePerType: Record<string, number>;
+        performanceTimers: Record<string, {
+            count: number;
+            totalTimeMs: number;
+            averageTimeMs: number;
+        }>;
+    }>;
+}
+/**
+ * Custom error class for all errors originating from the PromptEngine.
+ * This allows for specific catching and handling of prompt construction failures.
+ * @class PromptEngineError
+ * @extends {Error}
+ */
+export declare class PromptEngineError extends Error {
+    /**
+     * A specific error code for programmatic handling (e.g., 'TEMPLATE_NOT_FOUND', 'TOKEN_LIMIT_EXCEEDED_IRRECOVERABLY').
+     * @public
+     * @readonly
+     * @type {string}
+     */
+    readonly code: string;
+    /**
+     * Optional: The specific component or operation within the PromptEngine where the error occurred
+     * (e.g., 'TokenBudgeting', 'TemplateFormatting', 'ContextualElementSelection').
+     * @public
+     * @readonly
+     * @type {string | undefined}
+     */
+    readonly component?: string;
+    /**
+     * Optional: Additional structured details about the error, which might include relevant
+     * data like modelId, offending component, or configuration issue.
+     * @public
+     * @readonly
+     * @type {unknown | undefined}
+     */
+    readonly details?: unknown;
+    /**
+     * Creates an instance of PromptEngineError.
+     * @param {string} message - A human-readable description of the error.
+     * @param {string} code - A specific error code (e.g., 'TEMPLATE_NOT_FOUND').
+     * @param {string} [component] - The engine component where the error originated.
+     * @param {unknown} [details] - Additional context or the underlying error.
+     */
+    constructor(message: string, code: string, component?: string, details?: unknown);
+}
+export type { VisionInputData, AudioInputData };
+//# sourceMappingURL=IPromptEngine.d.ts.map

package/dist/core/llm/IPromptEngine.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"IPromptEngine.d.ts","sourceRoot":"","sources":["../../../src/core/llm/IPromptEngine.ts"],"names":[],"mappings":"AACA;;;;;;;;;;;;;;;;;;;;;;GAsBG;AAEH,OAAO,EACL,kBAAkB,EAClB,uBAAuB,EACvB,+BAA+B,EAChC,MAAM,uDAAuD,CAAC;AAC/D,OAAO,EAAE,cAAc,EAAE,MAAM,iDAAiD,CAAC;AACjF,OAAO,EAAE,KAAK,EAAE,MAAM,gBAAgB,CAAC;AACvC,OAAO,EAAE,eAAe,EAAE,cAAc,EAAE,MAAM,gCAAgC,CAAC;AACjF,OAAO,EAAE,mBAAmB,IAAI,OAAO,EAAE,MAAM,qCAAqC,CAAC;AACrF,OAAO,EAAE,WAAW,EAAE,MAAM,uBAAuB,CAAC;AAEpD;;;;;GAKG;AACH,oBAAY,qBAAqB;IAC/B,2EAA2E;IAC3E,wBAAwB,6BAA6B;IACrD,4EAA4E;IAC5E,gBAAgB,qBAAqB;IACrC,+DAA+D;IAC/D,mBAAmB,wBAAwB;IAC3C,wEAAwE;IACxE,yBAAyB,8BAA8B;IACvD,iFAAiF;IACjF,uBAAuB,4BAA4B;IACnD,gEAAgE;IAChE,0BAA0B,+BAA+B;IACzD,kFAAkF;IAClF,cAAc,mBAAmB;IACjC,mFAAmF;IACnF,iBAAiB,sBAAsB;IACvC,2EAA2E;IAC3E,kBAAkB,uBAAuB;IACzC,+FAA+F;IAC/F,kBAAkB,uBAAuB;IACzC,gFAAgF;IAChF,wBAAwB,6BAA6B;IACrD,sFAAsF;IACtF,6BAA6B,kCAAkC;CAChE;AAED;;;;;GAKG;AACH,MAAM,WAAW,sBAAsB;IACrC,8FAA8F;IAC9F,aAAa,EAAE,kBAAkB,CAAC;IAClC,2GAA2G;IAC3G,aAAa,EAAE,cAAc,CAAC;IAC9B,qFAAqF;IACrF,WAAW,CAAC,EAAE,MAAM,CAAC;IACrB,+FAA+F;IAC/F,cAAc,CAAC,EAAE,UAAU,GAAG,cAAc,GAAG,UAAU,GAAG,QAAQ,GAAG,MAAM,CAAC;IAC9E,qFAAqF;IACrF,QAAQ,CAAC,EAAE,MAAM,CAAC;IAClB,iFAAiF;IACjF,cAAc,CAAC,EAAE,QAAQ,GAAG,UAAU,GAAG,SAAS,GAAG,gBAAgB,GAAG,MAAM,CAAC;IAC/E,8EAA8E;IAC9E,QAAQ,CAAC,EAAE,MAAM,CAAC;IAClB,mGAAmG;IACnG,mBAAmB,CAAC,EAAE,MAAM,EAAE,CAAC;IAC/B,+EAA+E;IAC/E,aAAa,CAAC,EAAE,MAAM,CAAC,MAAM,EAAE,OAAO,CAAC,CAAC;IACxC,4FAA4F;IAC5F,eAAe,CAAC,EAAE,MAAM,CAAC,MAAM,EAAE,OAAO,CAAC,CAAC;IAC1C,gGAAgG;IAChG,yBAAyB,CAAC,EAAE;QAC1B,0BAA0B,EAAE,MAAM,CAAC;QACnC,2BAA2B,EAAE,MAAM,CAAC;QACpC,YAAY,CAAC,EAAE,MAAM,EAAE,CAAC;QACxB,wBAAwB,CAAC,EAAE,MAAM,CAAC;KACnC,CAAC;IACF,kEAAkE;IAClE,eAAe,CAAC,EAAE;QAChB,sBAAsB,EAAE,MAAM,CAAC;QAC/B,yBAAyB,EAAE,MAAM,CAAC;QAClC,mBAAmB,CAAC,EAAE,KAAK,GAAG,QAAQ,GAAG,MAAM,CAAC;KACjD,CAAC;CACH;AAED;;;;;GAKG;AACH,MAAM,WAAW,gBAAgB;IAC/B,+HAA+H;IAC/H,aAAa,CAAC,EAAE,KAAK,CAAC;QAAE,OAAO,EAAE,MAAM,CAAC;QAAC,QAAQ,CAAC,EAAE,MAAM,CAAC;QAAC,MAAM,CAAC,EAAE,MAAM,CAAA;KAAE,CAAC,CAAC;IAC/E,8EAA8E;IAC9E,mBAAmB,CAAC,EAAE,OAAO,EAAE,CAAC;IAChC,+BAA+B;IAC/B,SAAS,CAAC,EAAE,MAAM,GAAG,IAAI,CAAC;IAC1B,kEAAkE;IAClE,YAAY,CAAC,EAAE,eAAe,EAAE,CAAC;IACjC,uFAAuF;IACvF,UAAU,CAAC,EAAE,cAAc,CAAC;IAC5B,wFAAwF;IACxF,KAAK,CAAC,EAAE,KAAK,EAAE,CAAC;IAChB;;;OAGG;IACH,WAAW,CAAC,EAAE,KAAK,CAAC,MAAM,CAAC,MAAM,EAAE,OAAO,CAAC,CAAC,CAAC;IAC7C,+EAA+E;IAC/E,gBAAgB,CAAC,EAAE,MAAM,GAAG,KAAK,CAAC;QAAE,MAAM,EAAE,MAAM,CAAC;QAAC,OAAO,EAAE,MAAM,CAAC;QAAC,SAAS,CAAC,EAAE,MAAM,CAAC;QAAC,IAAI,CAAC,EAAE,MAAM,CAAA;KAAE,CAAC,CAAC;IAC1G,+EAA+E;IAC/E,gBAAgB,CAAC,EAAE,MAAM,CAAC,MAAM,EAAE,OAAO,CAAC,CAAC;IAC3C,6DAA6D;IAC7D,gBAAgB,CAAC,EAAE,MAAM,CAAC,MAAM,EAAE,OAAO,CAAC,CAAC;CAC5C;AAED;;;;GAIG;AACH,MAAM,WAAW,eAAe;IAC9B,+EAA+E;IAC/E,OAAO,EAAE,MAAM,CAAC;IAChB,+EAA+E;IAC/E,UAAU,EAAE,MAAM,CAAC;IACnB,+DAA+D;IAC/D,gBAAgB,EAAE,MAAM,CAAC;IACzB;;;OAGG;IACH,oBAAoB,CAAC,EAAE,MAAM,CAAC;IAC9B,sGAAsG;IACtG,YAAY,EAAE,MAAM,EAAE,CAAC;IACvB;;;;;;OAMG;IACH,gBAAgB,EAAE,aAAa,GAAG,oBAAoB,GAAG,oBAAoB,GAAG,QAAQ,GAAG,MAAM,CAAC;IAClG,uDAAuD;IACvD,WAAW,EAAE;QACX,SAAS,EAAE,OAAO,CAAC;QACnB,mEAAmE;QACnE,MAAM,EAAE,kBAAkB,GAAG,iBAAiB,GAAG,cAAc,GAAG,QAAQ,GAAG,MAAM,CAAC;QACpF,qFAAqF;QACrF,eAAe,CAAC,EAAE,MAAM,CAAC;KAC1B,CAAC;IACF,0CAA0C;IAC1C,aAAa,CAAC,EAAE;QACd,SAAS,EAAE,OAAO,CAAC;QACnB,SAAS,CAAC,EAAE,MAAM,CAAC;QACnB,gBAAgB,CAAC,EAAE,MAAM,EAAE,CAAC;QAC5B,kBAAkB,CAAC,EAAE,MAAM,CAAC;KAC7B,CAAC;IACF,kGAAkG;IAClG,YAAY,CAAC,EAAE;QACb,SAAS,EAAE,OAAO,CAAC;QACnB,qBAAqB,CAAC,EAAE,OAAO,CAAC;KACjC,CAAC;IACF,kFAAkF;IAClF,iBAAiB,CAAC,EAAE;QAClB,2BAA2B,CAAC,EAAE,MAAM,CAAC;QACrC,sBAAsB,CAAC,EAAE,MAAM,CAAC;QAChC,sBAAsB,CAAC,EAAE,qBAAqB,GAAG,eAAe,GAAG,UAAU,CAAC;KAC/E,CAAC;CACH;AAED;;;;;;;GAOG;AACH,MAAM,MAAM,eAAe,GACvB,WAAW,EAAE,GACb,MAAM,GACN;IACE,QAAQ,EAAE,WAAW,EAAE,CAAC;IACxB,MAAM,CAAC,EAAE,MAAM,CAAC;IAChB,KAAK,CAAC,EAAE,GAAG,EAAE,CAAC;IACd,CAAC,GAAG,EAAE,MAAM,GAAG,GAAG,CAAC;CACpB,CAAC;AAEN;;;;GAIG;AACH,MAAM,WAAW,kBAAkB;IACjC,4DAA4D;IAC5D,MAAM,EAAE,eAAe,CAAC;IACxB;;;OAGG;IACH,oBAAoB,CAAC,EAAE,GAAG,EAAE,CAAC;IAC7B,wFAAwF;IACxF,mBAAmB,CAAC,EAAE,MAAM,CAAC;IAC7B,gFAAgF;IAChF,UAAU,CAAC,EAAE,MAAM,CAAC;IACpB,kFAAkF;IAClF,MAAM,CAAC,EAAE,KAAK,CAAC;QACb,IAAI,EAAE,OAAO,GAAG,SAAS,GAAG,MAAM,CAAC;QACnC,IAAI,EAAE,MAAM,CAAC;QACb,OAAO,EAAE,MAAM,CAAC;QAChB,OAAO,CAAC,EAAE,OAAO,CAAC;QAClB,UAAU,CAAC,EAAE,MAAM,CAAC;QACpB,SAAS,CAAC,EAAE,MAAM,CAAC;KACpB,CAAC,CAAC;IACH,4EAA4E;IAC5E,wBAAwB,EAAE,OAAO,CAAC;IAClC,oGAAoG;IACpG,mBAAmB,CAAC,EAAE;QACpB,2BAA2B,CAAC,EAAE,MAAM,CAAC;QACrC,mBAAmB,CAAC,EAAE,MAAM,EAAE,CAAC;QAC/B,oBAAoB,CAAC,EAAE,MAAM,EAAE,CAAC;QAChC,iBAAiB,CAAC,EAAE,MAAM,EAAE,CAAC;QAC7B,yBAAyB,CAAC,EAAE,MAAM,EAAE,CAAC;KACtC,CAAC;IACF,mFAAmF;IACnF,QAAQ,EAAE;QACR,kBAAkB,EAAE,MAAM,CAAC;QAC3B,4BAA4B,EAAE,MAAM,EAAE,CAAC;QACvC,YAAY,EAAE,MAAM,CAAC;QACrB,yBAAyB,EAAE,MAAM,CAAC;QAClC,uBAAuB,EAAE,MAAM,CAAC;QAChC,oBAAoB,CAAC,EAAE,MAAM,CAAC;QAC9B,CAAC,GAAG,EAAE,MAAM,GAAG,OAAO,CAAC;KACxB,CAAC;IACF,mFAAmF;IACnF,QAAQ,CAAC,EAAE,MAAM,CAAC;CACnB;AAED;;;;GAIG;AACH,MAAM,WAAW,kBAAkB;IACjC,kGAAkG;IAClG,mBAAmB,EAAE,MAAM,CAAC;IAC5B,+EAA+E;IAC/E,kBAAkB,EAAE,MAAM,CAAC,MAAM,EAAE,sBAAsB,CAAC,CAAC;IAC3D,mDAAmD;IACnD,aAAa,EAAE;QACb,QAAQ,EAAE,WAAW,GAAG,SAAS,GAAG,QAAQ,CAAC;QAC7C,eAAe,CAAC,EAAE,eAAe,GAAG,OAAO,GAAG,UAAU,GAAG,SAAS,CAAC;KAGtE,CAAC;IACF,sEAAsE;IACtE,iBAAiB,EAAE;QACjB,kBAAkB,EAAE,MAAM,CAAC;QAC3B,mBAAmB,EAAE,MAAM,CAAC;QAC5B,yBAAyB,EAAE,MAAM,CAAC;QAClC,yBAAyB,EAAE,OAAO,CAAC;KACpC,CAAC;IACF,qEAAqE;IACrE,iBAAiB,EAAE;QACjB,mBAAmB,EAAE,MAAM,CAAC;QAC5B,wBAAwB,EAAE,MAAM,GAAG,UAAU,GAAG,cAAc,CAAC;QAC/D,kCAAkC,EAAE,OAAO,CAAC;QAC5C,4BAA4B,CAAC,EAAE,MAAM,CAAC;KACvC,CAAC;IACF,oEAAoE;IACpE,0BAA0B,EAAE;QAC1B,gDAAgD;QAChD,kBAAkB,EAAE,OAAO,CAAC,MAAM,CAAC,qBAAqB,EAAE,MAAM,CAAC,CAAC,CAAC;QACnE,yBAAyB,EAAE,MAAM,CAAC;QAClC,0BAA0B,EAAE,eAAe,GAAG,iBAAiB,GAAG,oBAAoB,CAAC;QACvF,0BAA0B,EAAE,kBAAkB,GAAG,kBAAkB,GAAG,mBAAmB,CAAC;KAC3F,CAAC;IACF,yCAAyC;IACzC,WAAW,EAAE;QACX,aAAa,EAAE,OAAO,CAAC;QACvB,mBAAmB,EAAE,MAAM,CAAC;QAC5B,iBAAiB,CAAC,EAAE,MAAM,CAAC;KAE5B,CAAC;IACF,sCAAsC;IACtC,SAAS,CAAC,EAAE;QACV,oBAAoB,CAAC,EAAE,OAAO,CAAC;QAC/B,4BAA4B,CAAC,EAAE,OAAO,CAAC;QACvC,6BAA6B,CAAC,EAAE,OAAO,CAAC;KACzC,CAAC;IACF;;;;;;;;;;;;;;;;;;OAkBG;IACH,kBAAkB,CAAC,EAAE,MAAM,CAAC,MAAM,EAAE;QAClC,cAAc,CAAC,EAAE,MAAM,EAAE,CAAC;QAC1B,eAAe,CAAC,EAAE,MAAM,EAAE,CAAC;QAC3B,cAAc,CAAC,EAAE,MAAM,CAAC,MAAM,EAAE,MAAM,EAAE,CAAC,CAAC;KAC3C,CAAC,CAAC;CACJ;AAED;;;;;;;;;;;;GAYG;AACH,MAAM,MAAM,sBAAsB,GAAG,CACnC,UAAU,EAAE,QAAQ,CAAC,gBAAgB,CAAC,EACtC,SAAS,EAAE,QAAQ,CAAC,eAAe,CAAC,EACpC,0BAA0B,EAAE,aAAa,CAAC,uBAAuB,CAAC,EAClE,MAAM,EAAE,QAAQ,CAAC,kBAAkB,CAAC,EACpC,oBAAoB,EAAE,CAAC,OAAO,EAAE,MAAM,EAAE,OAAO,CAAC,EAAE,MAAM,KAAK,OAAO,CAAC,MAAM,CAAC,KACzE,OAAO,CAAC,eAAe,CAAC,CAAC;AAE9B;;;;GAIG;AACH,MAAM,MAAM,cAAc,GAAG,CAAC,OAAO,EAAE,MAAM,EAAE,OAAO,CAAC,EAAE,MAAM,KAAK,OAAO,CAAC,MAAM,CAAC,CAAC;AAGpF;;;;;;GAMG;AACH,MAAM,WAAW,sBAAsB;IACrC;;;;;;;;;OASG;IACH,4BAA4B,CAC1B,QAAQ,EAAE,aAAa,CAAC,OAAO,CAAC,EAChC,gBAAgB,EAAE,MAAM,EACxB,SAAS,EAAE,QAAQ,CAAC,eAAe,CAAC,EACpC,yBAAyB,CAAC,EAAE,OAAO,GAClC,OAAO,CAAC;QAAE,eAAe,EAAE,OAAO,EAAE,CAAC;QAAC,kBAAkB,EAAE,MAAM,CAAC;QAAC,eAAe,EAAE,MAAM,CAAC;QAAC,kBAAkB,EAAE,MAAM,CAAA;KAAE,CAAC,CAAC;IAE5H;;;;;;;;OAQG;IACH,mBAAmB,CACjB,OAAO,EAAE,MAAM,GAAG,aAAa,CAAC;QAAE,MAAM,EAAE,MAAM,CAAC;QAAC,OAAO,EAAE,MAAM,CAAC;QAAC,SAAS,CAAC,EAAE,MAAM,CAAA;KAAE,CAAC,EACxF,gBAAgB,EAAE,MAAM,EACxB,SAAS,EAAE,QAAQ,CAAC,eAAe,CAAC,EACpC,yBAAyB,CAAC,EAAE,OAAO,GAClC,OAAO,CAAC;QAAE,OAAO,EAAE,MAAM,CAAC;QAAC,kBAAkB,EAAE,MAAM,CAAC;QAAC,eAAe,EAAE,MAAM,CAAC;QAAC,gBAAgB,CAAC,EAAE,MAAM,EAAE,CAAA;KAAE,CAAC,CAAC;IAElH;;;;;;;;OAQG;IACH,uBAAuB,CAAC,CACtB,OAAO,EAAE,MAAM,EACf,gBAAgB,EAAE,QAAQ,CAAC,sBAAsB,CAAC,EAClD,SAAS,EAAE,QAAQ,CAAC,eAAe,CAAC,GACnC,OAAO,CAAC;QAAE,cAAc,EAAE,MAAM,CAAC;QAAC,eAAe,EAAE,MAAM,CAAC;QAAC,QAAQ,CAAC,EAAE,MAAM,EAAE,CAAC;QAAC,MAAM,CAAC,EAAE,MAAM,EAAE,CAAA;KAAE,CAAC,CAAC;CACzG;AAED;;;;;;;;;;GAUG;AACH,MAAM,WAAW,aAAa;IAC5B;;;;;;;;;;;;;;;;;;OAkBG;IACH,UAAU,CAAC,MAAM,EAAE,kBAAkB,EAAE,SAAS,CAAC,EAAE,sBAAsB,GAAG,OAAO,CAAC,IAAI,CAAC,CAAC;IAE1F;;;;;;;;;;;;;;;;;;;;;;;;OAwBG;IACH,eAAe,CACb,cAAc,EAAE,QAAQ,CAAC,gBAAgB,CAAC,EAC1C,eAAe,EAAE,QAAQ,CAAC,eAAe,CAAC,EAC1C,gBAAgB,CAAC,EAAE,QAAQ,CAAC,sBAAsB,CAAC,EACnD,YAAY,CAAC,EAAE,MAAM,GACpB,OAAO,CAAC,kBAAkB,CAAC,CAAC;IAE/B;;;;;;;;;;;;;;;OAeG;IACH,gBAAgB,CACd,QAAQ,EAAE,QAAQ,CAAC,+BAA+B,CAAC,EACnD,OAAO,EAAE,QAAQ,CAAC,sBAAsB,CAAC,GACxC,OAAO,CAAC,OAAO,CAAC,CAAC;IAEpB;;;;;;;;;;OAUG;IACH,kBAAkB,CAAC,OAAO,EAAE,MAAM,EAAE,OAAO,CAAC,EAAE,MAAM,GAAG,OAAO,CAAC,MAAM,CAAC,CAAC;IAEvE;;;;;;;;;;;;;OAaG;IACH,gBAAgB,CACd,YAAY,EAAE,MAAM,EACpB,gBAAgB,EAAE,sBAAsB,GACvC,OAAO,CAAC,IAAI,CAAC,CAAC;IAEjB;;;;;;;;;;;;;OAaG;IACH,2BAA2B,CACzB,UAAU,EAAE,QAAQ,CAAC,gBAAgB,CAAC,EACtC,eAAe,EAAE,QAAQ,CAAC,eAAe,CAAC,EAC1C,gBAAgB,CAAC,EAAE,QAAQ,CAAC,sBAAsB,CAAC,GAClD,OAAO,CAAC;QACT,OAAO,EAAE,OAAO,CAAC;QACjB,MAAM,EAAE,KAAK,CAAC;YAAE,IAAI,EAAE,OAAO,GAAG,SAAS,CAAC;YAAC,IAAI,EAAE,MAAM,CAAC;YAAC,OAAO,EAAE,MAAM,CAAC;YAAC,UAAU,CAAC,EAAE,MAAM,CAAC;YAAC,SAAS,CAAC,EAAE,MAAM,CAAA;SAAE,CAAC,CAAC;QACrH,eAAe,CAAC,EAAE,MAAM,EAAE,CAAC;KAC5B,CAAC,CAAC;IAEH;;;;;;;;;;OAUG;IACH,UAAU,CAAC,gBAAgB,CAAC,EAAE,MAAM,GAAG,OAAO,CAAC,IAAI,CAAC,CAAC;IAErD;;;;;;;;OAQG;IACH,mBAAmB,IAAI,OAAO,CAAC;QAC7B,uBAAuB,EAAE,MAAM,CAAC;QAChC,yBAAyB,EAAE,MAAM,CAAC;QAClC,UAAU,EAAE;YAAE,IAAI,EAAE,MAAM,CAAC;YAAC,MAAM,EAAE,MAAM,CAAC;YAAC,WAAW,EAAE,MAAM,CAAC;YAAC,OAAO,CAAC,EAAE,MAAM,CAAC;YAAC,kBAAkB,EAAE,MAAM,CAAA;SAAE,CAAC;QAChH,kBAAkB,EAAE;YAAE,UAAU,EAAE,MAAM,CAAC;YAAC,eAAe,CAAC,EAAE,MAAM,CAAA;SAAE,CAAC;QACrE,sBAAsB,EAAE,MAAM,CAAC,MAAM,EAAE;YAAE,KAAK,EAAE,MAAM,CAAC;YAAC,uBAAuB,CAAC,EAAE,MAAM,CAAA;SAAE,CAAC,CAAC;QAC5F,gBAAgB,EAAE,MAAM,CAAC,MAAM,EAAE,MAAM,CAAC,CAAC;QACzC,iBAAiB,EAAE,MAAM,CAAC,MAAM,EAAE;YAAE,KAAK,EAAE,MAAM,CAAC;YAAC,WAAW,EAAE,MAAM,CAAC;YAAC,aAAa,EAAE,MAAM,CAAA;SAAE,CAAC,CAAC;KAClG,CAAC,CAAC;CACJ;AAED;;;;;GAKG;AACH,qBAAa,iBAAkB,SAAQ,KAAK;IAC1C;;;;;OAKG;IACH,SAAgB,IAAI,EAAE,MAAM,CAAC;IAE7B;;;;;;OAMG;IACH,SAAgB,SAAS,CAAC,EAAE,MAAM,CAAC;IAEnC;;;;;;OAMG;IACH,SAAgB,OAAO,CAAC,EAAE,OAAO,CAAC;IAElC;;;;;;OAMG;gBACS,OAAO,EAAE,MAAM,EAAE,IAAI,EAAE,MAAM,EAAE,SAAS,CAAC,EAAE,MAAM,EAAE,OAAO,CAAC,EAAE,OAAO;CAQjF;AACD,YAAY,EAAE,eAAe,EAAE,cAAc,EAAE,CAAC"}