* Get all memories with optional filtering * @param agentId - Optional agent identifier filter * @param category - Optional category filter *

* Get file path for a memory

* Get memory statistics * @returns Promise resolving to statistics object

* Initialize the storage system

* Initialize the file storage system

* Main entry point for the MCP task management server * Uses STDIO transport for communication with MCP clients

* Parse PRD content and extract task information * This is a simplified parser - in a real implementation, you might use NLP or LLM APIs

* Handle file name conflicts by adding numeric suffix

* Scan codebase for relevant files

* Score and rank tasks based on multiple criteria

* Update an existing memory * @param id - Memory identifier * @param updates - Partial memory object with updates * @returns Promise resolvin

@deprecated Use updateTask instead

* Update task dependencies with actual task IDs

* Update task levels for all tasks

* Validate title length (max 50 characters for file naming)

@deprecated Use createTask instead

* Delete all memories for a specific agent * @param agentId - Agent identifier * @returns Promise resolving to number of deleted memories

* Delete all memories for a specific agent (not applicable for simplified schema)

* Delete a memory

@deprecated Use deleteTask instead

@deprecated Use deleteTasksByProject instead

@deprecated Use deleteTasksByProject instead

@deprecated Use deleteTasksByParent instead

@deprecated Use deleteTasksByParent instead

* Get embedding corpus statistics (if supported)

* Get a specific memory by ID

* Get memory statistics

@deprecated Use getTask instead

@deprecated Use getTasks with parentId instead

* Initialize storage by validating working directory and loading data from file

* Search memories by text content

* Update an existing memory

@deprecated Use updateTask instead