Module memory 

Auto Memory system for MatrixCode.

This module implements automatic memory accumulation inspired by Claude Code. It captures user preferences, project decisions, key findings, and solutions across sessions, providing persistent context that survives conversation compression.

Structs

AiMemoryConfig

Configuration for AI-enhanced memory processing.

AiMemoryExtractor

AI-based memory extractor using a fast/cheap model.

AiMemoryProcessor

AI-enhanced memory processor. Provides advanced memory operations using AI.

AutoMemory

Manager for automatic memory accumulation.

MemoryConfig

Configuration for the memory system.

MemoryConflictResult

Result of AI conflict detection.

MemoryEntry

A single memory entry.

MemoryFileLock

File lock for preventing concurrent access to memory storage. Uses a simple lock file approach (.lock) with atomic operations.

MemoryMergeResult

Result of AI memory merge.

MemoryQualityResult

Result of AI quality assessment.

MemoryStatistics

Statistics about memory collection.

MemoryStorage

Storage for memory files (global and project-level) with file locking.

MemorySummaryResult

Result of AI memory summarization.

RewindResult

Result of a rewind/summarize operation.

SemanticUtils

Cosine similarity calculation utility. Used for vector-based semantic search when embedding API is available.

TfIdfSearch

Semantic search without AI (using TF-IDF like approach).

Enums

AiDetectionMode

AI memory detection mode. Controls whether AI is used for memory category detection.

AiKeywordMode

AI keyword extraction mode.

MemoryCategory

Categories for memory entries.

Constants

CONFLICT_OVERLAY_THRESHOLD

Topic overlap threshold for conflict detection.

CONFLICT_OVERLAY_THRESHOLD_WITH_SIGNAL

Lower topic overlap threshold when change signal is present.

CONTEXT_IMPORTANCE_WEIGHT

Weight for importance in contextual summary (1.0 - CONTEXT_RELEVANCE_WEIGHT).

CONTEXT_RELEVANCE_WEIGHT

Weight for relevance in contextual summary (relevance vs importance trade-off).

DEFAULT_FAST_MODEL

Default fast model for AI memory extraction.

DEFAULT_IMPORTANCE_DECISION

Default importance scores by category. Lower values allow for gradual importance growth through references.

DEFAULT_IMPORTANCE_FINDING

DEFAULT_IMPORTANCE_PREF

DEFAULT_IMPORTANCE_SOLUTION

DEFAULT_IMPORTANCE_STRUCTURE

DEFAULT_IMPORTANCE_TECH

DEFAULT_MEMORY_EXTRACTOR_MODEL

Default model for cost-effective memory extraction.

IMPORTANCE_STAR_THRESHOLD

Importance threshold for displaying star marker (⭐).

MAX_DETECTED_ENTRIES

Maximum entries to return from detection (to avoid overwhelming).

MAX_DISPLAY_LENGTH

Maximum length for display (shorter for terminal readability).

MAX_IMPORTANCE_CEILING

Maximum importance score ceiling (entries cannot exceed this).

MAX_MEMORY_CONTENT_LENGTH

Maximum length for memory content before truncation.

MIN_KEYWORDS_FOR_AI_FALLBACK

Minimum keywords threshold for triggering AI fallback. If rule-based extraction produces fewer keywords than this, AI is used.

MIN_MEMORY_CONTENT_LENGTH

Minimum content length for memory detection (to avoid capturing too generic content). Increased to 20 to filter out short fragments.

MIN_SIMILARITY_LENGTH

Minimum content length for similarity check (to avoid short words matching everything).

SIMILARITY_THRESHOLD

Similarity threshold for considering entries as duplicates (0.0-1.0). Higher value (0.85) reduces duplicate detection false negatives.

Traits

MemoryExtractor

Trait for memory extraction implementations.

Functions

calculate_similarity

Calculate word-based similarity between two strings (Jaccard coefficient). Returns a value between 0.0 (no similarity) and 1.0 (identical words). This is the public version for external use.

detect_memories_fallback

Detect potential memory entries from conversation content. This is the fallback method using rule-based detection (no AI). For AI-based extraction, use AiMemoryExtractor.

detect_memories_from_text

Detect memories from text using the rule-based fallback method. This is kept for backward compatibility and for cases where AI is unavailable.

detect_memories_smart

Smart memory detection that chooses the best method based on environment. Uses AI when MEMORY_AI_DETECTION=always and extractor is provided. Otherwise falls back to rule-based detection.

detect_memories_with_ai

Detect memories asynchronously using AI extractor. Falls back to rule-based detection if AI fails or is unavailable.

extract_context_keywords

Extract meaningful keywords from conversation context. Filters out common stop words and short tokens. Public for external use (e.g., TUI keyword display).

extract_keywords_hybrid

Extract keywords from context with hybrid approach.

extract_keywords_with_ai

Extract keywords from context using AI (for context-aware memory retrieval).

summarize_up_to

Summarize messages up to a specific index, keeping recent ones. Returns the new message list with summary + kept messages.