Engram is a production-ready memory system that gives AI agents persistent, searchable memory across conversations. Think of it as SQLite for agent state - simple, fast, and reliable. Features 馃 Hybrid Memory Retrieval - Combines semantic similarity, recency, confidence, and access patterns 馃攳 Full-Text Search - SQLite FTS5 for fast keyword search 馃幆 Vector Embeddings - Local embeddings with Xenova/all-MiniLM-L6-v2 (~23MB) 馃攼 Secret Detection - Automatic detection and redaction of API keys and credentials 馃Ч Auto-Consolidation - Duplicate detection, contradiction flagging, confidence decay 馃搳 Web Dashboard - Modern React UI for visualizing and managing memories 馃攲 MCP Server - Model Context Protocol integration for Claude Desktop/Code 馃殌 REST API - Full-featured HTTP API with CORS support 馃摝 Zero Config - Works out of the box with sensible defaults
Signals are stable with no major breakout or decline pattern detected in recent weekly snapshots.
Momentum changed by 0% due to steady ecosystem activity.
Operational posture, confidence, and momentum classification in one decision panel.
Momentum and health are visible, but confidence depth is still incomplete.
The final outlook blends raw trend signals with quality, confidence, and operational risk. These are the strongest drivers right now.
Engram: export and report actions do not influence ranking or scoring.