memorycrystal
Active·★ 12·MIT·Updated 2026-10-22
Memory Crystal provides persistent memory for AI agents, solving the problem of AI forgetting information between sessions. It captures conversations, distills durable knowledge into long-term memory, and recalls the right context when an agent needs it.
#Persistent Memory#AI Agent Memory#Context Management#Vector Search#Knowledge Base#LLM Integration#Conversational AI#Memory Stores
01
Features
01Two-layer memory architecture including Short-term (STM) for recent messages and Long-term (LTM) for durable knowledge.
02Five specialized memory stores (sensory, episodic, semantic, procedural, prospective) for different types of information.
03Adaptive recall with six automatically selected modes to prioritize broad recall, decisions, projects, people, workflows, or conversations.
04First-class immutable knowledge bases for structured reference data, policies, and runbooks, separate from conversational memory.
05Comprehensive set of memory management tools and an authenticated HTTP API for core operations like capture, recall, and search.
02
Compatibility
Claude Code
Integration
Verified via docs
Codex CLI
Integration
Verified via docs
Factory Droid
Integration
Verified via docs
OpenClaw
Integration
Verified via docs
MCP Host
API
Verified via docs
Cloud
Deployment
Verified via docs
03
Quick start
1
$ curl -fsSL https://memorycrystal.ai/crystal | bash
04
Use cases
↳Enabling AI agents to maintain continuity and recall past decisions across multiple sessions.
↳Integrating persistent memory with various AI tools and platforms like Claude Code, Codex CLI, and OpenClaw.
↳Storing and retrieving structured reference data, policies, and runbooks via immutable knowledge bases.
↳Performing decision archaeology to understand why past choices were made and tracing memories back to their source.
↳Providing pre-flight checks and guardrails for risky AI actions based on learned procedural memory.
05
Alternatives
okf-agent-memory★ 827
Git-native persistent memory for AI coding agents. Implements Google OKF v0.2 with sub-300µs in-memory BM25 search, embedded MCP server, and progressive disclosure. Slashes token bloat by 80% with zero external databases or dependencies. Built in pure Go.
Waggle-mcp★ 44
MCP server for external memory layer for AI agents + more . Download from pypi , and get started
fava-trails★ 26
🫛👣 FAVA Trails — Git-native, curated memory for AI agents via MCP. Draft isolation, promotion gate, thought lifecycle hooks, memory curation protocols, supersession chains.
mnemosyne★ 151
Evidence-backed memory for AI agents: local-first recall, temporal history, tested skills, and gradual migration.
rlm-claude★ 39
Recursive Language Models for Claude Code - Infinite memory solution inspired by MIT CSAIL paper
projectmem★ 868
Local-first AI coding memory for AI agents. Records issues, attempts, fixes and decisions, then warns your agent before it repeats an approach that already failed. Native MCP server for Claude Code, Cursor, Antigravity and Codex. 100% local, no cloud, no telemetry. MIT.
Comments
Log in to leave a comment
No comments yet. Be the first!