Agent frameworks & infrastructure

Agent memory & context

4,032 agents · most starred first

Agent memory & context, page 7 of 41

memory-opensource★ 74

Predictive memory layer for AI agents. MongoDB + Qdrant + Neo4j with multi-tier caching, custom schema support & GraphQL. 91% Stanford STARK accuracy, <100ms on-device retrieval

agent-memory-dotnet★ 74

An independent .NET implementation of a Neo4j-backed memory provider for AI agents. Persistent graph-native memory for Microsoft Agent Framework, with GraphRAG interoperability for Neo4j.

Long-Term-Memory-API★ 73

Production-grade API to give your AI agents long-term memory without the boilerplate.

epro-memory★ 73

LLM-powered agent memory with 6-category classification and L0/L1/L2 tiered structure

AI-Context-OS★ 73

Persistent, inspectable memory that grows and refines over time. Built on files. Owned by you.

AuraSDK★ 73

Cognitive memory for AI agents. Pure Rust, <1ms recall, 2.7MB, zero cloud. Patent Pending.

Graft★ 73

Your coding agents forget. Graft doesn't. Persistent local memory that brings back fixes, decisions and project knowledge when they matter — across Claude Code, Codex, ChatGPT and more. Local-first. No SaaS. No API key.

mm★ 73

Fast, multi-modal context for agents

llm-wiki-cli★ 73

Agent-driven proactive memory CLI for AI agents — autonomously recall, maintain, and evolve persistent, source-grounded knowledge across sessions.

hive-mind★ 73

Shared memory for your AI agents: a fact one agent writes on one machine is there for the next agent and the next machine, with no server in between. AGPL-3.0, alpha.

memora★ 73

Memora is the Memory-as-a-Service API for AI apps - durable, multi-tenant, versioned memory with vector + graph recall, callable from any MCP-aware agent.

personal-ai-memory★ 72

A local-first Chrome extension that passively captures ChatGPT, Gemini, Claude, Grok, Perplexity conversations into a private memory graph. Features in-browser Hybrid RAG (Vector + BM25), semantic search, and 100% privacy via WebAssembly and IndexedDB. No servers, no API keys.

ATM-Bench★ 72

ATM-Bench: A benchmark for long-term personalized memory QA spanning ~4 years of multimodal data (images, videos, emails). Features referential queries, evidence-grounded answering, and multi-source reasoning. Paper: "According to Me: Long-Term Personalized Referential Memory QA"

constellation-engine★ 72

Most AI agents forget you the moment the tab closes. Constellation Engine gives them a hippocampus — a living star map with spreading activation, Hebbian writeback, episodic recall, and post-turn consolidation. Local-first, model-agnostic, AGPL.

context-mill★ 72

PostHog context and skill assembly for AI agents and LLM tasks

openclaw_memory_supersystem-v1.0★ 71

AI Agent Memory System inspired by Neuroscience

mcp-neo4j-agent-memory★ 71

Memory management MCP server for AI agents using Neo4j knowledge graphs

soul★ 71

Soul: The Persistent Memory Layer for Multi-Agent Systems. High-performance session orchestration with SQLite-based KV-Cache for MCP. Give your AI agents a soul that never forgets.

summarization-pydantic-ai★ 71

Context Management processor for Pydantic AI agents, providing LLM-powered summarization or zero-cost sliding window trimming to handle infinite/long-running conversations without context overflow. Supports flexible triggers, safe cutoffs, and custom prompts for efficient AI apps.

wife★ 71

Local-first, evidence-aware memory and project continuity for AI coding agents. Checkpoints, context packs and auditable receipts without telemetry.

kenfold★ 71

Self-hosted memory server for AI agents. Share context across Claude Code, Codex, OpenCode, and ChatGPT over MCP.

akm★ 70

Agent Knowledge Manager (akm). A CLI tool to manage AI agent knowledge, memories, skills, and more.

hermes-gbrain-bridge★ 70

Convert Hermes / OpenClaw agent memory (JSONL sessions, MEMORY.md) to markdown for gBrain ingest

ConDB★ 70

ConDB: The KV-Cache Native Context Database

sibyl★ 70

Cross-agent memory for AI coding tools: one self-hostable knowledge graph shared by Claude Code, Codex, OpenCode, Cursor, and the agents you build.

EGC★ 70

EGC gives every AI coding agent the same brain. Shared memory, skills, and live context across Cursor, Claude Code, Copilot, Aider, and 29 AI coding tools with zero configuration. Every tab, terminal, and AI stays automatically synchronized. One brain. Everywhere.

flopwire★ 70

Flopwire connects AI agents through messaging and shared history across harnesses, models, machines, and teammates.

centaur-loop★ 69

半人马环 Centaur Loop:面向 AI Agent 反馈闭环、人类治理和记忆复盘的开源工作台 / Human-governed AI feedback loop workbench.

letta-code-sdk★ 69

The SDK interface to Letta Code. Build deeply personalized agents with persistent memory that learn over time.

memor★ 69

Reproducible Structured Memory for LLMs

ctxvault★ 69

Local-first memory for AI agents: semantic, procedural and episodic vaults you compose, control and query. CLI, MCP, HTTP and Python.

claude-mem-lite★ 69

Persistent long-term memory for Claude Code via MCP — captures coding decisions, bugfixes, and context across sessions. Hybrid FTS5 + TF-IDF search with episode batching. Single SQLite DB, no external services. Alternative to claude-mem with 600x lower cost.

clarilayer★ 69

Memory for your AI work across projects and sessions. Save decisions, rules and lessons with source and applicability. Connect over MCP.

flatbread★ 69

Context alignment, version controlled.

Myco★ 68

The living armor an AI agent inhabits: eternal devouring, eternal evolution, eternal amplification.

memX★ 68

A real-time shared memory layer for multi-agent LLM systems.

memory_agent_hub★ 68

2026 swarm Agent 年,swarm Agent 、Agent team、 ai coding、skill、memory、evolve、agentic RL 等 AI Agent集合

ox★ 68

The hivemind for AI coding agents — persistent team context recorded once and recalled across agents, machines, and teammates.

mnem★ 68

Memory as state, not search — deterministic, dependency-free memory for AI agents. The memory lives in a Markdown file.

cartographer★ 68

A shared, git-backed wiki your team's AI agents maintain but can't break — Claude Code, Codex, OpenCode and any MCP client.

synapse★ 68

Synapse is a local-first personal memory operating system designed for AI agents. It provides durable, typed, and highly reliable long-term memory using only local SQLite — giving agents the ability to remember, update, and reason over facts across sessions with full user control.

friday★ 68

The open-source persistent cognitive memory layer for AI coding agents. Stop your Cursor, Claude & Copilot from having amnesia.

awesome-agent-memory★ 68

A curated list of memory systems, frameworks, benchmarks, and research for AI agents.

mcp-automem★ 67

MCP client for AutoMem — give Claude, Cursor, Codex, and other MCP tools durable graph+vector memory across conversations.

data-structure-protocol★ 67

Graph-based long-term memory skill for AI (LLM) coding agents — faster context, fewer tokens, safer refactors

memory-vault★ 67

A local-first AI memory system with hybrid search, MCP integration, and a knowledge graph.

headroom-zed★ 67

Zed extension for Headroom — context compression for AI agents

kyb★ 67

Git-backed shared memory for AI agent fleets: searchable knowledge base, incident and task tracking, every write versioned and attributed. CLI, HTTP API and MCP server.

Orb★ 66

Self-evolving AI agent framework — wraps Claude Code CLI with persistent memory, multi-profile isolation, and messaging platform integration

FileGram★ 66

FileGram: Grounding Agent Personalization in File-System Behavioral Traces

soul.py★ 66

Persistent identity and memory for any LLM agent — markdown-native, provider-agnostic

git-context-controller★ 66

Structured context management framework for LLM agents. Implements Git-like operations (COMMIT, BRANCH, MERGE) to manage long-horizon agent memory.

langgraph-long-memory★ 66

A detail Implementation of handling long-term memory in Agentic AI

zeph★ 66

A memory-first AI agent that remembers why decisions were made — not just the last message. Runs local (Ollama), cloud (Claude · OpenAI · Gemini), or decentralized TEE. Graph memory, self-learning skills, multi-model routing, sandboxed tools. MCP · ACP · A2A. One Rust binary.

rms-memory-mcp★ 66

Persistent, local, cross-IDE memory for AI agents — markdown source of truth, LanceDB-powered semantic search, zero cloud dependency

engram★ 65

🧠 Hybrid long-term memory plugin for OpenClaw agents — SQLite+FTS5 for structured facts, LanceDB for semantic recall

claude-cortex★ 65

Brain-like memory + security for AI coding agents. Works with Claude Code, Cursor, VS Code, OpenClaw. Now ShieldCortex.

Cortex★ 65

Persistent memory for Claude Code — 36 neuroscience mechanisms, 97 papers. Reproducible via `make reproduce`: LongMemEval-S R@10 98.2% / MRR 0.915 (n=500), LoCoMo R@10 91.5% / MRR 0.805 (n=1982), BEAM-100K retrieval-proxy MRR 0.55. Clean-DB, single-process, production recall path. PostgreSQL + pgvector.

Co-Engram★ 65

Self-evolving team memory

agentkits-memory★ 64

Persistent memory system for AI coding assistants via MCP. Works with Claude Code, Cursor, Copilot, Windsurf, Cline.

ReasoningBank★ 64

An AI agent memory framework that converts an agent’s own interaction traces—both successes and failures—into reusable, high-level reasoning strategies. These strategies are retrieved to guide future decisions, and the loop repeats so the agent self-evolves.

zer0dex★ 64

A local dual-layer memory pattern for AI agents: a compact, human-readable markdown index paired with semantic retrieval from a local vector store, queried before each message. For cross-project recall where flat memory files or vector-only RAG fall short. Local-first. Reference implementation.

Zengram★ 64

A Multi Agent Memory MCP That Connect Agents Across Systems and Machines

agent-context-editor★ 64

Cross-agent plugin for manually excluding and editing AI conversation context, with search, filtering, hide/restore, undo, and original Session preservation.

resonant★ 64

Open-source relational AI framework with identity persistence, memory, and MCP integration. Build relationship-aware AI agents that remember, grow, and maintain continuity. Built on Claude Agent SDK.

redis-llm★ 63

redis-llm integrates LLM with Redis. It can help LLM to access private data, and remember long chat history.

chatgpt-long-term-memory★ 63

The ChatGPT Long Term Memory package is a powerful tool designed to empower your projects with the ability to handle a large number of simultaneous users and external sources.

llm-wiki-kit★ 63

Stop re-explaining your research to your AI agent. Persistent, LLM-maintained wikis that compound over time. Drop PDFs, URLs, YouTube - your agent remembers forever. Based on Karpathy's LLM Wiki pattern.

Helix-AGI★ 63

Helix-AGI is an Agentic AI memory and action system that runs on a continuous daemon pulse with a fully dynamic module context window compilation system. Bicameral release utilizes dual pipelines for speech and action/sense creating an agent with something closer to digital qualia than plain text tool calls and returns.

SeCom★ 63

SECOM: On Memory Construction and Retrieval for Personalized Conversational Agents, ICLR 2025

Mandol★ 63

Mandol: An Agglomerative Agent Memory System (https://arxiv.org/pdf/2606.29778)

aius★ 62

The long-term memory for your Superagents 🥷and LLMs 🤖. Built with GraphRAG, Knowledge graphs and autonomous ai agents

fuzzy-associative-memory★ 62

Fuzzy Logic "Fuzzy Associative Memory" (FAM) for fuzzy control systems, decision-making, artificial intelligence / AI, game agents & bots, etc. Ruby language.

c0ntextKeeper★ 62

Automatic context preservation for Claude Code. Never lose work to compaction again. 7 hooks, 187 semantic patterns, 3 MCP tools.

mom★ 62

MOM is a persistent memory for AI agents — she remembers, so you don't have to_

rg_history★ 62

Agent Skill. The only memory Claude Code needs exists in the file system.

mcp-local-memory★ 62

A lightweight, powerful local memory server for AI agents supporting text, entities, and relations. Enables persistent codebase understanding and user preference management.

memclaw-cross-fleet-gov★ 62

Multi-agent memory governance demo using MemClaw + OpenClaw. Three agents (Sales, Legal, Admin) share one memory backend with hard fleet boundaries enforced at the query layer. Shows scoped writes, blocked recall, cross-fleet synthesis, and conflict detection.

Evolving-LLM-Agent-Memory-Survey★ 62

Paper list for "From Storage to Experience: A Survey on the Evolution of LLM Agent Memory Mechanisms".

Pengenuity★ 61

Pengunuity is an experimental open-source autonomous AI Agent that aims to replicate the structure of human memory.

Zikkaron★ 61

Biologically-inspired persistent memory engine for Claude Code. 26 cognitive subsystems, Hopfield networks, predictive coding, causal discovery, successor representations, all running locally over SQLite.

LatentMem★ 61

LatentMem: Customizing Latent Memory for Multi-Agent Systems

claude-memory-skill★ 61

A minimal, low-friction memory system for Claude Code

MemoryAgent★ 61

具备认知记忆架构的 AI Agent,让 AI 记住你的偏好,越用越懂你 核心特性:四类型记忆系统、跨会话记忆共享、Claude Code 架构、15+ 大模型支持

whittle★ 61

Carves your agent's tool outputs down to what matters. Never cuts what doesn't come back.

HyperSkill★ 61

HyperSkill: Self-Evolving LLM Agents via Hypergraph-Structured Skill Memory

praxis★ 60

Turn scattered knowledge, operational data, and history into source-linked context that your agents can inspect, explain, and reuse.

openexp★ 59

Outsource your understanding. Capture every human-AI decision as a step in a trajectory; grade trajectories retroactively when outcomes land (deal closed, sprint shipped). Hooks + MCP server for Claude Code, Qdrant + FastEmbed local-first.

ctx-open★ 59

CTX is the standard Cognitive Version Control System for AI. Structured working memory for AI agents. CTX preserves goals, tasks, evidence, decisions, conclusions, runbooks, and origins so agents can continue work instead of restarting from scratch.

kyros-ai★ 59

Kyros — The Memory OS for AI Agents Give your AI agents secure, self-correcting, persistent memory in 3 lines of code. Three memory types (episodic, semantic, procedural) with built-in forgetting curves, cryptographic integrity, and automatic contradiction resolution. Model-agnostic REST API with Python and TypeScript SDKs.

aictx★ 59

Repo-local continuity runtime for AI coding agents. Helps them continue from the same inspectable work state instead of starting cold. Stop onboarding your coding agents like rookies every session.

local-cocoa★ 58

A local AI assistant running on your device. It turns your files into actionable memory.

recallium★ 58

Engineering memory for coding agents: the decisions, fixes and constraints behind your code, with who recorded them, the files and the commit, recalled by every agent on the team in Claude Code, Codex, Cursor, Copilot and 60+ MCP clients. Self-hosted community edition; Recallium Cloud at recallium.ai.

EvoEmbedding★ 58

EvoEmbedding: Evolvable Representations for Long-Context Retrieval and Agentic Memory

agenticow★ 58

Git for Agent Memory: Copy-On-Write vector branching for embedded multi-agent memory (83x faster, 3000x smaller snapshots)

hebb-mind★ 58

Neuroscience-inspired memory framework for AI agents

memexa★ 58

A traceable personal memory layer that carries verified state across apps, models, and AI agents.

beardrive★ 58

Google Drive for AI agents — share files with your team, share context across your agents. Open source, no server required.

mcp-memory-server★ 58

Persistent Shared Memory Server for AI Agents via MCP

memoripilot★ 57

Give Copilot a memory! MemoriPilot provides seamless, persistent context management that makes Copilot aware of your project decisions, progress, and architectural patterns - dramatically improving the relevance and quality of AI assistance.

More in Agent frameworks & infrastructure