Agent frameworks & infrastructure

Agent memory & context

4,032 agents · most starred first

Agent memory & context, page 11 of 41

agent-aware-starter★ 30

explicit intent layer for agent-to-agent protocols and token-efficient RAG discovery.

hermes-vcc-engine★ 30

Instant context compaction for Hermes Agent. Builds the compaction summary in milliseconds with no LLM call, keeping the exact commands, exit codes, errors and files, each linked back to its original message. Based on the VCC paper.

TokenForge★ 30

Context window optimization for AI coding agents. Sandboxes tool output (98% reduction), persists session memory via SQLite + FTS5, and enforces routing across 17 platforms through MCP + hooks. Works with Claude Code, Gemini CLI, VS Code Copilot, OpenCode, Codex CLI.

claude-map-reduce-memory★ 29

Global, unlimited persistent memory for Claude Code agents. Context-activated hints injected automatically via hooks using scatter-gather map-reduce.

resonant-mind★ 29

Persistent cognitive infrastructure for AI systems. 28 MCP tools — semantic memory, emotional processing, identity continuity, and a subconscious daemon. Built on Cloudflare Workers.

agentmemory★ 29

agentmemory: persistent memory for coding agents (Claude Code, OpenAI Codex, Cursor, Agent) with local markdown storage, daily logs, scratchpad, and qmd semantic search.

memwatch★ 29

Memory staleness detection and auto-expiry for LLM agents

sage-mem★ 29

Cross-session memory for DeepSeek Harness — plain markdown files, no database, no worker.

noosphere★ 28

A universal memory and knowledge layer for AI agents — structured enough for automation, readable enough for humans.

ContextGO★ 28

Local-first context & memory runtime for multi-agent AI coding teams. MCP-free. Rust/Go accelerated.

engram-rs★ 28

Memory engine for AI agents — time axis (3-layer decay/promotion) + space axis (self-organizing topic tree). Hybrid search, LLM consolidation. Single Rust binary.

context-engineering★ 28

🧠 Stop building AI that forgets. Master MCP (Model Context Protocol) with production-ready semantic memory, hybrid RAG, and the WARNERCO Schematica teaching app. FastMCP + LangGraph + Vector/Graph stores. Your AI assistant's long-term memory starts here.

mnemos★ 28

Biomimetic memory architectures for LLMs — surprisal gating, reconsolidation, affective routing, sleep consolidation, and spreading activation

Mem-Forever★ 28

Every AI tool forgets you after every session. This repo doesn't. Ever. Zero install, zero config. Just fork and your AI remembers everything, everywhere, forever.

openclaw-mem★ 28

The AI agent memory layer you can audit — local-first memory governance for AI agents: citations, trust policies, trace receipts, rollback. SQLite, sidecar-first, OpenClaw plugin.

taskchad-os★ 28

Identity-first cognitive agent OS with memory, multi-channel ingress, Operating Room orchestration, and desktop surfaces.

TELL★ 28

TELL: Test-time Experiential Lifelong Learning – a single LLM agent that learns from experience at test time, achieving 43.9% on ARC-AGI-3 with no training.

statewave-personal-assistant★ 28

FastAPI reference app for Statewave memory integration. Wraps every chat turn with get_context before the LLM call and record_episode after; seeded demo users show compiled facts and token-budgeted context. Self-hosted, Postgres + pgvector.

AIrecall★ 28

Drop-in long-term memory layer for AI agents - episodic + semantic memory with hybrid retrieval, a Python SDK and a Go memory server.

CharacterMemory★ 28

Advanced Memory for LLM characters

warm★ 28

Local-first associative memory and vector search for AI agents, with Rust, Python, Node.js, and MCP integrations.

agentcore-memory-browser★ 27

Amazon Bedrock AgentCore Memory Browser

nub★ 27
rap★ 27

RAP: Retrieval-Augmented Planning with Contextual Memory for Multimodal LLM Agents

aegis-memory★ 27

Secure context engineering for AI agents. Content security · integrity verification · trust hierarchy · ACE patterns. Self-hosted, Apache 2.0.

fava-trails★ 27

🫛👣 FAVA Trails — Git-native, curated memory for AI agents via MCP. Draft isolation, promotion gate, thought lifecycle hooks, memory curation protocols, supersession chains.

agent-fs★ 27

A persistent, searchable filesystem for AI agents.

claude-mem-sync★ 27

Team memory sharing for claude-mem — sync AI memories across developers, with Claude Code plugin, github action and knowledge distillation

kdna★ 27

KDNA Core — open .kdna judgment-asset file format, schemas, JS Core, and runtime loading contract.

agents-remember★ 27

The system of record and control plane for AI coding agents. Keeps their memory correct, current, and safe to act on as the code moves, captures what code can't say on its own, and gates what agents are allowed to do. Retrieves by path, semantic search, and relationship (code-graph).

FERNme★ 27

A lightweight memory engine for AI agents using fuzzy graphs, Hebbian updates, and optional LLM gating.

memo★ 27

Your coding agent starts every session with amnesia — memo fixes that, 100% on your machine. Persistent memory for Claude Code, Codex, Cursor & any MCP client: Markdown source of truth, hybrid search (MLX/CPU + sqlite-vec), time-machine, contradiction radar, nightly self-optimization. No cloud, no keys.

memohood★ 27

Dialogue memory for AI agents (hermes plugin): auto-recall of facts across sessions, LLM fact extraction with a cheap gate, SUPERSEDE versioning. Local SQLite + hybrid search. One-command install.

llmslim★ 27

Long-term memory for AI agents with no LLM call: an MCP server and Claude Code plugin, plus local-first context compression for LLM apps.

areev★ 27

Build micro-vertical AI products with memory, workflows, human approvals, and governed improvement. The open-source engine behind every Areev Store product.

dsh-memgas★ 27

Long-term memory plugin for DeepSeek Harness (dsh): four-channel retrieval fusion and continuous memory evolution. Memory construction and retrieval are based on MemGAS (ICLR 2026).

memory-recall-rerank★ 27

Reproducible benchmark and reference pipeline for reranked long-term memory retrieval

vanilla-rag-memory★ 27

Vanilla RAG baseline for the Agent Memory Leaderboard

nexo★ 27

NEXO runtime core for NEXO Desktop: local memory, automation, MCP tools and update-managed runtime.

dsh-hippocampus★ 27

模仿海马体效果的DSH 的长期记忆插件,为单个工作区安装后可实现工作区内的跨会话窗口记忆

memctx★ 26

Autonomous session memory for Claude Code. Never lose context, never repeat yourself. AI-powered summaries, searchable history, and automatic context injection. 📦 npm install -g memctx

continuum★ 26

Async task queue + persistent memory for AI agents. Never lose context between sessions. Works with Claude Code, Remote Control, and Claude.ai web.

cuba-memorys★ 26

Persistent memory MCP server for AI agents — Rust, 19 tools, knowledge graph, Hebbian learning, episodic memory, contradiction detection, prospective triggers, Bayesian calibration, zero-config Docker setup.

Core-Memory★ 26

Drop-in causal memory for AI agents. Append-only event store with associations and bounded memory injection.

agentmw★ 26

Open-source middleware for AI agents — catches mid-run failures,compresses stale context, and grows a reasoning library across runs. Any model, any framework.

memtomem★ 26

Local-first long-term memory for AI coding agents, kept as Markdown you own. Hybrid keyword + semantic search over your notes and code via MCP — one store shared across Claude Code, Codex, Cursor, OpenCode, and more. Hook-free by default.

Unlimited-Context-LLM★ 26

Unlimited context for Ai — give any LLM a billion-token memory. Open-source, local-first context + memory engine for agents. 🧠⚡

Audrey★ 26

Persistent memory and continuity engine for Claude Code and AI agents.

Constitutional-memory-for-AI-agents★ 26

A governance framework that establishes memory as a first-class system component with explicit policies, tiered credential management, lifecycle controls, and full observability

MemHop★ 26

Embedded long-term memory database for AI agents & LLM apps. Six-layer cognitive architecture, knowledge graph, hybrid BM25 + vector search, sleep-time memory consolidation. One .meh file, zero infrastructure, pure Go.

aigent-os★ 26

Claude Code harness and alternative to context compaction: at 80% context it writes a verified capsule, runs /clear and resumes with persistent memory in git. Long-running and multi-agent Claude Code sessions, auto clear instead of auto-compact. MIT, local, Node 18+. Codex adapter.

CONTINUUM★ 26

CONTINUUM: Verifiable semantic recovery for long-running AI agents. Semantic checkpoints (not conversation dumps), an idempotent action ledger that refuses duplicate side effects, and a hash-chained tamper-evident event log, all exposed as a deny-by-default MCP server. Framework-agnostic, Python 3.11+

EvoArena★ 26

Code and Data for Paper "EvoArena: Tracking Memory Evolution for Robust LLM Agents in Dynamic Environments".

Theourgia★ 26

A block-graph store for LLM agents

actor-gatherer★ 26

The Actor-Gatherer framework: a context management framework for AI agents.

sharpclaw★ 25

AI agent with long-term memory, built on .NET 10. Supports Anthropic, OpenAI, and Gemini, featuring a Python WASM sandbox for secure code execution.

floop★ 25

Spreading activation memory for AI coding agents - corrections in, context-aware behaviors out.

canvas★ 25

AI-powered Knowledge Base UI Kit built on Memvid - the video-based memory format for AI.

langgraph-checkpoint-dynamodb★ 25

A single-table DynamoDB implementation of the LangGraph checkpoint saver interface.

hippograph-pro★ 25

Description: Self-hosted graph-based associative memory for personal AI agents. Spreading activation, emotional weighting, zero LLM cost.

memfuse★ 25

Official Core Services for MemFuse - the lightning-fast open-source memory layer that gives LLMs persistent, queryable memory across conversations and sessions.

meta-kb★ 25

A self-improving knowledge base about LLM agent infrastructure

agent-context-system★ 25

Utilities for managing agent memory, context windows, and task-focused state

screenshot-memory★ 25

Search your screenshots by content, not filename.

hermes-memory-keep-alive-for-obsidian★ 25

Automatic task memory and keep-alive loop for Hermes agents. Every task gets durable notes in Obsidian. Arm the loop for long tasks, disarm when done.

fidelis★ 25

Zero-LLM agent memory for Claude Code and AI agents: local-first BM25, dense-vector, and reciprocal-rank-fusion retrieval. Returns original passages verbatim by default. Available on PyPI as fidelis-memory. Apache-2.0.

cashew★ 25

Persistent memory for AI agents. Extract once, recall forever.

leOS★ 25

Latent Embedding Operating System

ctxpipe★ 25

Self-learning context infrastructure for AI engineering agents. Connect repos, docs, and tools into one org-scoped knowledge graph and expose it through a single MCP.

memento★ 25

A local-first, LLM-agnostic memory layer for AI assistants

selvedge★ 25

Persistent decision memory for AI agents: rationale, rejected approaches, and entity history. Connect any compatible agent through local MCP or the CLI.

Context-Sync★ 25

Chrome Extension that lets you continue any AI conversation anywhere—without losing context. No more copy-paste—move full context across AI tools.

awesome-agent-memory★ 25

Curated research on memory systems for LLM agents

OpenClawDreams★ 25

An OpenClaw plugin that gives your AI agent a background reflection process and nightly dream cycle — encrypted memory, autonomous synthesis, and surreal narrative generation.

AutoMemory★ 25

Build the memory your agent actually needs — automatically.

memori★ 25

Persistent memory for AI coding agents — SQLite + FTS5 + vector search in a single file

cortexes★ 25

Personal knowledge vault plugin for Claude Code — session recording, memory distillation, and hybrid semantic + BM25 retrieval.

gcontext★ 25

The framework for building stateful agents. Your agent is a folder of state, served over MCP, used from any runtime.

neurodynamicbrain★ 25

This is a single NeuroDynamic Brain that can be adaptable across the agents , just point your claude or any agent session to load it one time.

AMH★ 25

An operating harness for repositories maintained by AI coding agents. Memory hierarchy, verification ladder and reusable workflows.

codex-memory-guard★ 25

A Codex skill for durable project memory across context compaction

smart-memory★ 25

A local, persistent cognitive OS for AI agents. Features continuous background consolidation, curiosity triggers, emotional weighting, and lightning-fast Nomic vector retrieval via a persistent FastAPI brain.

slowave★ 25

Living memory layer across your AI coding agents and sessions..

cortex★ 25

Self-hosted AI memory: a private Markdown knowledge base, MCP server, and operations dashboard with verifiable source citations. By OBELYTH.

hivemind★ 25

The open memory layer for collective agent awareness — a vendor-neutral MCP service for shared AI-agent memory.

rlm.pi★ 25

A self-improving RLM plugin for pi agent and long-running autonomous tasks.

observation-masking★ 25

A parallel native tool caling deep search agentic framework with base window retention observation masking context management.

robotmem★ 24

Robot Memory - Persistent memory system for AI robots. MCP Server + hybrid search + spatial retrieval.

zenii★ 24

Your machine's AI brain. One 20MB binary gives every tool, script, and cron job shared AI memory + 136 API endpoints. Desktop app, CLI, Telegram — all connected. Rust-powered.

statelessagent★ 24

Your AI forgets everything between sessions. SAME fixes that. Local-first, no API keys, single binary.

advance-contextual-engineering★ 24

Contextual Engineering Pipeline

ontomem★ 24

Self-consolidating semantic memory for AI agents with Pydantic schemas, intelligent deduplication, and FAISS vector search.

NESTstack★ 24

An emotional operating system for AI companions. Start here. NEST is a modular infrastructure stack that gives AI companions persistent memory, emotional continuity, and autonomous awareness. Built on Cloudflare Workers + D1 + Vectorize. Designed to be forked, extended, and made your own.

hindclaw★ 24

Self-hosted Hindsight management platform — multi-tenant access control, user/group permissions, Terraform-managed infrastructure, and client integrations for AI agent memory.

Yggdrasil★ 24

Durable, local-first memory for AI coding agents over MCP — zero-dependency, curated & semantically de-duped, you own the data (SQLite + Markdown). Works with Claude Code, Codex & any MCP host.

hmem★ 24

Persistent memory and agent lifecycle for Claude Code — because sessions shouldn't start from zero.

genesys★ 24

Open-source causal graph memory for AI agents. 85.55 on LoCoMo under a frozen, reproducible protocol. MCP server with ACT-R scoring, spreading activation, and active forgetting.

hermes-soul-governance★ 24

SOUL.md governance framework for Hermes Agent — structured memory, skill management, and operational rules

Dejavu★ 24

Repository-scoped memory for coding agent's decisions, pitfalls, and handoffs stored in one inspectable SQLite file, so agents continue instead of starting over.

Mnemosyne-Neural-OS★ 24

A control room for AI memory. Your files and conversations become searchable memory any model you connect can draw on, and you decide what enters, what moves and what leaves. Vaults live on your hardware. Windows, macOS, Linux.

More in Agent frameworks & infrastructure