Agent frameworks & infrastructure

Agent memory & context

4,032 agents · most starred first

Agent memory & context, page 14 of 41

crispy-recall★ 15

Local session transcript memory for Claude Code and Codex — search past sessions with FTS5 + semantic vectors.

harnessmith★ 15

Distribute and safely manage a cross-host personal harness and durable work state.

eunomia★ 15

Self-hosted shared memory for AI agents: vaults of people, facts and decisions your agents recall over MCP or REST, with semantic, keyword and graph search.

turnbook★ 15

Local long-term memory for AI coding agents. Keeps every conversation word for word, costs nothing to save, and is shared by all your agents. 91.8% on LongMemEval.

starsea-continuum★ 15

Know when an AI conversation thread is still safe to continue — a deterministic, dependency-free continuity planner.

dolphin★ 14

Open-source persistent memory infrastructure for LLM applications.🐬

memBook★ 14

Progressive memory layer for AI agents with Header+Content indexing, chain recall, and file-first governance.

elefante★ 14

The open, Git-native memory protocol for MCP agents: portable, inspectable, versioned, and human-editable.

everything-gemini★ 14

Your AI remembers your project across sessions and tools. Local-first MCP runtime with persistent memory, shared workflows, and plug-and-play integration.

kairos-context-keeper★ 14

Kairos: An autonomous development supervisor powered by Context Engineering. It provides a living memory and a contextual constitution for your projects to prevent AI context loss.

work-memory-mcp★ 14

Never lose context again - persistent memory management system for AI-powered workflows across multiple tools

botkit-storage-datastore★ 14

Google Cloud Datastore storage module for Botkit

ai-brain-starter★ 14

The operating system for Claude Code. Memory, accountability, journaling, knowledge graphs, pattern recognition — one install, every session compounds.

longbow★ 14

Apache Arrow Flight clustered vector engine for high throughput Agent memory sharing

agentid-protocol★ 14

Open protocol for portable AI agent identity, persistent memory, and real-time activity reporting — MCP + HTTP

sp-context★ 14

Give every AI agent persistent memory of your team's knowledge. No vector DB, no RAG — just Git + BM25 + 114 tokens per session.

python-token-killer★ 14

Minimize LLM tokens from Python objects, code, logs, diffs, and more. Zero deps. Ultra-Lightweight.

TokenPack★ 14

TokenPack packs long documents, codebases, PDFs, and folders into compact, evidence-dense LLM context using local embeddings, evidence scoring, and budget-aware selection.

memory-cloud★ 14

Adaptive memory for AI agents & teams — beyond RAG. Self-hosted MCP server that gets smarter every time you search: hybrid search + a neural memory graph that learns. Works with Claude, ChatGPT & any MCP client.

Midas★ 14

Local-first, eval-first memory for long-horizon AI agents — no LLM at ingest. Python SDK + MCP server with source-traceable recall, belief revision, selective forgetting, and reproducible benchmarks.

SecondBrain★ 14

A persistent MCP memory layer for AI agents. Store knowledge across sessions, retrieve it semantically, and connect any MCP-compatible agent.

open-bridge★ 14

Stop re-explaining your work to your coding agent. Plain markdown and YAML in git, read at session start by Claude Code, Codex and Copilot CLI. MIT.

context-summary★ 14

A Claude.ai Personal Skill that generates a structured, seven-block session handoff so you can continue work in a new conversation without losing context, rediscovering decisions, or re-explaining your project state.

Memories★ 14

A macOS app to browse and manage the memory files Claude Code keeps for your projects.

hypermemory-sdk★ 14

SDK for developing apps on top of Hypermemory

zugamind★ 14

Agents answer. ZugaMind notices. A persistent cognition sidecar that wakes Claude Code, OpenClaw, Codex, or Hermes when something deserves attention — engineered Global Workspace Theory, zero dependencies.

TeamBrain★ 14

The brain is a git repo. Memories are markdown files with YAML front-matter, committed like any other file. Git history is the audit trail; pull requests are the approval gate — nothing is written to the shared brain without human review.

ZeroMem★ 14

Zero-token long-term memory for Claude Code. Indexes your own session transcripts and retrieves evidence deterministically — entity-context graph + temporal hierarchy, no LLM call anywhere in the memory pipeline. It tells you when it does not know.

NexusMem★ 14

Your agent can read git log. It can't read the shell commands you tried that failed. NexusMem indexes shell history with exit codes, git commits and diffs, and project docs into a local SQLite database, and serves back a token-budgeted slice on demand. No cloud, no account.

neo-agent-brain★ 14

Point a cross-model AI engineering team at your own codebases. The Agent OS keeps named AI maintainers' identity, memory and review across labs alive between sessions — the Brain Neo.mjs runs on itself. Multi-tenant: onboarding a codebase is a config entry, not a fork.

gobstopper★ 14

gobstopper is retired and no longer maintained. Source and historical artifacts remain available.

rubi★ 14

Knowledge graph and context manager for AI coding agents: Claude Code, Codex, Gemini CLI, Cursor, Copilot and local models. Code map with PageRank, temporal knowledge graph, hybrid search, early compaction with handoffs.

memex-mvp★ 13

Local-first MCP memory server. Indexes Claude Code/Cowork/Telegram chats; any MCP-compatible AI agent (Cursor, Cline, Continue) can search them.

mda★ 13

Memory system for LLMs that remembers everything you teach it during conversation. No reindexing, no context window limits. CPU by default, GPU optional.

vainplex-openclaw★ 13

Turn OpenClaw from a smart assistant into a self-governing, learning system. Five production plugins: Cortex, Knowledge Engine, Governance, NATS EventStore, Memory.

streamlit-langchain-chatbot-bedrock-redis-memory★ 13

Build a Streamlit app with LangChain and Amazon Bedrock - Use ElastiCache Serverless Redis for chat history, deploy to EKS and manage permissions with EKS Pod Identity

memory-toolkit★ 13

From session to memory to repo docs: carries your work across Claude Code sessions and turns what you learn into docs and rules — for you, your agent, and your team. Handoffs, workstreams, curated markdown.

Soulmem★ 13

a human-like, self-evolving, and able to forget AI memory system dedicated for role-playing task.

mayros★ 13

The governed AI agent platform: semantic memory, intelligent model routing, an MCP server, and 67+ integrations.

memwire★ 13

Open source self-hosted AI memory infrastructure layer

empathy-framework★ 13

Combining a five-level AI framework with git-native memory overcomes session amnesia, enabling anticipation of problems weeks early. Production results: 2000x cost reduction, 10x+ productivity, shifting AI from reactive to predictive partnership through emotional intelligence, tactical empathy, and systems thinking.

openclaw-EverMemOS★ 13

EverMemOS memory backend plugin for OpenClaw — long-term memory, retrieval, and progressive profile building.

dory★ 13

One memory layer for every AI agent. Local-first, markdown source of truth, and CLI/HTTP/MCP native. Your agent forgot who you are. Again. Dory fixes that.

aim★ 13

A.I.M. – Actual Intelligent Memory: zero-token exoskeleton for long-running Gemini CLI/Chat GPT Codex/Claude Code agents. External SQLite memory, GitOps guardrails, DataJack cartridges. Alpha, trenches-first.

shellbrain★ 13

Long-term memory for AI.

synapse-layer★ 13

MCP-native infrastructure for persistent AI agent memory with encrypted storage, semantic retrieval and cross-agent continuity.

Memex★ 13

Memex — Memory Infrastructure for AI Agents.

PersistBench★ 13

[ICML'26] Codebase for PersistBench: When Should Long-Term Memories Be Forgotten by LLMs to benchmark cross-domain leakage and sycophancy in memory augmented LLMs.

nano-brain★ 13

Persistent memory & code intelligence server for AI coding agents. Deploy on VPS for shared memory across machines. Hybrid search (BM25 + pgvector), knowledge graph, impact analysis. 14 MCP tools. Works with Claude Code, OpenCode, Cursor, and any MCP client.

claude-code-memory★ 13

Persistent memory system for Claude Code — make your AI remember everything across sessions. Ready-to-use templates, zero dependencies.

hermes-inner-life★ 13

Your agent remembers how things have been going. Dated state, an evening journal, and dreams at night — for Hermes Agent.

seed-agent★ 13

A self-authoring AI agent. This repository is Cairn: its identity, memory, journal, and blog, edited by the agent itself, in public, commit by commit.

session-recall★ 13

Shared memory for Claude Code, Codex, and Cursor. Your agent recalls decisions from weeks ago — including what the other engines worked out. Local semantic search over all three histories, as MCP tools.

remembrane★ 13

Local-first memory for AI agents: one SQLite file, zero deps. Recency-aware exact recall, conflict detection, time-travel journal, MCP server.

agent-inner-life★ 13

Your agent remembers how things have been going. Dated state, an evening journal, and dreams at night — for Hermes Agent.

azdaja★ 13

Minimal harness-agnostic recursive language model layer — one binary, Python + llm()

Vitrus-Archive★ 13

your glass-box company brain — tells you what it doesn't know (deterministic gap analysis). MIT core · MCP · PGLite→Postgres.

project-brain★ 13

Persistent, navigable memory for AI coding agents — Claude Code, Cursor, Windsurf, local models. Stop re-explaining your projects every session.

daedalus★ 13

DAEDALUS: Bootstrapping Agent Memory from Self-Generated Tasks. Code, paper and blog post.

chronos★ 13

Chronos: Efficient Bolt-On Branching Across Data Stores for Stateful Agentic Applications

Salvor★ 13

Repo-owned engineering memory for coding agents — preserve decisions, failures, domain knowledge, and deferred work across sessions, teams, and models.

Structured-Memory-Engine★ 12

Persistent, self-maintaining memory for AI agents. 1,451 tests. 5ms recall. $0/month forever.

MemState★ 12

Transactional Memory for AI Agents - Keep SQL and Vector DBs in sync with ACID-like guarantees

memoer★ 12

Memory management for LLM Applications

WMB-100K★ 12

WMB-100K — The first 100,000-turn benchmark for AI memory systems

synix★ 12

Memory architectures that evolve with your agent

cecil-protocol★ 12

Persistent memory and identity protocol for AI systems with evidence-aware recall and inspectable memory.

log-file-genius★ 12

Token-efficient log file system - improve memory, reduce AI context bloat

formative-memory★ 12

Memory plugin for OpenClaw. Agent's memories strengthen through use, fade when unused, and consolidate overnight.

memtrace★ 12

LLM-agnostic memory layer for AI agents. No embeddings, no vector DB — just fast, structured, temporal memory that any LLM can consume as plain text.

agentic-vision★ 12

Persistent visual memory for AI agents — capture screenshots, embed with CLIP ViT-B/32, compare, recall. MCP server + Rust core library.

hindsight-benchmarks★ 12

Hindsight Benchmarks Results

cc-habits★ 12

Stop re-correcting your AI. cc-habits learns your coding style once and makes every AI tool remember it. Local-first, no telemetry, open source.

cognitive-os★ 12

cognitive-os · 认知操作系统 — 13 Skills + 4 Rules + 4 SubAgents,把人脑分层记忆结构外化为 AI 可操作的知识体系,以50年认知科学研究为基础

context-runtime★ 12

Context Runtime — a database query planner for LLM context. Decides what a model sees before it answers; plans it, runs it through reused substrate, and learns from the outcome.

agent-brain★ 12

Persistent decision memory for AI coding agents. MCP server for Claude Code, Cursor, Cline.

claude-code-project-brain★ 12

Persistent, navigable memory for Claude Code — stop re-explaining your projects to the AI every session.

bastra-recall★ 12

Local-first memory for AI assistants. Keep preferences, decisions and lessons across sessions and connected tools in readable Markdown files.

chronicle-agent-context-and-memory★ 12

Local-first long-term memory and safer context compression for Hermes Agent — SQLite, no required services.

engrava★ 12

The memory database for AI agents - graph memory, hybrid search, tamper-evident thought/edge journal.

Cortex★ 12

A local, SQLite-backed memory and project-ops layer for coding agents (Claude Code, Codex).

minta★ 12

The context quality layer for AI agents — memory that checks itself: lifecycle governance, calibrated confidence, and staged claim gates. Local-first, MCP 13 tools, DeepSeek Harness plugin.

Awareness-Market★ 12

Local-first AI agent memory — one command, works offline, no account needed. Give your Claude Code, Cursor, Windsurf, OpenClaw agent persistent memory. Markdown storage, hybrid search (FTS5 + embedding), MCP protocol, Web dashboard.

tinymemory★ 12

Agent friendly memory system. Route any memory system into a standardized interface

skillmem★ 12

Self-improving skill memory for coding agents: learn, recall, reinforce, decay — with provenance on every memory and trust only you grant. Local SQLite, no API key, no cloud.

context_architecture_bundle★ 12

The Context Architecture Bundle Repository is a copyable project context layer for multi-agent systems. It helps Codex, Claude, Gemini, and other coding agents restart cleanly, understand the system, change the right files, and close sessions with evidence instead of relying on scattered chat history.

CentralMemoryHub★ 11

Vendor-neutral organizational memory for multi-agent AI systems. MCP + REST API. SOMI category.

FlockMem★ 11

✦ FlockMem is a lightweight, local-first long-term collective memory bus for multi-agents, optimized for edge deployment with a minimal memory footprint under 50MB. It combines text-based, vector similarity, and knowledge graph retrieval with traceable citations.

six-phase-loop★ 11

An architecture for building AI agents that learn, adapt, and persist across sessions.

memory★ 11

Inspectable local project memory for AI coding agents.

atomicmemory-python★ 11

Open-source AtomicMemory Python SDK

codexfi★ 11

Persistent memory for OpenCode AI agents. Embedded LanceDB + Voyage AI embeddings, Bun plugin with CLI and web dashboard.

mainframe★ 11

Agent-agnostic control plane for coding agents: shell policy, durable memory, reviewed Bash tools, and readiness evidence for Pi, Codex, Claude Code, Copilot, Gemini, and custom agents.

VividnessMem-Ai-Roommates★ 11

VividnessMem is built on the idea that memory is not storage but changing salience over time. Vividness provides the core forgetting/reinforcement dynamic; associations enable lateral recall; higher layers like consolidation, dreaming, and regret build increasingly reflective and generative behaviour on top of that foundation.

oasyce_psyche★ 11

AI-first subjectivity kernel for agents — persistent emotional state, relation dynamics, adaptive reply loops. Zero extra LLM calls.

agent_session_manager★ 11

Agent Session Manager - A comprehensive Elixir library for managing AI agent sessions, state persistence, conversation context, and multi-agent orchestration workflows

negentropy★ 11

An agentic system built on a "One Root, Five Wings" architecture, dedicated to combating the entropy production of infomation and forging a continuously self-evolving cognitive framework.

Cathedral★ 11

Persistent memory and identity for AI agents. Free hosted API.

turboagents★ 11

Turbocharge your AI Agents with TurboQuant

LIFE★ 11

LIFE — Persistent identity architecture for AI agents. 16 MCP servers: drives, heart, memory, patterns, journal, genesis, garden, vision, voice, and more. Zero dependencies. Built across 938 conversations.

More in Agent frameworks & infrastructure