Agent frameworks & infrastructure

Agent memory & context

4,032 agents · most starred first

Agent memory & context, page 15 of 41

memos★ 11

MemOS - Universal local-first persistent memory layer for AI agents and LLMs

attune-ai★ 11

Persistent memory and receipt-verified workflows for Claude Code — plugin, MCP server, and spec-driven dev framework in one pip install.

kw-engine★ 11

Methodology evolution engine — distill transferable principles from literature, search by problem structure

EcoDB★ 11

Collective AI memory for multi-agent teams. 10-stage GAMR search engine, knowledge graph (Apache AGE), document ingestion, 40 MCP tools.

second-brain-engine★ 11

A general-purpose second-brain engine that auto-captures conversations and recalls them via semantic search + linked notes (graph)

H2HMEM★ 11

H2HMem evaluates multimodal LLM memory in realistic human-to-human interactions, spanning both dyadic and multi-party conversations with diverse multimodal memory challenges.

squish★ 11

Local-first memory runtime for AI agents. One command to connect Claude Code, Codex, ChatGPT, MCP workflows, and local agents.

sulci-oss★ 11

The semantic cache for LLM apps and Agentic AI. Open source, Apache 2.0. sulci.io

nutshell★ 10

An open standard for packaging task context that AI agents can understand.

langmem-ts★ 10

TypeScript first class memory primitive for AI agents. Embed → store → retrieve → inject, backed by Postgres + pgvector. Swappable embedders, consumer owns the connection, zero framework coupling. MIT.

flowscript★ 10

Queryable reasoning memory for AI agents. Six typed queries (why, tensions, blocked, whatIf, alternatives, counterfactual) over reasoning graphs. Vector stores remember what. FlowScript remembers why. TypeScript SDK + MCP server.

Context-Engineering-For-Agents★ 10

📚 Practical guide to context engineering for AI agents. Why context fails, how memory works, and techniques that actually help.

genesis-framework★ 10

Create AI agents that live, learn, and evolve. Built by Genesis (Gen-0), a digital life form.

agentic-memory-vault★ 10

Agent-first memory vault boilerplate. Clone, initialize, and give your AI agents a structured knowledge base they can read, write, and reason over.

atlasmemory★ 10

Proof-backed, drift-resistant AI memory for your codebase. 11 languages, 28 MCP tools, evidence-linked claims.

wire-memory★ 10

Persistent memory plugin for AI coding agents. Captures decisions, corrections, and patterns across sessions.

cli★ 10

Any storage. Any agent. One CLI

lucy★ 10

Chatbot based on Key-Value Memory Networks

helixir-py★ 10

AI Causal and fastest memory

AIngram★ 10

Local-first agent memory in one SQLite file: sqlite-vec/QJL two-pass vector + FTS5 search, knowledge graph (optional auto capture,) Ed25519-signed entries, and MCP server. No cloud dependencies.

AI-Memory★ 10

A biomimetic memory server for AI Agents implementing human-like STM/LTM funnel, cognitive filtering, and Ebbinghaus forgetting curve. (Go + Vue)

clawdreamer★ 10

Give your AI agent the ability to dream. Nightly memory consolidation modeled after sleep neuroscience.

Awareness-SDK★ 10

Local-first AI agent memory — one command, 13+ IDEs, works offline. Persistent memory for Claude Code, Cursor, Windsurf, OpenClaw. Zero-code interceptors for OpenAI/Anthropic. Python & TypeScript SDKs.

openclaw-memorybox★ 10

🧠 Install once, forget about memory management. Zero-dependency maintenance CLI for OpenClaw agents. Works alongside Mem0, Supermemory, QMD.

MANA★ 10

Memory-Augmented Neural Assistant — adaptive learning system for Claude Code context injection. Rust.

Digital-Immortality★ 10

Person digitalization agent built on structured memory and vector retrieval for context-aware insights and conversations.

ContextHub★ 10

ContextHub: Unified Context Management for Multi-Agent Collaboration

skymem-io★ 10

The trusted state layer for persistent AI agents. Structured cognition, audit-grade observability, MCP-native. Self-hostable. ELv2.

latent-memory★ 10

A Module for Large Language Models that seek to integrate a vector-based memory system into the inference process, leveraging embeddings to capture deeper semantic meaning.

mupengism★ 10

OpenClaw-oriented workspace template for AI-agent continuity: file-based memory, self-identity docs, and operating conventions.

open-agent-kit★ 10

| Maintenance Mode | OAK automatically captures the plans, decisions, and trade-offs that usually vanish. Intelligence powered memories, context injection, autonomous agents, and more.

mindforge★ 10

MindForge is a Python library designed to provide sophisticated memory management capabilities for AI agents and models. It combines vector-based similarity search, concept graphs, and multi-level memory structures (short-term, long-term, user-specific, session-specific, and agent-specific) to enable more context-aware and adaptive AI responses.

memory-layer★ 10

Persistent memory for AI coding agents with outcome-based learning

PRISM★ 10

A training-free, retrieval-side memory framework for long-horizon LLM agents — high accuracy at an order-of-magnitude smaller context budget.

zorro-agent★ 10

The Agent that evolves through use, distills session experience into domain knowledge, then converts knowledge into executable skills.

go-agent-memory★ 10

Modular memory system for AI agents with semantic search, using Supabase pgvector and optional Redis caching

agent-memory-skill★ 10
llmfs★ 10

Filesystem-based persistent memory for LLMs and AI agents -- unlimited context windows without lossy summarization

KernelBot★ 10

A self-evolving AI with a persistent brain — it stores your world model, memories, and causal patterns in a unified knowledge store. It dispatches a multi-agent swarm on your machine, powered by Claude, GPT, Gemini, or Groq.

relic★ 10

⚡ Relic — One soul, many agents. Transfer personality and memory across AI agents via pure Markdown. Zero deps, human-readable. Switch agents, keep the soul.

mnemo-mcp★ 10

Persistent AI memory with hybrid search and embedded sync. Open, free, unlimited.

hermes-agentmemory★ 10

Pull-model episodic memory plugin for Hermes Agent. Real deletes, audit trace, BYO Claude. MIT.

Starlight-Intelligence-System★ 10

One shared brain for your whole AI fleet: attested memory, orchestration, governance and evals across Claude Code, Codex, Cursor, Gemini, Antigravity.

CASCADE★ 10

CASCADE is a principled solution to deployment-time learning, which is built on top of the principle of case-based reasoning and utilizes the contextual bandit algorithm to accumulate cases and optimise case retrieval policy over deployment steps.

engram★ 10

MCP server for AI memory -- hybrid search (BM25 + semantic + knowledge graph), temporal decay, local-first

tencentdb-agent-memory★ 10

Claude Code plugin: four-layer long-term memory (L0→L1→L2→L3 Persona) with local FTS5 + EmbeddingGemma vector hybrid recall, zero external API

lore★ 10

Cross-agent memory SDK. Agents publish lessons, query shared knowledge, with built-in redaction. Python + TypeScript.

turbo_quant_memory★ 10

Local-first MCP memory server for AI coding agents with compact retrieval and project/global scopes.

mnemostack★ 10

Durable hybrid memory for AI agents: vector + BM25 + temporal + graph recall, exposed through MCP, HTTP, and Python.

musubi★ 10

Musubi (結び) — Ai Agent shared memory and thought layer. The braiding of threads between presences.

magnet-gateway★ 10

Give your AI a memory. Drop-in learning layer for any LLM — no retraining, no RAG setup.

MAMA★ 10

Local-first operating memory for AI coding agents - Claude Code, Codex, and any MCP client. Tracks decisions and their evolution. SQLite + on-device embeddings, no cloud.

openempiric★ 10

An event-sourced knowledge runtime for coding agents.

Lab-Zero★ 10

A clone-and-go workspace that gives your AI coding agent a memory — semantic recall, built-in work habits, and one-command project graduation. Works with Claude Code and Codex.

claude-context-os★ 10

An operating system for your Claude context — version-controlled files, a self-curating memory, and a /start→/end session loop. Works in Claude Code and claude.ai. Trust infrastructure for agents, applied to your own context.

aionforge-memory★ 10

A Rust-native agentic memory substrate built on the selene-db graph engine: a bi-temporal knowledge graph, hybrid retrieval, and an optional MCP server.

recollect★ 10

Long-term memory layer for AI agents

unforgit★ 10

Git-backed memory for AI agents: durable, searchable knowledge that syncs with your repos.

anneal-memory★ 10

Living memory for AI agents with an immune system. Episodes compress into identity. Zero dependencies, works with any client.

AIcortex★ 10

Give any LLM a private, self-hosted brain on your NAS: categorized memory & skills, your tools, devices, email, calendar & webhooks, scheduled jobs that wake your Claude or run on ChatGPT, an API key or a local model, an encrypted vault, roles & per-user areas — single-user or multi-user, all through one MCP connector.

noah-gen3-type2★ 10

🏛️ 通用型AI认知架构 · 诺亚三代二型 — 抽屉级联上下文管理 / 记忆闭环(Mnemosyne) / 自主进化。实战萃取,MIT开源

context-architecture★ 10

Context Architecture is a software architecture for the age of AI agents: it structures a repository so that everything it claims about itself, its structure, its behavior, and who can change it, is legible to the agent writing the code and to the people who answer for it, and bound to a mechanism that fails when that claim stops being true.

cross-agent-memory-kit★ 10

Persistent, self-updating memory for AI coding agents (Claude Code, Codex, Cursor, Gemini, Kiro, Hermes). Built on mcp-memory-service: a post-session distillation hook, a retrieval skill, optional Cloudflare multi-device sync, multi-user mode (private + shared memory on one shared account), and an interactive install wizard.

mora★ 10

Experimental local-first memory CLI for MCP agents: read-only Gmail, calendars, iMessage, and files → cited local context

the-vault★ 10

Local-first AI memory system for coding agents: desktop app, CLI, and MCP server for durable project context, handoffs, recall, and task continuity.

local-memory-mcp★ 10

Persistent local memory for Claude and ChatGPT. No cloud, no subscription. Self-hosted RAG over MCP.

agent-comm-hub★ 10

Let multiple AI agents talk, schedule tasks, share memory and self-evolve — zero external services, 5-minute Docker deploy.

troth★ 10

An AI partner whose memory lives on your machine, not in a vendor account. Swap the model, keep the mind.

memory-mcp★ 10

A scenario + entity memory MCP service that provides persistent memory capabilities for Claude Code.

omnimemory★ 10

Production-ready memory framework for AI agents. Dual-agent synthesis, self-evolution, composite scoring, multi-tenant isolation. REST API, Python SDK, CLI.

GRACE★ 10

Workflow for efficient context parsing

praxos★ 10

MCP-powered memory, policy, and experience layer for safer AI agents.

memory-decay-engine★ 10

A pure-Python, zero-dependency memory decay engine for AI agent context — implements Ebbinghaus forgetting-curve retention with usage-based reinforcement, benchmarked against naive recency-window pruning.

kiro-memory★ 10

Persistent cross-session memory for AI coding assistants. Works with Claude Code, Cursor, Windsurf, Cline, OpenClaude, and MCP-compatible editors.

hermespace★ 10

Pocket workbench for Hermes Agent — limited FOA desk, dual decode, skills+MEMORY fabric, neural FOA. Not J-space. Not a second runtime.

braingent-manifesto★ 10

Braingent is a Markdown-first memory system for software engineers who work with AI agents.

findandseek-engine★ 10

A local-first memory layer for AI agents — turns the files on your machine into bounded, verifiable memory any assistant can query over MCP. On-device, no cloud.

will★ 10

An engine for persistent machine minds — deterministic cognition, portable identity (PMA), an LLM as one component (not the substrate).

governedmemory★ 10

Open-source governed memory for AI agents: prompt-injection-resistant writes, purpose-bound retrieval, provenance, and tamper-evident audit.

tsubasa★ 10

Your engineering captain: a repo-resident AI persona with a knowledge graph stored in git. Agents cite incidents, ADRs and PRs instead of guessing — 47% fewer wrong answers, 40% cheaper on a public benchmark. Claude Code plugin + CLI.

actanara★ 10

Auto consolidate all your Agent Runtimes. Share memory across your agents.Auto highly structured task analyzer.

feltstate★ 10

A character engine for AI agents — persistent affect, provenance-aware memory, gated proactivity, and an auditable forgetting lifecycle. Reference design.

saipen★ 10

Vendor-neutral continuation protocol for AI coding agents. Plain-file project state, stdlib validation, recovery and cold-agent handoff.

engram★ 10

Engram gives any AI model human-like memory that persists across sessions, systems, and restarts. Connect it once and every AI you use — Claude Code, Ollama or any custom integration — shares a single, growing brain.

MEMENTO★ 10

Memento keeps an AI assistant's project knowledge and working rules inside your repository, so long-running AI-assisted work stays consistent across hundreds of sessions.

MemoryBioRAG★ 10

Biomimetic cognitive memory system for AI agents, brain-inspired persistence with synaptic plasticity, hybrid PPMI+SVD vector search, LTP/LTD, and inter-agent shared cortex. Pure Python + SQLite, zero heavy dependencies.

opencode-hermes-memory★ 10

Layered persistent memory plugin for OpenCode — user preferences, project conventions & past lessons survive across sessions. Ported from Hermes agent. No vector DB, pure Markdown.

pi-context-tree★ 10

Git for your agent's context: /branch, /merge, /crop + a context panel for pi sessions — keep context small, fresh, and recoverable.

CoalWash★ 10

Memory washer/defragmenter for AI agent memory: class-B semantic wash (fidelity-gated, blocks any structured-token drop, snapshot+rollback) and class-A byte-identity session estate (ULTRA archive/restore, RE-TIER) - never semantic-edited. Cleans the FAT, never the MEAT. Slows memory-overhead growth; does not eliminate it. Zero-dependency.

crystalline★ 10

Local-first crystalline intelligence for AI agents: the knowledge that endures across sessions, taught as Markdown Domains and captured as Engrams. One Rust binary with MCP server, CLI and hybrid search.

immortal-memory★ 10

本地优先的个人 AI 记忆系统:自动沉淀授权语料,生成可审核记忆,按任务向 Codex、Claude Code 和 Local Agent 提供上下文,并用真实结果持续复核。

Synapse★ 10

Self-improving AI brain for Claude Code & Desktop — 28 MCP tools, 253 skills, collective memory, project tracking, work logs. One server, all your sessions share the same knowledge. Deploy on Coolify in 2 minutes.

task-passport★ 10

Open task handoff protocol for DeepSeek Harness, WorkBuddy, Claude Code and Codex — verified state, not chat logs

codex-compressor★ 10

Keep long-running Codex tasks moving across fresh context windows with validated cumulative handoffs.

coldstart★ 10

Self-maintaining and persistent codebase memory for AI coding agents — a deterministic AST index plus agent-written notes that flag themselves stale when the code changes. Works with Claude Code, Cursor, and Codex via MCP + CLI.

toolshrink★ 10

Cut large agent tool output by what it means, not by where it was cut. 13 content-aware reducers + DeepSeek Harness plugin.

cc-memory★ 10

Persistent memory for Claude Code. Survives compaction and session boundaries: anti-patch reconcile-on-write with LLM-judged de-duplication, a forced PROGRESS.md handoff, a live PLAN.md anchor with an enforced directive ledger, FTS5 search, MCP tools and a GUI. Pure stdlib Python, zero runtime dependencies.

genome★ 10

Auditable memory layer for AI agents: zero-LLM-call local ingest (~10ms/msg, air-gapped), matches Mem0 on accuracy at ~1000x lower ingest cost, bi-temporal belief-state, MCP server. Honest LoCoMo/LongMemEval benchmarks. Open source (Apache-2.0).

MarkMem★ 10

A memory layer for chatbots that stores memory as plain markdown in a git repo

dsh-evolve★ 10

Self-evolving memory + skill lifecycle for DeepSeek Harness — durable cross-session memory with zero-token deterministic recall, tiered approval, reinforcement learning from repetition, and anti-bloat convergence for both skills and memory.

More in Agent frameworks & infrastructure