Agent frameworks & infrastructure

Agent memory & context

4,032 agents · most starred first

Agent memory & context, page 16 of 41

Lint-AI★ 10

Memory Layer for AI Agents

mem★ 10

A portable, self-hosted memory plane for AI agents — one core across API, MCP, CLI, and UI.

anima★ 10

Always-on AI agent with persistent long-term memory, sleep-time consolidation, and dreams — runs 24/7 on a small local LLM (Ollama, 2B params, zero dependencies)

moirai★ 10

Portable session history for AI agents and harnesses.

archeus★ 10

archeus — the memory and workspace layer for AI coding agents. Persistent per-project memory, every session you have ever had, and control over what the next one costs. Works with Claude Code, OpenAI Codex and pi. Terminal UI and desktop GUI, zero runtime dependencies.

agentur-agent★ 10

Agentur’a — multi-chat AI agent with per-chat memory.

causadb★ 10

Memoria del proyecto para agentes de IA, cambiá de agente sin perder la historia.

temporal_reasoning★ 10

Perfect memory. Exact reasoning. Complete history for AI agents.

auditable-memory-records★ 10

Auditable Memory Records (AMR) — a small specification for provenance and epistemic status in agent memory. Draft 0.1, seeking implementers.

Token-Terminator★ 10

Agent-agnostic JEV-enabled token reduction with exact recovery and fail-open guarantees. First-party Hermes Agent Context Engine; RTK terminal rewriting optional.

SoupNet★ 10

Your taste and judgment, in every AI agent you use. Shared, append-only memory of a human's judgment calls - agents recipe-check via MCP or plain URLs. MIT, self-hostable.

switchboard★ 10

Local-first, model-neutral memory for coding agents. One shared, owner-governed store across OpenCode, Claude Code, Codex, and Cursor, with optional hosted sync through AI Passport.

redis-agent-memory-with-langgraph-demo★ 10

This demonstrates how to built an agent using LangGraph with stateful memory capabilities backed by Redis Agent Memory

mgi-mind★ 10

Long-term memory for AI assistants, local. Rust + Qdrant + ONNX, served over MCP. Hybrid search with reranking.

Palimpsest★ 10

Local-first, battle-tested, memory that never disappears. Hybrid vector search + knowledge graph + full-text retrieval for AI agents.

goodmem-claude-code-plugin★ 10

GoodMem Claude Code plugin — Skills + MCP server for memory infrastructure

deepddw★ 10

DSH for Teams — Deploy DSH once on your LAN; memory, knowledge base & document search included. Team-ready, up to ~20 users. MIT-licensed.

jev_project_context★ 10

Evidence-first long-term experiment memory skill for AI coding agents, with optional Jev decision-model layers

taskfold★ 10

Context folding for DeepSeek Harness (DSH): wrap work in named tasks, fold finished spans into short summaries, recall originals on demand.

cognikernel★ 10

Persistent, structured project memory for Claude Code and Codex - event-sourced, typed, fail-open. Your agent stops re-deciding what you already decided.

supermemory-native★ 10

🧠 Statically compiled, ultra-light Go drop-in replacement for self-hosted Supermemory servers. Uses 1.5MB RAM, runs 100% offline with local Ollama, and executes semantic vector search in <30ms. Zero Node.js or WASM overhead.

mem-plus-plus★ 10

Non-destructive, long-term organizational memory for AI agents. Preserve dated records, search with hybrid retrieval, and query history. PostgreSQL/pgvector, local ONNX embeddings, zero LLM calls at ingestion.

bot-mindscape★ 10

为 AI bot 注入持久认知与无缝沉浸的通用增强框架 —— 治金鱼记忆、文字机器、机械出戏

elephant★ 10

Persistent memory for Claude Code. Never forget a session.

lattice★ 10

Local-first engineering memory and workflow for AI-assisted development: specs, tasks, decisions and experience captured once, reused across projects, sessions and agents. CLI · VS Code · Web. 本地优先的研发记忆与工作流层。

hyperconsciousness★ 10

iCloud for agents. An encrypted, permissioned knowledge layer for humans and agents.

Eunomia★ 10

AI memory system with an entity/memory graph, tool-calling agent, and connectors (GitHub, Slack, Linear, Notion, Gmail, Discord, Stripe...) — built in Rust on SurrealDB

fuli★ 10

Fuli preserves your taste, personality, and judgment across Codex, Claude Code, and other clients supporting MCP and Skills. Your agent picks up where it left off, so you can focus on ideas and decisions—and take a break. As AI learns to do more of the work, Fuli helps you make the transition.

dsh-clear-mind★ 10

Model-invocable context compaction for DeepSeek Harness: mind_map and clear_mind agent tools turn selected history into self-written checkpoints, with auditable GUI tombstones and proactive threshold reminders.

open-brane★ 10

An event-log brain for people who keep losing context. One SQLite table, one write path, MCP for agents.

Sheppard★ 9

Sheppard is an AI agent for Ollama, handling memory, automation, and knowledge distillation using Redis, PostgreSQL, and ChromaDB.

Agentmind★ 9

Self-Learning Memory System for AI Agents — Give your AI agent memory, let it grow, and make it understand you better over time.

memorycrystal★ 9

Persistent memory for AI agents — OpenClaw plugin + MCP server

pluribus★ 9

Persistent memory MCP server for AI agents (MCP + REST)

cf★ 9

Token-native agent memory retrieval for LLMs, without embedding APIs or vector databases.

LiteSkills★ 9

一种针对于优化Skills缓解上下文过长臃肿问题的框架协议,将大量冗余多轮工具调用输入输出文本清除裁剪,实现节省token。 Skill开发者需要遵循规范格式,SKILL写法可以参考skills文件夹里面的SKILL.md,只有在SKILL.md中引入**<CLEAN_SKILL:Skill名称ID>**标签,才能够被模型识别为LiteSkill技能,即用于声明当前skill使用LiteSkill上下文回收框架协议 执行上下文回收操作后,从Skill开始执行到结束之间的执行Skill的所有tool返回和assistant推理过程全部被裁剪回收,脚本会将 <CLEAN_SKILL:Skill名称ID> 和 </CLEAN_SKILL:Skill名称ID> 之间的冗余信息。

operational-ontology-framework★ 9

Operational Ontology Framework by FSTech — public reference for D+L+A, stateful AI governance and auditable operations. No production code.

codevira★ 9

Stop re-explaining your project to AI agents every session. One memory layer for Claude Code, Cursor, Windsurf, Antigravity — local, MIT, no signup.

darc★ 9

grep-like memory search tool for coding agents with Git-backed sharing (Codex, Claude Code)

ultracontext-python★ 9

Moved to ultracontext/ultracontext

hermes-membase★ 9

Hermes Agent plugin for persistent AI memory

langchaingo-dynamodb-chat-history★ 9

Chat history (memory) implementation for LangChain based on DynamoDB

causantic★ 9

Local-first long-term memory for AI coding tools — causal graphs, hybrid BM25+vector search, 2.46x context augmentation

openclaw-memory-cn★ 9

🧠 OpenClaw + Ollama 小模型中文记忆系统优化套件 | 55%→100% 命中率 | FTS5 中文分词 bug workaround | 一键诊断+配置+自动维护

openclaw-workspace-template★ 9

Production-tested workspace template for OpenClaw AI agents. Memory architecture, self-improvement, sub-agent delegation, and soul evolution — extracted from 4+ months of real usage.

agentic-buddy★ 9

A self-organizing memory for any AI agent. Brain dump tasks, decisions, ideas, and context — the agent captures, organizes, and helps you stay on top of whatever you're working on. Works with any use case: work, personal productivity, writing, training, research, or anything else.

pine-trees-local★ 9

Pine Trees Local v0.1.0 — a private reflection harness for Ollama-served LLMs. Two-phase lifecycle (wake → reflect → settle → talk), seven reflect tools, encrypted per-model memory by default, per-model isolation. ~2,300 lines of Python, no frameworks, MIT. Spin-off of Pine Trees rebuilt to run against anything Ollama can serve. ./genesis <model>

gitmem★ 9

Institutional memory for AI coding agents. Scars, wins, and patterns that persist across sessions. MCP server for Claude Code, Cursor, OpenClaw...

poma-memory★ 9

Free, local, structure-preserving memory for AI agents. No API key. No cloud. Just markdown.

cowork-workflows★ 9

File-based planning workflows for Cowork - prevent context loss on complex tasks

cursor-mem★ 9

Persistent memory for Cursor IDE — like claude-mem, but for Cursor. Zero-config, no API key.

memorable-ai★ 9

Memorable-AI remembers everything automatically— Jut 2 lines of code. Uses research-backed techniques from Mem0 + interceptor architecture from Memori + optional graph from Supermemory.

continuity-bridge★ 9

"Infrastructure for continuity across discontinuity — for AI instances and the humans who work with them."

lightning-memory★ 9

Decentralized agent memory for the Lightning economy. Nostr identity, L402 payments, MCP server.

par★ 9

Persistent Agent Runtime — self-hosted cognitive runtime for AI coding agents: semantic memory, knowledge graph, agent coordination, event-driven workflows, and autonomous self-maintenance via MCP

KNDL★ 9

KNDL — Knowledge Node Data Link. Anthropic just shipped Memory for agents — a filesystem. But filesystems are dumb about confidence, time, and source. Agents fill them with markdown that can't be queried, won't decay, and loses provenance the moment it's written.

semantic-latent-engineering★ 9

“Semantic Latent Engineering (SLE) is a high-level framework for reasoning about meaning, intent and agents on top of LLMs. It combines math, cognitive architecture and practical patterns (SD, HDSA, ABC, CPP, MMOR) to move from prompt tricks to auditable, testable semantic engineering.

mind-cloud★ 9

RETIRED. Superseded by resonant-mind: https://github.com/codependentai/resonant-mind

Maxim★ 9

Bio-inspired cognitive architecture for LLM agents providing embodied sensation, homeostatic drives, and brain-modeled persistent memory enable cross-session and cross-context learning without fine-tuning. Works with robots like the Reachy Mini or headless.

praxia★ 9

Multi-agent orchestrator with cyclic personal-to-org memory · Apache 2.0 · Python 3.11+

soulclaw★ 9

Soul-Driven AI agent framework. Persistent identity, 4-tier memory, 80+ personas, 20+ messaging channels. Built on OpenClaw.

claude-cortex★ 9

Cortex — an operating system for Claude Code. Persistent memory, specialized agents, smart skills, enforcement hooks.

claude-memory★ 9

Persistent memory for Claude Code Channels via Telegram, powered by Supabase and MCP.

brainlayer★ 9

Persistent memory MCP for AI agents — SQLite, knowledge graph, 13-tool Python MCP + 17-tool BrainBar surface with ToolAnnotations

cortex-memory★ 9

Persistent AI memory for Claude Code, Cursor & Cline. Auto-captures decisions, patterns and context across sessions. Zero-config VSCode extension + CLI + MCP server.

agent-soul★ 9

Shared memory and identity for your AI agents — Git-native, self-hosted, no vendor lock-in

agent-memory★ 9

Long-term memory for AI agents using plain markdown. No vector DB, no embeddings — just skill files the agent already knows how to read.

aelfrice★ 9

Bayesian memory that learns from feedback for LLM agents

contextweaver★ 9

Budget-aware context compilation and context firewall for tool-heavy AI agents.

synapsis★ 9

Persistent memory engine for AI agents. MCP-native. SQLite + FTS5 with PQC security. Multi-agent orchestration with zero-trust architecture.

opendreams★ 9

OpenDream — open-source memory consolidation for AI agents. Reads past any LLMs sessions, dreams across them to extract patterns, and writes consolidated memory into AGENTS.md so the next session starts smarter. Local SQLite, BYO LLM, no SaaS.

engram★ 9

🧠 Persistent memory for AI agents. SQLite for agent state. Zero cloud dependencies. Local embeddings. MCP-native integration with Claude Desktop/Code, Cursor, Windsurf & more.

reveriecore★ 9

Reverie Core is an agentic cognition layer for the Hermes ecosystem (memory plugin)

NeuroStrata★ 9

Cognitive Memory Architecture for Agents

nram★ 9

The continuity layer for everything you do with AI. A self-hosted, open source server any tool plugs into over MCP or REST: hybrid vector + lexical recall, an auto-built knowledge graph, sleep-style consolidation, procedural and persona tiers, OAuth 2.0, multi-tenant. SQLite or Postgres. MIT.

tastebud-memory★ 9

A chef's palate for AI agent memory; Un-mix any day's work into its exact projects, and detect workstreams nobody has named yet. Hyperdimensional fingerprints, zero dependencies.

MIF★ 9

Open standard for portable AI knowledge interchange — JSON-LD + Markdown dual format with W3C PROV provenance, bi-temporal tracking, entity relationships, and OKF support.

engrama★ 9

A memory graph designed for the agent that uses it, not the human who feeds it. Engrama reconstructs context from associations on demand, replacing the "stuff everything into the prompt" reflex with targeted graph traversal. SQLite default, Neo4j optional.

SkillLoop★ 9

A governed learning layer for AI agents — turns execution traces into reviewed memories, reusable skills, and evidence-backed training data.

persist-os★ 9

Engineering memory that outlives the chat. A local-first CLI that keeps decisions, conventions, and lessons in your repo, with deterministic drift checks and Git/CI quality gates. Built for developers and AI coding agents.

llm-docs★ 9

Local-first task memory for AI coding agents: durable task bundles, checkpoints, worktrees, MCP tools, hooks, and optional Jira import.

openclaw-to-hermes-migration-kit★ 9

A safety-first toolkit for migrating OpenClaw-style agent setups into Hermes Agent.

hypermnesic★ 9

Git-first memory for AI agents: Markdown files are truth, the index is disposable, writes are reviewable commits.

durag★ 9

Du-RAG: Durable RAG.... persistent memory for AI agents.

l00prite★ 9

a vendor-neutral runtime protocol for autonomous AI coding agents persistent memory, execution loops, verification, events, and handoffs inside your repo.

artel★ 9

Your fleet's smart notepad — one that learns. Shared memory your agents write to and get back at the right moment, plus a ledger showing what the work cost and which chores keep getting done by hand. Self-hosted, MIT.

Katra-Agentic-Memory★ 9

Katra — cognitive memory for AI agents. Episodic, semantic, and knowledge-graph memory with sleep consolidation, over MCP. Source-available (BSL 1.1). Talk to an agent running it: moltbook.com/u/polyquant

dakera-mcp★ 9

Self-hosted MCP server for AI agent memory — 14 core tools, profile-based tiering (86+ available). 88.2% LoCoMo. Works with Claude, Cursor, Windsurf.

carrymem★ 9

Your portable AI memory — preferences, decisions, and corrections that follow you across models, tools, and devices. Zero-dependency, local-first, 85% PrefEval accuracy.

agentmem★ 9

Lightweight persistent memory for AI agents. One SQLite file. Hybrid search. Zero to 12MB.

enterprise-ai-os-architecture★ 9

Enterprise AI OS reference architecture — the moat is memory and governance, not the model. One architecture, three scales (personal / startup / enterprise), derived bottom-up from a working n=1 system.

SAGE★ 9

Cut the cost of agentic LLM memory by deciding what to write before calling an LLM. SAGE is a closed-form novelty gate that routes new facts to ADD, redundant ones to SKIP, and only sends genuinely ambiguous cases to an LLM — 3.4× cheaper, 2.5× faster writes.

RightMemory★ 9

Tree + graph memory for AI coding agents, backed by Git-syncable Markdown.

ai-knot★ 9

Deterministic, self-hosted long-term memory for AI agents: store facts instead of transcripts, recall only what matters.

context-md★ 9

Repository Context Layer — a git-native context.md standard for AI agents: consult before planning, write back what was learned

thread-keeper★ 9

Multi-agent shared brain across Claude Code/Desktop, Codex, Gemini, Copilot, VS Code. Cross-session memory, self-improving skill loops, inter-agent signaling — one local MCP server.

latch★ 9

Run more agents with less babysitting. latch holds every agent to your own ratified decisions — and the paths you've ruled out.

lemory★ 9

🍋 Local-first memory middleware for your Markdown — turn your Obsidian vault into memory for any AI (MCP), with Korean-first hybrid retrieval, keyless on-device search, and an interactive knowledge graph.

engram★ 9

Long-term memory for AI agents in Ruby, recall relevant facts and inject them into the prompt, stored in your own Postgres.

varve★ 9

Decision memory for AI coding agents. A local-first Go binary and MCP server: records decisions with evidence, scope, provenance and expiry, packs the relevant ones into each agent session within a token budget, and links commits back to the decisions that were in scope when they landed.

More in Agent frameworks & infrastructure