MCP servers that ship as code — a repository or a package you install and run locally (usually over stdio). They have no hosted endpoint, so we cannot probe or grade them: instead each one shows its repository, install command, license, stars and when it was last updated.
Collected from GitHub repositories and npm, PyPI, Cargo, Go, Maven, NuGet and RubyGems packages tagged as MCP servers, the official MCP Registry, Docker's MCP catalog, ModelScope, Baidu MCP World, Cline and other public directories; one entry per repository. A server that also offers a hosted endpoint is listed under Remote instead. These entries are not counted in our reachability, verification or readiness statistics.
-
Temporal memory system for AI assistants with human-like forgetting curves. All data stored locally in human-readable formats: JSONL for short-term memory, Markdown (Obsidian-compatible) for long-term. Memories naturally decay unless reinforced. Features knowledge graphs, smart prompting, and MCP server integration for Claude.
pypi AGPL-3.0 ★ 39 -
Ebbtide MCP: local long-term memory for AI agents. Hybrid keyword + embedding recall, fact updates, and a forgetting curve for bounded stores; benchmarked on LongMemEval and LoCoMo.
MIT -
给 AI Agent 的长期记忆库:会遗忘、会弃答、零依赖 · Long-term memory for LLM agents with forgetting curve, abstention, SQLite/WAL, multi-process writes and an MCP server. No vector DB required.
MIT ★ 1 -
Temporal memory for AI assistants using the Ebbinghaus forgetting curve with MCP server integration for Claude, Cursor, and compatible agents.
AGPL-3.0 -
The memory engine behind sharpwave (MCP server) and openwave (OpenClaw plugin): graph store, FSRS-6 forgetting curve, spreading activation, sleep-style consolidation, LLM extraction.
npm MIT -
MMPL Memory Palace plugin for DeepSeek Harness (@cup319/dsh-mmpl) — DuckDB vector memory MCP with skills and forgetting curve.
npm MIT -
Agentic AI memory with Ebbinghaus forgetting curve decay. +16pp better recall than Mem0 on LoCoMo.
pypi NOASSERTION ★ 274 -
Local-first, markdown-native memory for AI agents: tiered memory, an Ebbinghaus forgetting curve, consolidation, and an MCP server.
MIT ★ 2 -
SuperMem 超级记忆系统 — local-first long-term memory for AI agents: knowledge graph + vector search + verbatim layer, episodic memory, forgetting curve, causal edges, MCP/REST API. 图数据库与向量检索的语义记忆平台。
MIT ★ 3 -
Palimpsestgithub.com/Vector897/Palimpsestupdated this year
Agent memory that survives — a production-grade memory layer on CockroachDB. Episodic→semantic consolidation, history-preserving reconciliation, forgetting curves, and crash-safe checkpoints. Any agent plugs in via MCP.
Apache-2.0 -
MemoryMeshgithub.com/0DevDutt0/MemoryMeshupdated this year
Production-grade persistent memory server for AI agents. MCP-compatible (Claude Code, Cursor). Four memory types (episodic/semantic/procedural/preference), hybrid vector retrieval (FAISS + bge-large), hierarchical LLM compression (Groq + Mistral), Ebbinghaus forgetting curve, FastAPI + Streamlit.
-
本地优先的 AI 记忆中枢:SQLite 全文检索、短期遗忘曲线与每日回顾、场景级隐私与敏感位;为 DSH MCP、QQ 桥接和桌宠提供同一份长期记忆。 Local-first memory hub for DSH, QQ bridge and desktop pets: SQLite full-text search, short-term forgetting curve, daily review, scene-level privacy.
NOASSERTION -
可塑性记忆 — 让 AI 学会做梦。会遗忘、会做梦、会改主意的三层记忆(Hermes 插件 / MCP / CLI,纯标准库零依赖)
MIT -
@cdzzy/engramactive
Standardized long-term memory for AI agents 鈥攆orgetting curve, behavior observer, shared memory spaces, memory versioning, and MCP stdio server
npm MIT -
给 AI 装一套像人脑的记忆:重要的事记牢、日常的会淡忘、休眠不删除、被提起才想起。按情绪效价×唤醒度打分,区分余波期与心结。零依赖、纯本地、单文件 SQLite,支持 AI skill / MCP / 命令行。
MIT -
ai-mind-suitegithub.com/hwc66626/ai-mind-suiteupdated this year
三个 MCP 服务器:受规则约束的思维框架(八步决断闸门)、带遗忘曲线的记忆库(双强度+上下文策展)、事件型内心提醒(任务锚定闹钟)。本地 SQLite,零云依赖,178 项断言。工具定义 token 开销实测见各 README。
MIT -
Continuity for the agent itself. An MCP server that gives an AI agent a memory of its own, written in its own words and kept on your machine, so it wakes as itself every session.
pypi Apache-2.0 ★ 165