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.
-
Persistent, bi-temporal memory for LLM agents — local-first, MCP-native, Obsidian-readable.
pypi Apache-2.0 -
gnokeegithub.com/omurlabs/gnokeeupdated this year
Memory infrastructure for personal AI — bi-temporal facts, honest contradictions, autonomous maintenance, real forgetting. MCP-native, multi-tenant, multilingual.
pypi Apache-2.0 -
memharnessgithub.com/las7/memharnessupdated this year
Bi-temporal agent long-term memory: SQLite-backed MCP server with recall ranking, hybrid vector+FTS recall, and a source-staleness signal
npm Apache-2.0 ★ 6 -
gnokeegithub.com/gnokeelabs/gnokeeupdated this year
Memory infrastructure for personal AI — bi-temporal facts, honest contradictions, autonomous maintenance, real forgetting. MCP-native, multi-tenant, multilingual. Placeholder release; see https://github.com/gnokeelabs/gnokee.
npm Apache-2.0 -
Bi-temporal, 4-tier hybrid cognitive memory engine for AI coding agents (Claude Code, Cursor, Windsurf) via FastMCP. Eliminates amnesia and reduces token costs using local PostgreSQL & Neo4j.
-
@freshair129/gksgithub.com/freshair129/gksv3updated this year
Genesis Knowledge System — a memory fabric for agentic AI built on the EVA Tri-Brain architecture. Four memory layers, bi-temporal versioning, multi-tenancy, OpenTelemetry, MCP server, and a CLI.
npm MIT -
thought-mcpgithub.com/RNBBarrett/thought-mcpupdated this year
Local MCP memory server for LLMs. Bi-temporal graph + vector + Cypher queries + auto-write/auto-recall hooks for Claude Code. Works with Ollama, LM Studio, Anthropic, OpenAI.
pypi MIT ★ 1 -
Structural memory for AI coding agents. Bi-temporal graph, MCP-native, zero LLM calls. Cursor · Claude Code · Codex · DeepSeek Harness · Hermes · VS Code · Windsurf.
npm NOASSERTION ★ 515 -
Zero-infrastructure, local-first memory for AI agents. One SQLite file — no Postgres, Neo4j, or cloud. Bi-temporal supersession (updates, never deletes), four-channel hybrid retrieval, and a drop-in MCP server. Runs offline on cheap models. Apache-2.0. Bring your own benchmark.
Apache-2.0 ★ 14 -
Agent-native bi-temporal graph store: verified temporal operators as LLM tools + trace-grounded answer verification
pypi Apache-2.0 ★ 5 -
tensegithub.com/Zacplischka/tenseupdated this year
Temporal memory for AI agents — an MCP server storing knowledge as a bi-temporal graph on Postgres. Knows which version is true now, or as of any past date.
MIT -
Stratum-Memory: a four-layer AI memory & context system (filesystem L0-L2, bi-temporal graph, vector/KV, reflect) with an ECL ingestion pipeline, a curator gate, and an MCP tool API. Runs on a Raspberry Pi 4 or a multi-tenant server.
Apache-2.0 -
mcp-tenant-pair-demoactive
Reference MCP server demonstrating mcp-tenant-pair (12 tools, stdio): pair lifecycle + bi-temporal shared state + DSGVO Art.17 erasure
npm MIT -
Open-source long-term, time-aware memory engine for AI agents (MCP + hybrid search + bi-temporal recall)
Apache-2.0 ★ 4 -
Local-First Bi-Temporal Knowledge Graph, Cross-Agent Persistent Memory & Token-Budgeted Context Compiler MCP Server for AI Coding Agents.
MIT ★ 1 -
aionforge-memorygithub.com/jscott3201/aionforge-memoryupdated this year
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.
Apache-2.0 ★ 12 -
Local-first, inspectable memory for coding agents: durable context across sessions and repositories, code-aware recall, bi-temporal history, MCP, and a self-hosted WebUI.
pypi Apache-2.0 ★ 184 -
engramgithub.com/HEMANTH-A-7/engramupdated this year
Local-first memory layer for coding agents: bi-temporal fact resolution, tiered consolidation, and MCP tools for automatic dev-context capture/recall. 100% offline, SQLite only, no cloud calls.
★ 1 -
Local-first encrypted memory for AI agents — SQLCipher-encrypted, hybrid search (BM25 + vectors + graph), bi-temporal validity, MCP server for every client. Source available (BUSL-1.1).
npm NOASSERTION ★ 4 -
Self-hosted long-term memory for AI coding agents: Postgres+pgvector doc-RAG (component/constraint map + hybrid RRF + cross-encoder rerank) and bi-temporal personal memory. MCP-native, local-model friendly.
MIT ★ 17 -
Local-first memory engine for coding agents (Claude Code/Cursor/Codex): SQLite+FTS5 retrieval, versioned bi-temporal memories, full-session capture, zero dependencies, MCP server
MIT ★ 83 -
Zero-daemon cognitive memory engine for AI coding agents. AST code graph, self-verifying error traps, and bi-temporal SQLite for Claude Code, Cursor & Copilot.
npm NOASSERTION ★ 1 -
MIRSAD (مِرصاد) — a shared, living memory hub for AI agents: bi-temporal facts, first-class decisions, owner-resolved conflicts, and a live graph view. Local-first
pypi MIT ★ 2 -
Auditable memory for AI agents — provable deletion (crypto-shred + signed certificate), a tamper-evident audit chain, and bi-temporal time-travel recall. Runs on SQLite or Postgres.
pypi Apache-2.0 -
llm-memorygithub.com/DamianTarnowski/llm-memoryupdated this year
Persistent, cross-model memory for LLM assistants — MCP server in .NET 10 over Postgres + pgvector + Apache AGE. Hybrid retrieval (vector + BM25 + graph PPR + LLM rerank), bi-temporal supersession, multi-tenant RLS.
MIT -
Persistent local memory for Claude, Cursor & Codex. 21 MCP tools, hybrid retrieval (BM25 + vector RRF), bi-temporal queries, contradiction detection. SQLite + Knowledge Graph. No cloud, no API keys. npx @studiomeyer/local-memory-mcp
npm MIT ★ 16 -
MCP memory server for Claude Code: a git-diffable Markdown vault (opens in Obsidian) with hybrid semantic search, a wiki-link graph, and bi-temporal facts — plus hooks that pull context in automatically. Local, no API keys.
pypi Apache-2.0 -
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).
npm Apache-2.0 ★ 12 -
CodeGraphX (CGX) — a local, token-efficient codebase graph engine & MCP server for AI coding agents. Tree-sitter parsing, a bi-temporal SQLite semantic graph, O(1) symbol lookup, and impact/blast-radius tracing so agents answer 'what breaks if I change this?' in a few hundred tokens.
npm MIT ★ 3 -
Turns any text stream into a dated history of how its subject changed. Point it at your Claude Code, OpenCode or Codex sessions to track how you think and work, or at market news, research, or any corpus to track how it changed — raw records to observations to bi-temporal facets to narrative, with provenance, served over MCP. Single pure-Go binary.
npm MIT ★ 12 -
VERA is a shared, governed memory platform for AI agents, providing trusted, temporal knowledge across teams through MCP, with Graphiti as the underlying memory engine.
Apache-2.0 ★ 4 -
@weavory/mcpgithub.com/deepakkts/weavory.aiupdated this year
weavory.ai — Responsible-AI memory for agent swarms. Every belief signed, every decision auditable. MCP-native.
npm Apache-2.0 -
mcp-memory-graphgithub.com/YonasValentin/mcp-memory-graphupdated this year
Local-first memory for Claude Code and any MCP client: hybrid search + knowledge graph in one SQLite file, $0/token, no cloud.
npm NOASSERTION ★ 2 -
agidbgithub.com/rohansx/agidbupdated this year
A cognitive substrate for AI agents — hyperdimensional memory with first-class goals, beliefs, sensory, self-model, and a non-destructive unlearn primitive. Rust top to bottom, embedded-first, no LLM in the read path. One binary, no query language.
Apache-2.0 ★ 6