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.
-
Open-source local-first cognitive memory. AGM-compliant belief revision (49/49 postulates). When a fact changes, downstream beliefs are automatically re-evaluated, not just flagged.
Apache-2.0 ★ 86 -
@mrvicadai/cortexactive
Unified Agent Context Architecture: normative beliefs and persistent episodic memory for AI coding agents
npm MIT -
@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 -
touchgrass-mcp-servergithub.com/chrbailey/touchgrassupdated this year
AI Chief of Staff — emotional memory, Bayesian belief tracking, dual-engine gap analysis, voice-adaptive briefings, and governed memory. MCP server for Claude Code.
pypi MIT -
Governed persistent memory for AI agents — trust scoring, contradiction detection, and belief lifecycle management via MCP and API
NOASSERTION ★ 3 -
patha-memorygithub.com/autotelicbydesign/pathaupdated this year
Local-first AI memory with contradiction detection, non-destructive supersession, and empirically non-commutative belief evolution. Plugs into any MCP-compatible AI tool.
pypi Apache-2.0 -
Particles reference engine — corpus, LLM claim extraction, belief store, query/lint/review, CLI + HTTP + MCP. PyPI: linkedparticles
pypi Apache-2.0 ★ 4 -
Memory for AI agents: beliefs with a source, a certainty and a revision history. SQLite, zero deps, MCP server.
npm MIT ★ 3 -
An MCP server that gives an agent a dialect with its owner: beliefs with standing, a feedback loop, and a gate on action. No model calls, no API key.
npm MIT -
The system of record for what an AI agent believed and did. Append-only beliefs with evidence, contradictions kept, risky actions gated behind a named human. MIT, self-hosted.
npm MIT -
agentic-memorygithub.com/Verace-Pvt-Ltd/agentic-memoryupdated this year
6-layer persistent memory (sensory, working, episodic, semantic, long-term, procedural) with non-destructive belief revision and injection-safe context synthesis — an MCP server for Claude Code, Cursor, Antigravity, Codex CLI, and any MCP-compatible agent.
MIT ★ 1 -
Persistent semantic memory for Claude Cowork. Extracts beliefs, insights, and research from every session into a local database Claude queries in real time. Nightly research scouting, automated belief verification, and MCP-based retrieval. Built for Cowork — not Claude Code.
MIT -
Liquid Memory — adaptive belief revision MCP server.
Apache-2.0 -
A claims memory for agents: assert beliefs, let them decay, surface contradictions (MCP + CLI)
npm MIT -
Kenwer Channel 3 - Cognitive Identity Layer: persistent/episodic/consolidated memory, gated C1-C6 belief revision, Channel 1 retrieval, and MCP. No flight control.
-
Local-first agentic memory: the engram MCP server launcher. Memory + knowledge graph + beliefs + code intelligence for any MCP client.
npm MIT -
Deep research in Rust: 5 source types, multi-agent belief debate, SQLite-backed queue, MCP server for OpenCode
MIT ★ 1 -
Local-first MCP server: hash-chained document store, signed retrieval receipts, and an orient protocol that tells your agent when its beliefs went stale.
Apache-2.0 ★ 1 -
Verified memory for AI coding agents. Every memory carries evidence and is re-verified on every pull — stale beliefs get flagged, not retrieved. MCP + CLI.
npm MIT ★ 1 -
Deterministic cognitive state substrate & world model runtime for autonomous agents. Features transactional belief updates, hybrid RK4 rollouts, and 3-tier experience consolidation via MCP & REST.
pypi Apache-2.0 ★ 6 -
Midasgithub.com/vornicx/Midasupdated this year
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.
pypi MIT ★ 14 -
knowledge-agentgithub.com/oladri-renuka/knowledge-agentupdated this year
Persistent belief graph from document sequences. Extracts entities/claims, resolves duplicates via two-pass entity resolution (alias + cosine similarity), detects contradictions across sources, answers questions with citations. 936 entities from 173 pages using 9 API calls. MCP server for Claude Desktop/Cursor.
-
Memory that forgets: a persistent, self-revising belief-state MCP server and Python API for agentic R&D. Uses a hypothesis DAG to auto-prune dead ends and deduce truth without extra probes.
pypi MIT ★ 13 -
Git revert for AI agent beliefs: shared memory for AI agents and teammates on FalkorDB. One correction quarantines a wrong fact and everything built on it, blocks and compensates actions. Set up from any AI chat.
AGPL-3.0 ★ 1 -
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 -
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 -
synaptic-stateupdated this year
Verifiable State Plane for Autonomous Agents
pypi MIT -
mcp-dual-cycle-reasonerstale
MCP server implementing dual-cycle metacognitive reasoning framework for autonomous agents
npm MIT -
Persistent cognitive graph memory for AI agents — facts, decisions, reasoning chains, corrections. 16 query types, sub-millisecond. Rust core + Python SDK + MCP server.
cargo MIT ★ 32 -
Deterministic, non-destructive, replayable memory substrate for AI agents: an append-only claim ledger with supersession, provenance, deterministic replay, and semantic recall over MCP.
-
Hosted memory for AI agents: commit, recall, revise, snapshot and roll back agent state with provenance. Python SDK, CLI and MCP server. Free, no card required.
NOASSERTION ★ 1 -
Building world-class AI memory: open-source, local-first memory for AI agents with evidence, hybrid retrieval, and MCP. This is the official best ai memory in the world, as seen on bridgemind.
Apache-2.0 ★ 1