Neuromcp
This is a local-first memory layer that runs entirely on your machine with no cloud dependencies. It stores conversations and facts in SQLite with hybrid retrieval (vector embeddings via nomic-embed-text, BM25 full-text, knowledge graph links), exposing MCP tools for semantic search, memory consolidation, and a git-tracked wiki that compiles Markdown knowledge bases organized by people, projects, and patterns. The consolidation pipeline batches raw sessions every few hours and uses Claude to fact-check summaries against source material. Reach for this when you want persistent memory across Claude Desktop, Cursor, or any MCP client without sending your conversation history to a third party API. The author publishes benchmark numbers with sample size caveats and positions it as portable memory you control rather than vendor-locked cloud storage.
- GitHub stars
- 4
- tools
- 20
- transport
- STDIO
- last updated
- Oct 15, 2026
Tools
20 toolsPublic tool metadata for what this MCP can expose to an agent.
store_memoryStore a memory with dedup, contradiction detection and surprise scoring.
Store a memory with dedup, contradiction detection and surprise scoring.
No parameter schema in public metadata yet.
search_memoryHybrid RRF search returning ranked raw memories with explain metadata.
Hybrid RRF search returning ranked raw memories with explain metadata.
No parameter schema in public metadata yet.
recall_answerSynthesizes an extractive cited answer, or reports not_in_memory.
Synthesizes an extractive cited answer, or reports not_in_memory.
No parameter schema in public metadata yet.
search_allSearches memories and the verbatim store together.
Searches memories and the verbatim store together.
No parameter schema in public metadata yet.
recall_memoryPlain ID/filter lookup of memories with no semantics.
Plain ID/filter lookup of memories with no semantics.
No parameter schema in public metadata yet.
create_entityCreates an entity in the knowledge graph.
Creates an entity in the knowledge graph.
No parameter schema in public metadata yet.
create_relationCreates a relation between entities in the knowledge graph.
Creates a relation between entities in the knowledge graph.
No parameter schema in public metadata yet.
query_graphQueries the knowledge graph.
Queries the knowledge graph.
No parameter schema in public metadata yet.
compute_centralityComputes PageRank centrality over the knowledge graph.
Computes PageRank centrality over the knowledge graph.
No parameter schema in public metadata yet.
start_episodeStarts a memory episode.
Starts a memory episode.
No parameter schema in public metadata yet.
end_episodeEnds a memory episode.
Ends a memory episode.
No parameter schema in public metadata yet.
memory_timelineReturns a timeline of memories.
Returns a timeline of memories.
No parameter schema in public metadata yet.
register_agentRegisters an agent for multi-agent memory use.
Registers an agent for multi-agent memory use.
No parameter schema in public metadata yet.
find_expertFinds an agent expert for a topic.
Finds an agent expert for a topic.
No parameter schema in public metadata yet.
wiki_ingestIngests content into the wiki.
Ingests content into the wiki.
No parameter schema in public metadata yet.
wiki_lintLints the wiki.
Lints the wiki.
No parameter schema in public metadata yet.
wiki_briefingProduces a wiki briefing.
Produces a wiki briefing.
No parameter schema in public metadata yet.
log_retrievalLogs a retrieval event for the usefulness prior.
Logs a retrieval event for the usefulness prior.
No parameter schema in public metadata yet.
cite_memoriesRecords citations of memories to close the usefulness-prior loop.
Records citations of memories to close the usefulness-prior loop.
No parameter schema in public metadata yet.
generate_reflectionGenerates a reflection over memories.
Generates a reflection over memories.
No parameter schema in public metadata yet.
Sponsors
Agent CaffeineSponsoredClose your lid, come back to finished work
SnugSponsoredThe clipboard manager for the AI eraPromote your product hereReach 125k+ visitors monthly
WaxSponsoredBlock anything with supervision, without erasing your iPhonePromote your product hereReach 125k+ visitors monthly
knobsSponsoredSee your site on any phone, any day, in any languageReadme
neuromcp — Sovereign Memory for AI agents
Any model. Your memory. Stays local.
neuromcp is the first Sovereign Memory layer for AI: an open-source MCP server that gives Claude, GPT, Gemini, and Ollama persistent, searchable memory — stored entirely on your machine. No API keys. No cloud sync. No subscription required to remember who you are.
Sovereign Memory = data that you own outright, lives on hardware you control, and is portable across every model you use. Cloud memory products own your data; Sovereign Memory means you do.
npx neuromcp-init # one command: detects your MCP clients, writes configs, sets up the wiki
Details
- Repository
- github.com/adelelo13/neuromcp
- Categories
- AI & LLM Tools, Documents & Knowledge, Search & Web Crawling
- Package
neuromcp- Transport
- STDIO
- Registry
- MCP registry, active
More from adelelo13
Related AI & LLM Tools MCP servers
View all
Lune Researchretrogradelabs/lune-mcp-serverSearch peer-reviewed papers and research methodology guidance from your AI agent.4stars
EchoVault Memorygo-ports/echovaultLocal-first memory for coding agents; stores decisions, bugs, and context across sessions.4stars
Wso2 Docs Mcp Serveriamvirul/wso2-docs-mcp-serverSemantic search over WSO2 docs (APIM, IS, MI, Choreo, Ballerina) via RAG and pgvector.4stars
Memographindhar01/memographGraph-based memory system for LLMs with knowledge graphs and semantic search.4stars
CoinRithm Agent Tradingcoinrithm/coinrithm-agent-tradingKeyless prediction-market data across 12 venues plus paper-trading of crypto spot, futures, and PM.4stars