Semantic memory for AI agents with hybrid search, knowledge graph, and consolidation
Neuromcp MCP server is a locally run integration for AI assistants that speak the Model Context Protocol. Semantic memory for AI agents with hybrid search, knowledge graph, and consolidation.
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.
Once Neuromcp is connected, these are the calls the assistant has available:
macOS — a launchd agent that fires every 4 hours (com.neuromcp.consolidate)Linux — prints a cron snippet to add manuallyThreshold — skip if fewer than 5 unprocessed sessionsv0.27 — security release: CWE-22 path-traversal fix in wiki_ingest,v0.26 — recall_answer (deterministic extractive answers withv0.20 — 0.25** — shared HTTP daemon, session isolation, critic hook,Run — ScoreMode — R@5Embedder — DistractorsWhen — What happensCrash — Checkpoint every 5 tool calls to file. Git history for rollbackFeature — Claude CodeYou will need 2 environment variables: NEUROMCP_DB_PATH, NEUROMCP_NAMESPACE. Keep credentials in your client's env block or a secrets manager rather than in a file you might commit.
Installation goes through your MCP client rather than a global install: point it at neuromcp-init on npm and it is fetched when the client starts. The copy-paste blocks for Claude Desktop, Claude Code and Cursor are further down this page.
This sits in the knowledge and memory group, where several servers overlap in what they claim to do but differ sharply once you actually set them up. Neuromcp's toolset — macOS, Linux, Threshold and 11 more — is a fair guide to whether it matches your workflow. It is maintained by AdelElo13; worth a glance at recent repository activity before you build anything load-bearing on it.
This entry was verified against Neuromcp's own documentation before publication; SyncDev keeps the directory reviewed rather than auto-generated.
| Tool | What it does |
|---|---|
| macOS | a launchd agent that fires every 4 hours (com.neuromcp.consolidate) |
| Linux | prints a cron snippet to add manually |
| Threshold | skip if fewer than 5 unprocessed sessions |
| v0.27 | security release: CWE-22 path-traversal fix in wiki_ingest, |
| v0.26 | recall_answer (deterministic extractive answers with |
| v0.20 | 0.25** — shared HTTP daemon, session isolation, critic hook, |
| Run | Score |
| Mode | R@5 |
| Embedder | Distractors |
| When | What happens |
| Crash | Checkpoint every 5 tool calls to file. Git history for rollback. |
| Feature | Claude Code |
| Component | macOS |
| Family | Tools |
### Claude Desktop
```jsonc
// ~/Library/Application Support/Claude/claude_desktop_config.json
{
"mcpServers": {
"neuromcp": {
"command": "npx",
"args": ["-y", "neuromcp"]
}
}
}Configuration as documented by the project. Restart the client after saving.
| Variable | Description | Required |
|---|---|---|
| NEUROMCP_DB_PATH | Filesystem location the server is allowed to use. | Optional |
| NEUROMCP_NAMESPACE | Configuration value read at startup. | Optional |
A knowledge graph your assistant keeps between sessions — entities, relations and observations that persist.
Kill hallucinated APIs — version-accurate, up-to-date library documentation injected straight into context.
Your workspace, on speaking terms with AI — search, read and write Notion pages and databases.
A structured scratchpad for hard problems — stepwise reasoning with revisions, branches and visible logic.
Symbol-level code navigation, refactoring and memory for coding agents — the IDE brain your assistant has been missing.
Chat with your second brain — search, read and write vault notes through the Local REST API.