FRENES

Databases · MCP server

Memory Vault: MCP server, setup and security

Memory Vault is an open-source MCP server that lets an AI agent use self-hosted AI memory with hybrid Postgres and pgvector search. Its repository has 65 stars and its last commit is from October 19, 2026 and it is licensed under MIT.

Actively maintainedNeeds secrets

What Memory Vault does

According to its repository: “Local-first AI memory layer with hybrid search. Postgres + pgvector. Self-hosted, MIT.”

An MCP server gives an AI agent extra tools it calls during the conversation: the agent reads their description, decides to use them and gets the result back. Memory Vault is listed under Databases, a category of 138 servers in the directory, where it ranks #91 by stars.

Topics declared by the repository: ai-memory, claude, hybrid-search, knowledge-graph, open-source, pgvector.

Installing and connecting Memory Vault

Distribution modes declared in the official MCP registry:

  • Docker (OCI) image

Like any MCP server, you declare it in your client's configuration (Claude Desktop, Claude Code, Cursor, VS Code, Codex…): a name, the command that starts it or its URL, and any environment variables. The client then starts it and lists its tools. Project language: Python.

The registry lists one secret variable to provide (API key, access token). Keep them in the client's environment variables, never in a shared file or a repository.

Security: what to check

An MCP server runs code with the rights you give it, and the model reads its tool descriptions: a malicious or compromised server can hijack the agent. Read the code, or pick a maintained and widely used server, before connecting it.

Points found for Memory Vault:

  • Secrets: the server acts with your credentials. Create a dedicated key with minimum rights and an expiry date.

Good practice: grant only the rights needed (least privilege), require human approval for irreversible actions, and run the AI agent security check.

Project activity and maintenance

Created on March 21, 2026, the MihaiBuilds/memory-vault repository has 65 stars and 14 forks, with 5 open issues. It has more stars than 57% of the servers in the directory. It received changes this month. Version published in the registry: 1.6.1 (October 9, 2026).

Stars measure attention, not quality: a recent, heavily promoted project can collect them fast, while a useful but niche server will have few.

DbxDbx : 27.7K27.7KMCP Toolbox for DatabasesMCP Toolbox for Databases : 16.7K16.7KNhostNhost : 9.4K9.4KJitsuJitsu : 5.1K5.1KPostgres MCP (crystaldba)Postgres MCP (crystaldba) : 3.4K3.4KLINQ to DBLINQ to DB : 3.3K3.3KSupabaseSupabase : 3K3KMindwtrMindwtr : 2.3K2.3KMemory VaultMemory Vault : 6565
GitHub stars of the most followed servers in the category (log scale); Memory Vault in orange. Checked October 25, 2026.

Verdict

Memory Vault is a reasonable choice; check the points below before connecting it.

Strengths

  • Active project: recent commit
  • Listed in the official MCP registry
  • Permissive licence (MIT)

Limitations

  • Requires an API key or token

Avoid if…

  • You do not want to give credentials to an agent

Frequently asked questions

How do I install Memory Vault?

Memory Vault is distributed as Docker (OCI) image. Add it to your client's MCP configuration (Claude Desktop, Cursor, VS Code…) following the README of MihaiBuilds/memory-vault.

Is Memory Vault free?

The code is published under the MIT licence and can be used free of charge. It needs an API key: the service it connects to may be paid.

Is Memory Vault safe?

Points to watch: secrets. Give it minimum rights and approve sensitive actions.

Is it an official server?

Nothing says so: it is a project published by MihaiBuilds on GitHub. Check who maintains it before giving it access.

What are the alternatives to Memory Vault?

In the same category: Nhost, ArcadeDB MCP Server, Jitsu, and Supabase. Nhost and ArcadeDB MCP Server score higher on our grid.

Related articles

Neutrality. No affiliate links. No vendor pays to be reviewed, rated or ranked. Reviews are directed by François Chollet, neutral and factual, and every data point is dated and sourced. How we score