Skip to content
MCP server

JFSecure

Let agents run commands with your secrets without ever seeing the values

Runs locallyStatus and history
Pricing
Free
Platforms
Windows, macOS, Linux
Runs
Locally
Version
2.10.0
Listing updated
3 days ago

About JFSecure

JFSecure is a free MCP server for security. It runs locally on your machine and is started by your AI client.

Pasting tokens into chats or letting an agent read a .env puts them into transcripts, logs and the model provider's context. Here the agent asks for a secret by name and a window appears showing the program, the exact command and the secret, with choices to allow once, allow for 15 minutes or deny. The value goes straight from the app into the process environment, and every secret in the output is replaced with a marker such as [JFSecure: TOKEN].

There is deliberately no get tool. SSH keys never leave the app: ssh and git push use its own agent to sign after approval. Creating a secret also needs approval, and the agent cannot change or delete existing ones or read values back. A template tool fills references into files like .env.

It needs the JFSecure desktop app, an offline password manager that is free and needs no account, running and unlocked. The jfs command and MCP server are MIT licensed, a thin client with no vault access, and set up with one claude mcp add or MCP config entry.

Install

Run in a terminal

claude mcp add jfs -- npx -y https://github.com/theone55/jfs/releases/download/v2.10.0/jfs-2.10.0.mcpb

Add to claude_desktop_config.json

{
  "mcpServers": {
    "jfs": {
      "command": "npx",
      "args": [
        "-y",
        "https://github.com/theone55/jfs/releases/download/v2.10.0/jfs-2.10.0.mcpb"
      ]
    }
  }
}
Install in CursorOpens Cursor, which asks you to confirm. Or add it by hand:

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "jfs": {
      "command": "npx",
      "args": [
        "-y",
        "https://github.com/theone55/jfs/releases/download/v2.10.0/jfs-2.10.0.mcpb"
      ]
    }
  }
}
Install in VS CodeOpens VS Code, which asks you to confirm. Or add it by hand:

Add to .vscode/mcp.json

{
  "servers": {
    "jfs": {
      "type": "stdio",
      "command": "npx",
      "args": [
        "-y",
        "https://github.com/theone55/jfs/releases/download/v2.10.0/jfs-2.10.0.mcpb"
      ]
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "jfs": {
      "command": "npx",
      "args": [
        "-y",
        "https://github.com/theone55/jfs/releases/download/v2.10.0/jfs-2.10.0.mcpb"
      ]
    }
  }
}

Add to your client's MCP settings (most use this shape)

{
  "mcpServers": {
    "jfs": {
      "command": "npx",
      "args": [
        "-y",
        "https://github.com/theone55/jfs/releases/download/v2.10.0/jfs-2.10.0.mcpb"
      ]
    }
  }
}

Tools this server provides (4)

Taken from the project documentation; the running server may differ.

list_secrets
Lists secret and key names without values.
run_with_secrets
Runs a command with a secret in its environment and returns masked output.
render_template
Fills secret references in a template and writes the file.
create_secret
Adds a new secret after approval; never returns values.

What it does

Masked output
Secret values in command output are replaced by labels.
Per-use approval
Each use is approved in a window showing the command.
Environment injection
Runs commands with secrets as environment variables.
Template rendering
Fills secret references into files such as .env.
Name-only listing
Agents can see secret names, never values.
SSH signing
Git and ssh use a signing agent so keys never leave the app.

Good for

  • Deploy with a database password without exposing it to the model
  • Call an API with a token the agent never reads
  • Generate a .env from stored secrets
  • Cut credential leakage into agent transcripts

Pricing

Free

Free, offline, no account

Prices change. Check secure.jfolio.org before you buy.

JFSecure FAQ

Can the agent read a secret value?

No. There is no get tool and output is masked.

Do I have to approve each use?

Yes, with options to allow once, for 15 minutes or deny.

What does it require?

The JFSecure desktop app running and unlocked, on Windows, macOS or Linux.

Is it free?

The site says offline and free, and the jfs MCP server is MIT licensed.

Is JFSecure free?

Yes. JFSecure is free to use.

How do I add JFSecure to Claude Code?

Run this command in your terminal: claude mcp add jfs -- npx -y https://github.com/theone55/jfs/releases/download/v2.10.0/jfs-2.10.0.mcpb

What are the best alternatives to JFSecure?

The closest alternatives to JFSecure are REA, Codebase Memory, GitHub, Serena MCP: the IDE for your agent and XERJ.

Alternatives to JFSecure

  1. REA

    MCP server

    CLI and MCP server that lets coding agents reverse engineer apps and binaries

    Runs locally
    • Open source
    • 464K GitHub stars
    • Developer tools
  2. Codebase Memory

    MCP server

    Local MCP server that builds a code knowledge graph for AI coding agents

    Runs locally
    • Open source
    • 48.7K GitHub stars
    • Knowledge and memory
  3. GitHub

    MCP server

    Official MCP server connecting AI assistants to GitHub repos, issues and workflows

    Not checked yet
    • Remote
    • Open source
    • 34.5K GitHub stars
    • Developer tools
  4. MCP toolkit giving coding agents symbol-level code retrieval and editing

    Runs locally
    • Open source
    • 31.5K GitHub stars
    • Developer tools
  5. XERJ

    MCP server

    Local search engine for AI agents with cited answers over your code and docs

    Runs locally
    • Open source
    • 19.2K GitHub stars
    • Knowledge and memory
  6. Ouroboros

    MCP server

    Spec-first layer that pins requirements and verifies results of AI coding agent work

    Runs locally
    • Open source
    • 6.4K GitHub stars
    • Developer tools