🇬🇧 English · 🇷🇺 Русский

Passbolt MCP

Work with your password vault from your AI assistant.

Find the right account, organise vault entries and prepare access for a colleague in the same conversation. Passbolt MCP connects your assistant to your vault and keeps passwords out of MCP responses. When a task needs a secret, the server delivers it to a handler you have configured.

Quick start · Connect your assistant · Issues

Try asking your assistant:

Find the vault entry for https://example.com. Show its name and username so I can choose the right account.

What you can do

The protected endpoint walkthrough includes private configuration, example requests and the local passbolt-mcp-doctor command.

Related MCP server: Bitwarden MCP Server

Quick start

Prefer a ready package? Install the release and generate your client configuration. No source checkout is required.

Install Python 3.12–3.14 and uv. Clone with Git or download the ZIP. With a ZIP, open the extracted directory and skip the first two commands.

Install GnuPG and prepare a Passbolt service account, its GPG key and passphrase, and the vault access policy. Follow the Passbolt setup instructions before running the commands below. The configuration wizard uses these prepared files; it does not create the account or its keys.

git clone https://github.com/zai-one/passbolt-mcp.git
cd passbolt-mcp
uv sync --frozen --extra standalone
uv run --frozen --extra standalone python scripts/configure.py
uv run --frozen --extra standalone passbolt-mcp --config mcp.local.json --check-config

The wizard creates a local configuration and stores secrets in private files. It refuses to overwrite an existing setup. --check-config validates local settings; the first request below checks your account connection.

Connect your assistant

Add this configuration to an MCP client that uses mcpServers, such as Claude Desktop or Cursor. Replace /ABSOLUTE/PATH/ with your absolute path; Windows JSON paths can use forward slashes, such as D:/Tools/.

{
  "mcpServers": {
    "passbolt": {
      "command": "uv",
      "args": [
        "--directory",
        "/ABSOLUTE/PATH/passbolt-mcp",
        "run",
        "--frozen",
        "--extra",
        "standalone",
        "passbolt-mcp",
        "--config",
        "/ABSOLUTE/PATH/passbolt-mcp/mcp.local.json"
      ]
    }
  }
}

The client starts the MCP server for you. Refresh its tool list, then make your first request. For clients with a different config format, reuse the same command and args; uv must be available to the client process.

First request

Check the Passbolt connection with passbolt_status, then search for an entry I have access to.

A configured vault returns connection/decryption status and matching entry metadata. Passwords are not part of the response. If setup is incomplete, use the status fields to finish the vault configuration.

If tools do not appear, check the absolute path, whether the client can find uv, and the --check-config result. For access errors, check account credentials and permissions. Installation and troubleshooting.

Access and limits

Start with read-only access. Creating or updating entries and delivering secrets require separate permissions and server settings. The handler is configured by the operator; this project does not automatically fill every browser or application login form.

New secrets must be generated by the server or supplied through a separately prepared server-side file. This version has no MCP tool for submitting a plaintext password. See secret handling and approvals.

Authenticated HTTP is available for a server deployment. See HTTP setup, configuration and permissions and Python package integration.

uv sync --frozen --all-groups --extra standalone
uv run --frozen --extra standalone python scripts/verify.py
uv run --frozen --extra standalone python scripts/verify_install.py

Tests use synthetic fixtures. A passing test run does not establish live provider connectivity.

Built by ZAI.ONE

ZAI.ONE is a digital agency working on websites, SEO, advertising and analytics. We also build tools that connect AI assistants to everyday work. Talk to us on Telegram about setup, automation or an integration for your team.

Use and feedback

You may install and use this project for your own accounts under LicenseRef-ZAI-ONE. This is not an open-source license. Third-party notices remain in NOTICE. If it helps, give the repository a ⭐. Missing something or found a bug? Open an issue. I'm working on this project; accepted improvements are implemented here. Support is not guaranteed.