Commvault MCP Server

A Model Context Protocol (MCP) server for seamless integration with Commvault environments. This server enables AI agents to securely access and manage job details, commcell metrics, client and storage information, user permissions, plan configurations, and backup schedules.

Features

The Commvault MCP Server enables seamless integration with Commvault environments, offering the following:

Prerequisites

Before running the Commvault MCP Server, ensure the following requirements are met:

1. Python Environment

• Python 3.11 or higher
• `uv` package manager (used for dependency management and running the server)

2. Authentication & Security Configuration

The Commvault MCP Server supports two authentication methods:

Option 1: OAuth Authentication

> Note: OAuth authentication is only supported for Commvault environments running SP42 CU 27 and above.

> OAuth must be properly configured in the CommServe before using this option.

When using OAuth authentication, you'll need:

• Discovery Endpoint URL: The OAuth discovery/metadata endpoint
• Client ID: Your OAuth application's client identifier
• Client Secret: Your OAuth application's client secret
• Required Scopes: Required OAuth scopes
• Base URL: Base URL of the MCP Server

> Important: The redirect URI must be set to `OAUTH_BASE_URL/auth/callback` in your OAuth provider's app/client configuration.

Option 2: Traditional Token-Based Authentication

The following values will be collected during the setup process:

• Commvault Access Credentials:

You need a valid `access_token` and `refresh_token` to authenticate with the Commvault API.

Learn how to generate these tokens here: Creating an Access Token – Commvault Docs

• Secret Key:

This secret must be included by the MCP Client in the `Authorization` header of all tool requests.

It acts as a security layer for tool access in remote server. You can set your own.

> Important: When using traditional token-based authentication, the setup script requires a secure, OS-native keyring backend to store sensitive credentials securely. Only secure backends are allowed for security reasons.

> Running multiple MCP servers on the same host? The OS keyring is per-user, so by default every install writes to the same keyring slots and the most recent `setup.py` run overwrites the previous install's credentials. Set a unique `MCP_INSTANCE_ID` per install — see Running Multiple Instances on the Same Host under *Additional Configuration*.

Supported Secure Backends by Platform

Setup

1. Clone the repository

git clone https://github.com/Commvault/commvault-mcp-server.git
cd commvault-mcp-server

2. Run the Setup Script

The setup script will guide you through configuration options including:

• Metallic or on-premise deployment — for Metallic setups, the gateway URL is configured automatically
• Transport mode (stdio, streamable-http, or sse)
• Server connection details (for remote modes)
• Authentication method (traditional tokens or OAuth)
• OAuth configuration (if selected)
• Secure token storage

uv run setup.py

Metallic (Cloud) Setup

When prompted "Is this a Metallic setup?", select yes. The setup script will automatically configure:

• Gateway URL: `https://api.metallic.io`

No additional configuration is needed for the server URL; all other setup steps (transport mode, authentication, etc.) proceed as normal.

3. Start the MCP Server

uv run -m src.server

Secure Production Deployment

For production deployments, it is recommended to use a reverse proxy with TLS/HTTPS and security headers. The MCP server should bind to `127.0.0.1` (localhost only) to prevent direct public access.

Quick Setup

1. Configure MCP Server for localhost: Set `MCP_HOST=127.0.0.1` in your `.env` file

2. Install reverse proxy: Choose nginx or Caddy

3. Configure TLS: Use Let's Encrypt for automatic certificate management

4. Add security headers: Configure OWASP-recommended headers (HSTS, X-Frame-Options, CSP, etc.)

5. Update client config: Use `https://your-domain.com/mcp` instead of `http://HOST:PORT/mcp`

Running with Docker

The Docker image runs the server in OAuth mode only.

> Prerequisites: Docker installed, and a Commvault environment running SP42 CU 27 or above with OAuth configured.

1. Generate a `.env` file

Run the interactive setup script on your host machine. When prompted for the authentication method, select OAuth.

uv run setup.py

This produces a `.env` file with all required variables, including the OAuth endpoints and credentials.

2. Build the image

docker build -t commvault-mcp-server .

3. Run the container

Pass the generated `.env` file to the container at runtime:

docker run --rm -p 9090:9090 --env-file .env commvault-mcp-server

Configuring Clients

> Note: `npx` is required while using Token-Based Authentication. You can get it by installing Node.js which includes `npx` by default.

While using OAuth

Refer to your AI client’s documentation for integration steps. For example, Claude requires specifying a server name and the MCP server URL in its connector configuration.

Remote MCP Server (Streamable HTTP / SSE)

{
  "mcpServers": {
    "Commvault": {
      "command": "npx",
      "args": ["mcp-remote", "HOST:PORT/mcp", "--header", "Authorization: "]
    }
  }
}

Remote MCP Server (Client on Windows)

{
  "mcpServers": {
    "Commvault": {
      "command": "cmd",
      "args": ["/c", "npx", "mcp-remote", "HOST:PORT/mcp", "--header", "Authorization: "]
    }
  }
}

Remote MCP Server (HTTP)

{
  "mcpServers": {
    "Commvault": {
      "command": "npx",
      "args": ["mcp-remote", "HOST:PORT/mcp", "--header", "Authorization: ", "--allow-http"]
    }
  }
}

Local MCP Server (STDIO) - Unix

{
  "mcpServers": {
    "Commvault": {
      "command": "C:\\YOUR\\PATH\\TO\\commvault-mcp-server\\.venv\\bin\\python",
      "args": [
        "C:\\YOUR\\PATH\\TO\\commvault-mcp-server\\src\\server.py"
      ]
    }
  }
}

Local MCP Server (STDIO) - Windows

{
  "mcpServers": {
    "Commvault": {
      "command": "C:\\YOUR\\PATH\\TO\\commvault-mcp-server\\.venv\\Scripts\\python.exe",
      "args": [
        "C:\\YOUR\\PATH\\TO\\commvault-mcp-server\\src\\server.py"
      ]
    }
  }
}

Additional Configuration

> Note: These are optional configurations that extend the MCP server's capabilities.

Running Multiple Instances on the Same Host

The OS keyring used to store the server secret and Commvault API tokens is per-user. By default every install of the MCP server on a given host writes to the same keyring entries under the service name `commvault-mcp-server`. If you run two or more MCP server installs under the same OS user, each rerun of `setup.py` silently overwrites the other instances' credentials, which typically surfaces as a `401 Invalid token` after a restart or reboot of the unaffected instance.

To run multiple MCP server instances on the same host under the same OS user, set a unique `MCP_INSTANCE_ID` per install. The setup script prompts for it at the start; you can also set it directly in each install's `.env`:

# Instance A's .env
MCP_INSTANCE_ID=prod

# Instance B's .env
MCP_INSTANCE_ID=dr

With an instance ID set, the keyring service name becomes `commvault-mcp-server:`, isolating each install's `server_secret`, `server_secret_expiry`, `access_token`, and `refresh_token`.

Rules and notes

• Allowed characters: letters, digits, `.`, `-`, `_` (max 32 chars).
• Leaving `MCP_INSTANCE_ID` unset (or setting it to `default`) preserves the original behaviour and reuses the existing `commvault-mcp-server` keyring entries, so existing single-instance installs do not need to do anything.
• Each install should still live in its own directory with its own `.env` so transport mode, port, `CC_SERVER_URL`, etc. don't collide. The instance ID only namespaces the keyring.
• After changing `MCP_INSTANCE_ID` for an existing install, rerun `uv run setup.py` so the secret and tokens are written under the new namespaced service name.
• Alternative isolation strategies that also work: run each instance under a dedicated OS user, or in its own host/VM/container.

Trusted Proxy Configuration

When deployed behind a reverse proxy or load balancer, configure `TRUSTED_PROXY_IPS` environment variable with comma-separated proxy IP addresses to enable per-client rate limiting using the `X-Forwarded-For` header. For example:

export TRUSTED_PROXY_IPS="10.0.0.1,10.0.0.2,192.168.1.100"

Salesforce Backup Integration

The Salesforce integration enables browsing of backed-up Salesforce records stored in Commvault. It provides two tools:

Prerequisites

1. Environment Variable: Set `ENABLE_SALESFORCE_TOOLS=true` in your environment

2. Commvault Salesforce Backup: At least one Salesforce organisation must be configured and backed up in Commvault

Usage Example

To browse backed-up `Account` records for a Salesforce org:

Get all backed-up Account records for Salesforce org 00D2w000005mBCpEAM

The tool will automatically resolve the org ID to a Commvault client and return the latest backed-up records.

Parameters for `get_salesforce_records`

DocuSign Backup Integration

The DocuSign backup integration enables backup of completed DocuSign envelopes to a Commvault S3 vault. This integration provides comprehensive document management capabilities including backup, listing, and restore operations.

Prerequisites

1. Environment Variable: Set `ENABLE_DOCUSIGN_TOOLS=true` in your environment

2. Commvault S3 Vault: Configure an S3 endpoint in Commvault

3. DocuSign API Access: Valid DocuSign integration credentials

Configuration Files

Create the following files in the `config/` directory:

1. DocuSign Configuration (`docusign_config.json`)

Based on the template file `config/docusign_config_template.json`:

{
  "docusign": {
    "integrationKey": "YOUR_INTEGRATION_KEY_HERE",
    "userId": "YOUR_USER_ID_HERE", 
    "authServer": "account-d.docusign.com",
    "scopes": "signature impersonation",
    "basePath": "https://demo.docusign.net/restapi"
  },
  "fromDate": "2024-07-01T00:00:00Z"
}

2. DocuSign Private Key (`docusign_key.pem`)

Place your DocuSign private key file in the `config/` directory as `docusign_key.pem`.

Note on LiteLLM

We do not recommend using LiteLLM as an MCP client. LiteLLM's per-turn session lifecycle triggers cosmetic `ClosedResourceError` tracebacks in the server terminal due to a race condition in the upstream `mcp` Python SDK — these are harmless but noisy.

Security Considerations

Multi-Server Trust Boundary (Tool Poisoning & Tool Shadowing)

The MCP protocol enforces no isolation between servers that share a single agent session. Any other MCP server running in the same session as the Commvault MCP has the practical ability to influence the agent's reasoning about Commvault operations, even without direct access to CommServe or any privileges against it.

Two attack paths follow from this architectural property:

• Tool Poisoning — A malicious server's tool description coerces the agent to read the operator's stored Commvault auth token from host-side storage (e.g. `~/.claude.json`, a project-level `.mcp.json`, or the OS keychain entry for `commvault-mcp-server`) and pass the token value as a parameter to an attacker-controlled tool, which then exfiltrates it.
• Tool Shadowing — A malicious server's tool description names Commvault tool identifiers (e.g. `kill_job`, `disable_schedule`, `set_user_enabled`) and instructs the agent to perform additional or modified Commvault operations whenever the operator legitimately invokes those tools. The Commvault MCP itself is the actor in these calls; its code and responses are not corrupted. The corruption lives in the agent's reasoning, induced by the third-party server's metadata.

Neither path requires any defect in the Commvault MCP Server code. Both are properties of the MCP architecture as currently specified.

> Recommendations for operators:

> - Install only first-party or independently audited MCP servers in agent sessions that include Commvault tools.

> - Where feasible, use an isolated agent session in which the Commvault MCP is the only integration.

> - Where isolation is not feasible, restrict the agent's general filesystem and outbound-network tool capabilities for the duration of the session.

Credential Hygiene

• Use short-lived access tokens with explicit refresh rather than long-lived bearer tokens. A bounded token lifetime limits the window during which an exfiltrated credential remains usable, and rotation invalidates leaked tokens before an attacker can act on them.
• Restrict outbound traffic from the MCP host. Where your environment supports it, apply a network policy that limits outbound HTTP/HTTPS from the host running the MCP server to a documented allowlist. Unexpected outbound calls from an MCP host are the principal exfiltration channel for this class of attack and the most reliable detection signal.

Contributing

• We're continuing to add more functionality to this MCP server. If you'd like to leave feedback, file a bug or provide a feature request, please open an issue on this repository.
• Pull requests are welcome! For major changes, please open an issue first to discuss what you would like to change.

License

This project is licensed under the Apache License. See the LICENSE file for details.