Pangea

Pangea MCP Server

A Model Context Protocol (MCP) server that provides integration with Pangea APIs.

Prerequisites

• Node.js v24.11.0 or greater.
• A Pangea API token with access to all of AI Guard, Domain Intel, Embargo, IP Intel, Redact, Secure Audit Log, URL Intel, and Vault. This token needs to be stored in Pangea Vault. See Service Tokens for documentation on how to create and manage Pangea API tokens.
• A Pangea API token with access to Vault. This will be used to fetch the above token at runtime.
• A Pangea Secure Audit Log configuration with the "Standard Audit Log Config" schema. See Multiple Service Configurations for documentation on how to create a new Secure Audit Log configuration. Note down the configuration ID for later.

The first API token may look like:

Pangea setup from scratch

This section will go over how to set up a brand new Pangea account for this MCP server.

1. Create a Pangea account at https://pangea.cloud/signup. During the account creation process, an organization (top-level group) and project (individual app) will be created as well. On the "Get started with a common service" dialog, just click on the Skip button to get redirected to the developer console.
2. In the developer console, there will be a list of services in the left hand panel. Click on the Vault service to enable it. Vault provides secure storage of secrets, cryptographic keys, and Pangea API tokens.
3. In the modal, there will be a prompt to create a new Pangea API token or to extend an existing one. Choose Create a new token and click on Done. This is the token that will be used later as the PANGEA_VAULT_TOKEN environment variable.
4. In the left hand panel, click on the Secure Audit Log service to enable it. Secure Audit Log provides transparent, immutable, and cryptographically verifiable tamperproof audit logging.
5. Continue with the default Standard Audit Log schema.
6. In the modal, there will be a prompt to create a new Pangea API token or to extend an existing one. Choose Create a new token. Ensure that the Store token in Vault checkbox is checked, then click Done.
7. The Config ID displayed on the destination page is what will be used later as the PANGEA_AUDIT_CONFIG_ID environment variable.
8. Now the second token that was created must be extended to the rest of Pangea's services. For each of the following services, click on its respective item in the left hand panel and, in the modal that appears, choose Extend an existing token. Select the token that was created in the previous step (not the first token that was created for Vault alone), then click Done.
   1. AI Guard — Protects data and interactions with LLMs.
   2. Domain Intel — Retrieves intelligence data for submitted domains. Supported providers: DomainTools, CrowdStrike, WhoisXML API.
   3. Embargo — Checks IPs and country codes against known sanction and trade embargo lists.
   4. IP Intel — IP monitoring. Supported providers: Team CYMRU, CrowdStrike, Digital Element.
   5. Redact — Controls sensitive information by performing redaction using defined rules.
   6. URL Intel — Retrieves intelligence about known URLs. Supported provider: CrowdStrike.
   7. Vault — Secure storage of secrets, cryptographic keys, and Pangea API tokens.
9. Navigate to https://console.pangea.cloud/project/credentials. The token that has been extended to all of the above services should be displayed with a "Stored" indicator under the Vault column. Click on that to navigate straight to the Vault item that contains that token. The ID of this item is what will be used as the PANGEA_VAULT_ITEM_ID environment variable.

Usage with Claude Desktop

Edit the following configuration file (create it if it does not exist):

• macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
• Windows: %APPDATA%\Claude\claude_desktop_config.json

Replace (or merge) the file contents with the following:

{
  "mcpServers": {
    "pangea": {
      "command": "npx",
      "args": ["-y", "@pangeacyber/mcp-server"],
      "env": {
        "PANGEA_VAULT_TOKEN": "pts_00000000000000000000000000000000",
        "PANGEA_VAULT_ITEM_ID": "pvi_00000000000000000000000000000000",
        "PANGEA_AUDIT_CONFIG_ID": "pci_00000000000000000000000000000000"
      }
    }
  }
}

1. Update the PANGEA_VAULT_TOKEN value to the Pangea Vault API token.
2. Update the PANGEA_VAULT_ITEM_ID value to the Vault item ID that contains the API token that will be used to call all other Pangea services.
3. Update the PANGEA_AUDIT_CONFIG_ID value to the Secure Audit Log configuration ID.
4. Restart Claude Desktop.

Running as an HTTP server

npx @pangeacyber/mcp-server --transport httpStream --port 8080

Adding Pangea AuthN

When running the server with the streamable HTTP transport, it is possible to optionally add authentication to the server via Pangea AuthN. To enable this feature, set the --authn flag and set the following environment variables:

• PANGEA_AUTHN_ISSUER — Pangea AuthN issuer URL.
• PANGEA_AUTHN_CLIENT_ID — Pangea AuthN client ID.
• PANGEA_AUTHN_CLIENT_SECRET — Pangea AuthN client secret.

Client setup

1. Navigate to the Pangea AuthN dashboard.
2. In the left hand panel, click on OAuth Server.
3. Click on the + OAuth Client button on the right to begin creating a new OAuth client.
   • Name: Assign a recognizable name to your client as it will appear in the list of clients in the OAuth Server settings. This name may be updated at any time.
   • Grant Type: must be Authorization Code.
   • Response Types: only Code is required.
   • Allowed Redirect URIs: enter http://localhost:8000/pangea/callback. Note that for a production MCP server, this should use the remote address of the server (e.g. https://mcp.example.org/pangea/callback) instead of a localhost address.
   • Allowed Scopes & Default Scopes: openid.
4. The Client ID and Client Secret should be the PANGEA_AUTHN_CLIENT_ID and PANGEA_AUTHN_CLIENT_SECRET environment variables respectively. The Hosted Login URL that is displayed on the AuthN Overview page is the PANGEA_AUTHN_ISSUER environment variable.

Tools

AI Guard

• prompt_guard — Analyze and redact text to avoid manipulation of the model, addition of malicious content, and other undesirable data transfers.

Domain Intel

• lookup_domain_reputation — Look up reputation score(s) for one or more domains.
• whois — Retrieve WHOIS (an Internet resource's registered users or assignees) for a domain.

Embargo

• check_ip_embargo — Check one or more IP addresses against known sanction and trade embargo lists.
• check_iso_code_embargo — Check a country code against known sanction and trade embargo lists.

File Intel

• lookup_file_reputation — Retrieve a reputation score for a set of file hashes.

IP Intel

• lookup_ip_address_reputation — Look up reputation score(s) for one or more IP addresses.
• lookup_domain_from_ip_address — Retrieve the domain name associated with one or more IP addresses.
• is_proxy — Determine if one or more IP addresses originate from a proxy.
• is_vpn — Determine if one or more IP addresses originate from a VPN.
• geolocate — Geolocate, or retrieve location information associated with, one or more IP addresses.

Redact

• redact — Redact sensitive information from provided text.

Secure Audit Log

• log_entry — Create a log entry in the Secure Audit Log.
• search_log — Search the Secure Audit Log.

URL Intel

• lookup_url_reputation — Look up reputation score(s) for one or more URLs.

Vault

• get_vault_item — Retrieve details for a Vault key, secret, token, or folder.
• list_vault_items — Retrieve an array of Vault items matching a given filter, including secrets, keys, tokens, and folders, along with their common details.
• delete_vault_item — Delete a Vault key, secret, token, or folder.
• generate_key — Generate a symmetric or asymmetric key.