Skip to main content
Glama
WatchGuard Cloud MCP Server

WatchGuard Cloud MCP Server

by WYRE-AI

WatchGuard Cloud MCP Server

MCP server for WatchGuard Cloud's REST API - MSP account management, device/license activation, and operator (user) management - for AI assistants and the WYRE Conduit gateway.

Authentication

Enable API access in WatchGuard Cloud (Account Settings > API Access) to receive two independent credentials:

  1. An AccessID/Password pair, exchanged for a short-lived (1-hour) OAuth access token via POST /oauth/token (grant_type=client_credentials, HTTP Basic auth). This server handles that exchange, and its refresh on expiry, internally.

  2. A separate WatchGuard-API-Key, sent as its own header on every request alongside the OAuth bearer token.

Callers only ever need to supply the three long-lived values below - not the short-lived access token.

Related MCP server: mcp-server-action1

Configuration

Copy .env.example and set the variables below. Leave CONDUIT_S2S_SECRET empty in that file.

Env var

Description

WATCHGUARDCLOUD_ACCESS_ID

AccessID from WatchGuard Cloud API Access. Required for stdio and AUTH_MODE=env.

WATCHGUARDCLOUD_PASSWORD

Password paired with the AccessID. Required for stdio and AUTH_MODE=env.

WATCHGUARDCLOUD_API_KEY

WatchGuard-API-Key generated when API access is enabled. Required for stdio and AUTH_MODE=env.

WATCHGUARDCLOUD_REGION

usa (default), deu (EMEA), or jpn (APAC) - your account's data-storage region.

MCP_TRANSPORT

stdio (default) or http.

AUTH_MODE

env (default, reads the vars above) or gateway (credentials arrive per-request via X-WatchGuardCloud-* headers, injected by the Conduit gateway). The container image and Compose set gateway. Gateway mode never falls back to WATCHGUARDCLOUD_*.

CONDUIT_S2S_SECRET

Required to start the HTTP transport. The process logs an error and exits non-zero when this is empty or a known placeholder (changeme, your-s2s-secret, secret, placeholder, and the other example values). Compose will not start when it is unset. Never commit a real value. The value is never logged.

MCP_ALLOW_INSECURE_DEV

Set to exactly 1 only for local development to start HTTP without CONDUIT_S2S_SECRET. The process logs a warning, does not check X-Gateway-S2S, and binds 127.0.0.1 only. A non-loopback MCP_HTTP_HOST (including 0.0.0.0) is ignored in that mode. A placeholder secret is still refused.

MCP_HTTP_HOST

Bind address when a real CONDUIT_S2S_SECRET is set. Unset keeps the historical default of 0.0.0.0. The container image and Compose set 0.0.0.0. Without a real secret, the listener is 127.0.0.1 only.

MCP_HTTP_PORT

HTTP port. Defaults to 8080.

LOG_LEVEL

debug | info (default) | warn | error.

/health stays unauthenticated and does not read vendor credentials. The stdio transport does not use CONDUIT_S2S_SECRET.

HTTP authentication

With MCP_TRANSPORT=http and a real CONDUIT_S2S_SECRET, every /mcp request must include X-Gateway-S2S:

X-Gateway-S2S: t=<unix seconds>,v1=<hex hmac-sha256>

v1 is the hex HMAC-SHA256 of the literal string t=<unix seconds> keyed with CONDUIT_S2S_SECRET. A missing or invalid header returns 401. The timestamp must be within five minutes of the server clock.

The gateway signs each /mcp call:

t=$(date +%s)
sig=$(printf 't=%s' "$t" | openssl dgst -sha256 -hmac "$CONDUIT_S2S_SECRET" -hex | awk '{print $NF}')
curl -sS -X POST http://127.0.0.1:8080/mcp \
  -H "content-type: application/json" \
  -H "accept: application/json, text/event-stream" \
  -H "X-Gateway-S2S: t=${t},v1=${sig}" \
  -H "X-WatchGuardCloud-Access-ID: $WATCHGUARDCLOUD_ACCESS_ID" \
  -H "X-WatchGuardCloud-Password: $WATCHGUARDCLOUD_PASSWORD" \
  -H "X-WatchGuardCloud-API-Key: $WATCHGUARDCLOUD_API_KEY" \
  -H "X-WatchGuardCloud-Region: usa" \
  -d '{"jsonrpc":"2.0","id":1,"method":"tools/list"}'

In AUTH_MODE=gateway, /mcp also requires X-WatchGuardCloud-Access-ID, X-WatchGuardCloud-Password, and X-WatchGuardCloud-API-Key (X-WatchGuardCloud-Region is optional). A request missing any of those three returns 401. Those headers are the only credential source for that request.

AUTH_MODE=env still requires a valid X-Gateway-S2S header, then uses WATCHGUARDCLOUD_* from the process environment.

Request limits

  • HTTP request bodies are capped at 256 KiB (262144 bytes). After the X-Gateway-S2S check passes, a Content-Length above the cap returns 413 with a JSON-RPC error; bodies without Content-Length are capped by the MCP SDK transport. Unauthenticated requests still get 401 first.

  • tools/call arguments are capped at 10,000 elements (every array item and object key at any depth). Larger inputs return an isError tool result and are never dispatched.

These limits address CVE-2026-111768 / GHSA-46jr-4wx8-v453 (memory exhaustion during validation) and GHSA-f897-62hf-6gxq (no request body / batch cap).

Tools

Accounts

  • watchguardcloud_get_account - get an account's details.

  • watchguardcloud_create_account - create a managed (child) account.

  • watchguardcloud_update_account - update a managed account's name/primary contact.

  • watchguardcloud_delete_account - delete a managed account.

  • watchguardcloud_list_managed_accounts - list accounts managed by (delegated from) an account.

Activations

  • watchguardcloud_create_activation - activate a Firebox/hardware device or SaaS license by activation key.

  • watchguardcloud_list_recent_activations - list activation batches from roughly the last 30 days.

  • watchguardcloud_get_activation_status - per-line-item status for an activation batch.

Operator Management

  • watchguardcloud_create_operator - create an operator (user) account with a privilege role.

  • watchguardcloud_update_operator - update an operator's name/phone/role.

  • watchguardcloud_delete_operator - delete an operator account.

  • watchguardcloud_get_operator_transaction_status - poll an async create/update/delete result.

  • watchguardcloud_list_operators - list every operator and its role/MFA status for an account.

Scope

This is a v1 surface covering the three best-documented WatchGuard Cloud platform APIs (account, activation, and operator management). Explicitly out of scope for now: Firebox device configuration/management and Endpoint Security policy management - both exist as separate WatchGuard APIs, but their public documentation does not (at the time of writing) carry enough concrete request/response detail to implement with confidence. They can be added as a follow-up once better-documented.

Development

npm install
npm run build
npm test
npm run lint   # tsc --noEmit

HTTP refuses to start until CONDUIT_S2S_SECRET is a real secret:

CONDUIT_S2S_SECRET="$CONDUIT_S2S_SECRET" npm run start:http

Local HTTP without a service-to-service secret is development-only. It binds 127.0.0.1 and logs a warning:

MCP_ALLOW_INSECURE_DEV=1 npm run start:http

Docker

The image listens on 0.0.0.0 inside the container (MCP_HTTP_HOST) and defaults to AUTH_MODE=gateway. Publish the port on loopback and pass the S2S secret. In gateway mode, WatchGuard credentials arrive on each request as headers; they are not read from the environment.

docker build -t watchguard-cloud-mcp .
docker run \
  -e CONDUIT_S2S_SECRET="$CONDUIT_S2S_SECRET" \
  -p 127.0.0.1:8080:8080 \
  watchguard-cloud-mcp

Compose refuses to start when CONDUIT_S2S_SECRET is unset, sets AUTH_MODE=gateway, publishes 127.0.0.1:8080:8080, and does not inject static vendor credentials:

CONDUIT_S2S_SECRET="$CONDUIT_S2S_SECRET" docker compose up

Single-tenant HTTP (AUTH_MODE=env) still requires the S2S secret, then uses WATCHGUARDCLOUD_ACCESS_ID, WATCHGUARDCLOUD_PASSWORD, and WATCHGUARDCLOUD_API_KEY. This is separate from the gateway Compose service:

docker run \
  -e CONDUIT_S2S_SECRET="$CONDUIT_S2S_SECRET" \
  -e AUTH_MODE=env \
  -e WATCHGUARDCLOUD_ACCESS_ID="$WATCHGUARDCLOUD_ACCESS_ID" \
  -e WATCHGUARDCLOUD_PASSWORD="$WATCHGUARDCLOUD_PASSWORD" \
  -e WATCHGUARDCLOUD_API_KEY="$WATCHGUARDCLOUD_API_KEY" \
  -e WATCHGUARDCLOUD_REGION=usa \
  -p 127.0.0.1:8080:8080 \
  watchguard-cloud-mcp

Available Tools

13 tools
watchguardcloud_create_accountB

Create a new managed (child) account under a parent WatchGuard Cloud account.

Inputs: name (required), type (required), email (required), lastName (required), +2 more.

watchguardcloud_create_activationB

Activate one or more Firebox/hardware devices or SaaS licenses by activation key.

Inputs: items (required).

watchguardcloud_create_operatorB

Create a WatchGuard Cloud operator (user) account with an assigned privilege role. Role options depend on whether the account is a Service Provider (OWNER, SALES, HELPDESK, AUDITOR, NO_ACCESS) or a Subscriber (ADMINISTRATOR, ANALYST, OBSERVER, NO_ACCESS).

Inputs: role (required), email (required), phone (required), lastName (required), +4 more.

watchguardcloud_delete_accountA

Delete a managed WatchGuard Cloud account. Set force to also delete its child accounts.

Inputs: accountId (required), force.

watchguardcloud_delete_operatorB

Delete a WatchGuard Cloud operator (user) account.

Inputs: username (required), accountId (required).

watchguardcloud_get_accountA

Get a WatchGuard Cloud account by ID, optionally expanded with contacts/addresses/service properties.

Inputs: accountId (required), fields.

watchguardcloud_get_activation_statusB

Get per-line-item status for an activation batch by batch ID.

Inputs: batchId (required), limit, offset, sortBy.

watchguardcloud_get_operator_transaction_statusA

Poll the async result of a prior create/update/delete operator call by transaction ID.

Inputs: transactionId (required).

watchguardcloud_list_managed_accountsB

List accounts managed by (delegated from) a WatchGuard Cloud account, with paging/sorting/filtering.

Inputs: accountId (required), name, type, limit, +4 more.

watchguardcloud_list_operatorsA

List every operator (user) account and its role/MFA status for a WatchGuard Cloud account.

Inputs: accountId (required).

watchguardcloud_list_recent_activationsB

List activation batches submitted in roughly the last 30 days for the account.

Inputs: limit, offset, sortBy.

watchguardcloud_update_accountB

Update a managed account's name and primary contact.

Inputs: name (required), email (required), lastName (required), accountId (required), +1 more.

watchguardcloud_update_operatorB

Update an existing WatchGuard Cloud operator's name, phone, or privilege role.

Inputs: username (required), accountId (required), role, phone, +2 more.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    A3.7/5.0

    Scored across 13 tools

    Disambiguation5/5

    Each tool targets a clear resource-action combination: accounts, activations, or operators. Even the two status tools are distinguishable by batch/transaction context. There is no meaningful overlap between tool purposes.

    Naming Consistency5/5

    All tools follow the consistent watchguardcloud_ verb_noun pattern. Verbs are get, create, update, delete, list, which map predictably to the resource nouns. This makes the toolset easy to navigate.

    Tool Count5/5

    13 tools is well-scoped for a cloud management server covering accounts, activations, and operators. Each tool handles a distinct operation, and none feel redundant or unnecessary. The count supports the domain without becoming unwieldy.

    Completeness5/5

    The toolset covers the main lifecycle operations for accounts, activations, and operators: create, read, update, delete, and list where applicable. Async operator operations are supported via transaction status polling, closing an important workflow gap. The domain appears thoroughly covered for its stated purpose.

    Maintenance

    ActivityMaintained
    ResponsivenessNo recent community issues

    Related MCP Connectors

    Related MCP Servers

    • F
      license
      Not graded
      quality
      D
      maintenance
      Enables AI agents to investigate threats, pull reports, manage security policies, and administer deployment infrastructure through the Cisco Secure Access REST API.
      -
    • A
      license
      A
      quality
      C
      maintenance
      Enables management of endpoint inventory, patching, and vulnerabilities through the Action1 REST API, with read-only access by default and optional write support for update approvals.
      15
      1
      Apache 2.0
    • A
      license
      Not graded
      quality
      B
      maintenance
      Enables AI assistants to manage MSP-multi-tenant email security: organizations, domains, users, licensing, features, subscription packages, and email flow reporting.
      Apache 2.0
    • A
      license
      Not graded
      quality
      B
      maintenance
      Enables AI agents and applications to interact with UniFi network infrastructure through the UniFi Network Controller API, supporting device, client, network, firewall, QoS, backup, multi-site, and topology management.
      Apache 2.0