check_dns
Resolve a hostname and report its DNS records (what the domain points to). Works with no Drumbeats account or API key — useful for "what does this domain resolve to?".
LABSCO SUMMARY
Monitoring you can try before you sign up, then run entirely from the assistant.
Leaving the three diagnostic tools keyless is a real design decision rather than a trial: "is this site up, and is its certificate about to expire" is answerable the moment the server is added. Once a key is in place the same conversation creates the monitor and acknowledges the incident it later raises.
What it is
The official MCP server for Drumbeats, which watches background jobs and endpoints — cron jobs, queues, scheduled tasks and HTTP checks — and raises an incident when something stops checking in. Monitors and incidents need an account; the three diagnostic checks do not.
What you get
- check_http, check_ssl and check_dns — is the URL reachable and how fast, when does the certificate expire and who issued it, and what does the hostname resolve to; all three work with no account and no key
- list_projects, create_project and update_project — the projects that group monitors, incidents and notification settings
- create_monitor — one tool for all four kinds: a cron job on a schedule, a basic event-driven job, a heartbeat, or an HTTP uptime check
- list_monitors, get_monitor and update_monitor — find a monitor, read its type, status and schedule, and patch only the fields you change
- pause_monitor and resume_monitor — stop checks and alerts, then restart them
- get_monitor_history — recent pings, warnings, checks or response times for one monitor
- get_uptime_summary — project-wide uptime and SLA across every uptime monitor, with per-monitor figures over a lookback window
- list_incidents and manage_incident — downtime and missed-run events newest first, filterable by status, and one tool to read, acknowledge or resolve one
Requirements
Node.js 22 when running via npx -y @drumbeats/mcp over stdio; Claude Desktop can install the .mcpb bundle instead. The monitoring tools need an account-scoped API key (dk_...) in DRUMBEATS_API_KEY, from Account > API keys; creating projects also needs the manage_projects scope. Leave the key unset and the HTTP, SSL and DNS checks still work. Self-hosted deployments override the service URLs through DRUMBEATS_API_BASE_URL and the per-service variables.
Setup effort
One command — npx -y @drumbeats/mcp
USE CASES
01
Checking a site right now
Confirm the URL responds, the certificate is not expiring, and DNS points where you expect — with no account.
02
Making sure a nightly job actually ran
Create a cron monitor on the schedule it should keep, then read its ping history.
03
Reporting uptime for the month
Pull the project-wide SLA rollup with per-monitor figures over the window.
04
Handling an outage
List the open incidents, acknowledge the live one, and resolve it once the fix lands.
05
Quieting alerts during a deploy
Pause the affected monitor and resume it when the window closes.
16 tools
Counted once with only startup setup. Other versions, modes or optional credentials may expose a different list.
check_http
Check whether an http(s) URL is reachable and report its status code and response time. Works with no Drumbeats account or API key — handy for a quick "is this site up?" before signing up.
check_ssl
Inspect a host's TLS/SSL certificate — validity, expiry date, and issuer. Works with no Drumbeats account or API key — useful for "is my SSL certificate about to expire?".
create_monitor
Create a Drumbeats monitor. One tool for all four types, chosen via "type":
- JOB_CRON: a job that should run on a cron schedule (schedule = cron expression).
- JOB_BASIC: an event-driven job with no fixed schedule (e.g. triggered by user action). "schedule" is required but ignored for timing — pass any placeholder like "@event". grace_period_seconds and schedule_tolerance don't apply; only failure_tolerance (and, if max_duration_seconds is set, a run-timeout) count against it.
- JOB_HEARTBEAT: a passive endpoint your job pings (schedule = expected interval).
- UPTIME_HTTP: Drumbeats polls a URL (schedule = poll interval; requires uptime_url).
All uptime_* fields are only valid when type is UPTIME_HTTP. Wire alerts at create time with notification_channel_ids / notification_group_ids (UUIDs discoverable via list_projects include=channels,groups). Notification channel types: EMAIL, SLACK, TELEGRAM, DISCORD, WEBHOOK, PAGERDUTY, SMS, PUSH.
create_project
Create a new Drumbeats project. A project groups monitors, incidents, and notification settings. Requires an account-scoped API key with the manage_projects scope. Returns the created project id, which you can then pass to create_monitor. Creating a project may hit your plan's project limit — the result will say so and point to the upgrade path.
get_monitor
Get a single Drumbeats monitor by id, including its type, status, schedule, and alerting metadata.
get_monitor_history
Fetch a monitor's recent history. One tool with a "kind" argument: pings | warnings | checks | response_times. Defaults to pings. Use it to answer "what happened with this monitor lately?".
get_uptime_summary
Get the project-level uptime / SLA summary across all of a project's uptime monitors: overall uptime percentage plus per-monitor uptime, check counts, and response times over a look-back window. The answer to "what is my uptime this month?".
list_incidents
List incidents for a Drumbeats project (downtime / missed-run events), newest first and paginated. Requires a project_id from list_projects. Filter by status (OPEN, ACKNOWLEDGED, RESOLVED) or a single monitor.
list_monitors
List all monitors in a Drumbeats project (jobs and uptime checks), with their type, status, and schedule metadata. Requires a project_id from list_projects. Use it to find a monitor id before fetching details, history, or incidents.
list_projects
List all Drumbeats projects the configured account API key can access. Returns each project id, name, plan, and basic metadata — use it to find the target project id before creating or querying monitors. Pass include=["channels","groups"] to also return each project's notification channels and groups, whose ids are needed to wire alerts when creating a monitor.
manage_incident
Triage a Drumbeats incident: get its details, acknowledge it, or resolve it. One tool with an "action" argument.
pause_monitor
Pause a monitor by id. It stops running checks and sending alerts until resumed.
resume_monitor
Resume a paused monitor by id.
update_monitor
Update an existing Drumbeats monitor. Provide monitor_id plus only the fields to change. Set status to PAUSED or UP to pause/resume (or use pause_monitor / resume_monitor). The uptime_* fields apply to UPTIME_HTTP monitors.
update_project
Update an existing Drumbeats project. Provide project_id plus only the fields to change (name and/or description). Requires an API key with the manage_projects scope and MANAGER or OWNER role on the project.
BEFORE YOU INSTALL
What to know first
Tools you get
16 tools when we connected. check_http, check_ssl and check_dns work without an account or key; the other thirteen need DRUMBEATS_API_KEY, and create_project needs a key with manage_projects.
Requires
npx on PATH
Optional
DRUMBEATS_API_KEY in your environment
Where it runs
The documented client runs the server on your machine.
CONTEXT COST
Moderate Standing cost
Rated by the size of the 16 tool definitions that ride along on every turn, not by how many there are
INSTALLATION
Connect your client
Server setup
npx
TERMINAL
npx -y @drumbeats/mcpHosted endpoint
ENDPOINT URL
https://api.drumbeats.io/mcpAdd this hosted endpoint to your MCP client.
MCP Clients
Any MCP client
ANY MCP CLIENT · JSON
{
"mcpServers": {
"drumbeats": {
"command": "npx",
"args": [
"-y",
"@drumbeats/mcp"
],
"env": {
"DRUMBEATS_API_KEY": "<YOUR_DRUMBEATS_API_KEY>"
}
}
}
}
