FlowRunner
PricingContact
Theme
Start Free

Pingdom

Developer Tools

Connect AI agents to Pingdom uptime monitoring. Agents create, update, and pause checks as infrastructure changes, pull uptime, outage, and performance summaries for SLA reporting, inspect per-probe raw results, and audit alerting actions.

15 actions API key available
Pingdom website ↗ Platform Documentation ↗ Capability data verified 2026-07-27
A scheduled sweep polls List Checks and finds a check reporting down
Get Raw Results pulls the per-probe rows for the last few minutes
Agent confirms probes in multiple regions report the failure, not one flaky location
List Maintenance Windows rules out planned downtime before anyone is paged
Agent opens a PagerDuty incident with the probe evidence attached
Ops channel receives the outage window, affected regions, and response times
The on-call engineer decides to escalate or stand down, with raw probe data in front of them

What This Integration Enables

Pingdom is the outside-in view: geographically distributed probes measuring what your customers actually experienced, not what your servers claim. FlowRunner agents work that data in both directions. As an evidence source, they pull average, outage, and performance summaries for SLA reporting, drill into raw per-probe results to see exactly which locations failed, and audit the alerting actions log to answer who was notified, how, and when. As a control surface, they create, update, pause, and retire uptime and transaction checks as infrastructure changes, so monitoring coverage tracks reality instead of trailing it. The connector has no event triggers; flows drive it on a schedule or from other systems' events, polling List Checks and the summary operations for state. That fits how the data is best used: verified, in context, and on a cadence you control.

Without FlowRunner

Alert then scramble A down alert arrives with no evidence of which regions actually failed
Blips that page people Single-probe noise wakes the on-call for outages that never happened
SLA math by hand Quarterly uptime numbers get assembled from screenshots and memory

With FlowRunner

Evidence travels with the page Per-probe results and outage windows are attached to every escalation
Verified before anyone wakes Multi-region confirmation and maintenance windows are checked first
Reports straight from the source Get Average Summary and Get Performance Summary feed SLA reports on a schedule

Use Case Scenarios

Verified Outage Escalation

A sweep finds the checkout check down. Before paging anyone, the agent pulls Get Raw Results and sees probes in three regions reporting failures, confirms no window in List Maintenance Windows covers it, and pulls Get Outage Summary for the incident start time. Then it opens the incident in [PagerDuty](/integrations/pagerduty) and posts the evidence to [Slack](/integrations/slack). The on-call engineer starts with region data and a timeline instead of a bare red alert.

SLA Reporting on a Schedule

Monthly, the agent walks every customer-facing check, pulls Get Average Summary for uptime and response times and Get Performance Summary for the weekly buckets, and appends the numbers to the SLA workbook in [Google Sheets](/integrations/google-sheets). Downtime windows come from Get Outage Summary, so disputed minutes can be traced to specific state changes. The report the account team sends is the same data Pingdom recorded, untouched by hand.

Monitoring That Tracks the Infrastructure

When a new service ships, the agent runs Create Check with the right check type, resolution, and probe regions, and tags it to its owning team. When a host migrates, Update Check repoints the hostname. Get Account Credits is consulted first so the agent knows whether plan capacity covers the new checks. And when infrastructure is decommissioned, the agent proposes retirements but does not execute them alone, because deleting a check is the one operation here that cannot be walked back.

Human-in-Loop Highlight

Delete Check permanently removes an uptime check and stops all monitoring for it, and it cannot be undone. The check's summaries are also the SLA evidence, so deleting the wrong one costs both coverage and history. After a migration, the cleanup agent therefore stops before deleting anything. It posts: "The migration removed four hosts. Checks 201 and 205 have returned no traffic in Get Raw Results since cutover; checks 203 and 209 still show live results. Delete Check cannot be undone. Confirm which to delete, and I will pause the rest with Update Check instead." The infrastructure owner makes the call. That [human-in-the-loop](/concepts/human-in-the-loop/) gate is the difference between tidy monitoring and a quiet hole in your coverage.

Agent processes routinely
Detects exception requiring judgment
Clear match Continues automatically
Ambiguous Routes to human via preferred channel
Human decides
Agent resumes with decision

Agent Capabilities

15 actions

Uptime Checks

5
  • List Checks Lists the account's uptime checks with id, name, hostname, type, status, resolution, and last response time, with optional tag filtering. The entry point for sweeps and audits.
  • Get Check Returns one check's full configuration: type-specific settings, notification thresholds, assigned probes, integrations, and tags.
  • Create Check Creates an uptime check of any supported type (HTTP, HTTP Custom, TCP, Ping, DNS, UDP, SMTP, POP3, IMAP) with resolution, notification behavior, response-time threshold, probe region filters, and tags.
  • Update Check Updates an existing check: rename it, repoint its hostname, change resolution, pause or resume it, or adjust notifications and tags. Empty fields are left unchanged.
  • Delete Check Permanently deletes a check and stops all monitoring for it. Cannot be undone, which is why decommission flows route it through a human gate.

Reporting and Analysis

4
  • Get Average Summary Returns average response time and optional uptime and downtime totals over a range, with country and hour-of-day breakdowns. The SLA reporting workhorse.
  • Get Outage Summary Returns the state changes for a check over a range, each with status and start and end timestamps, so incident durations and downtime windows can be computed exactly.
  • Get Performance Summary Returns response time and uptime metrics bucketed by hour, day, or week, ready for uptime graphs and trend reports.
  • Get Raw Results Returns individual per-probe test rows with probe id, timestamp, status, and response time. The ground truth for verifying which locations actually failed.

Transaction Monitoring

2
  • List Transaction Checks Lists Transaction Monitoring (TMS) checks, each simulating a multi-step user journey such as login and checkout, with status and average response time.
  • Get Transaction Check Returns one transaction check's full configuration, including its ordered steps, probe region, interval, and severity.

Account and Operations

4
  • List Maintenance Windows Lists scheduled maintenance windows during which monitoring is suppressed, so planned downtime is ruled out before an escalation.
  • Get Account Credits Returns current plan usage: how many uptime checks, SMS credits, and transaction checks are available versus used. Checked before creating new coverage.
  • List Actions Returns the alerting actions log: every notification Pingdom sent, filterable by time, check, channel, and delivery status. The audit trail for who was alerted and when.
  • List Probes Lists Pingdom's global probe servers with location and status. Probe ids scope summaries and raw results to specific regions.

Frequently Asked Questions

What can FlowRunner do with Pingdom?

FlowRunner agents can run List Checks, Get Check, and Create Check in Pingdom, plus 12 more actions.

Does connecting Pingdom to FlowRunner require OAuth?

No. Pingdom connects to FlowRunner with an API key, no OAuth flow required.

Can Pingdom trigger a FlowRunner workflow automatically?

Pingdom doesn't currently expose triggers in FlowRunner. It connects as an action step inside workflows started by another trigger.

Start building with Pingdom

$100 in credits. No card required. Connect in minutes.