← Explore all skills

OBSERVABILITY / Grafana Labs

Synthetic monitoring checks

Separates a reliability probe from a load test, and names the assertion mistake that leaves probe_success green.

“Author Grafana Cloud Synthetic Monitoring checks, with deep coverage of k6 scripted and browser checks: SM's single-VU/single-iteration execution model, assertions that actually fail probe_success (expect() and fail() vs bare check()), secrets, deterministic scripts, robust browser locators, local validation with k6 run, deployment via UI/API/Terraform, verifying probe_success, and rollback. …”

01 / THE REASONING

Why this made the selection.

  • The first heading says reliability monitoring, not load testing.
  • A heading names what agents get wrong about how an execution fails.

02 / THE REVIEW RECORD

What we actually inspected.

Source review has boundaries.
A clear record is more useful than a “safe” badge.

Material inspected

  • skills/grafana-cloud/synthetic-monitoring-checks/SKILL.md
  • LICENSE
  • README.md

Our findings

  • Headings include Execution model and Scripted check authoring.
  • The deployment section distinguishes local smoke, unsaved probe tests and saved checks, and documents millisecond API durations, base64 scripts and per-probe validation.
  • OpenAPI guidance says to monitor one to three customer journeys, not every path, and not to run a destructive operation against production only because the spec lists it.
  • The deployment guidance keeps Terraform-owned checks under Terraform changes to avoid drift, while permitting gcx for live inspection and ad-hoc test execution.

Not established by this review

  • The skill was not executed inside a host agent.
  • Not executed: it was not installed or run.

The review applies to the material and revision named here. A newer upstream release can change its behavior.

03 / PUT IT TO WORK

Add a skill to your workflow.

Upstream setup instructions ↗

DOCUMENTED COMMAND

npx skills add grafana/skills

Run in your terminal to choose your AI tools, project or global scope, and installation method. This short command follows the current upstream version; the review covers the revision shown above.

  1. Install the repository skills.
  2. Use expect or fail where the skill says a bare check does not fail the probe.

Before you start

  • A target you are allowed to probe

04 / FOLLOW THE EVIDENCE

The source trail.

Our notes are separate from the original resource.
Check upstream before adopting a new version.

  1. Skill definition ↗Checked

    https://github.com/grafana/skills/blob/1ccacf29049fde66637fe01ab93a83a772da323b/skills/grafana-cloud/synthetic-monitoring-checks/SKILL.md

    Supports: Workflow and stated scope

  2. Repository licence ↗Checked

    https://github.com/grafana/skills/blob/1ccacf29049fde66637fe01ab93a83a772da323b/LICENSE

    Supports: Redistribution terms

  3. https://raw.githubusercontent.com/grafana/skills/1ccacf29049fde66637fe01ab93a83a772da323b/skills/grafana-cloud/synthetic-monitoring-checks/SKILL.md

    Supports: review.findings, bestFor, upstreamDescription

Evidence & Ask