agentleFS
itentialitential/infra-agent-skills/skills/alert-monitor-management/SKILL.md

alert-monitor-management

Skill1 starsChanged 33 days agoApache-2.0View on GitHub

Create a monitor or check that watches a metric/resource for a threshold breach, and acknowledge or resolve an alert it raises, on a supported observability platform — Datadog, New Relic, LogicMonitor, Paessler PRTG, Centreon, OP5 Monitor, or Thruk. Use when a caller needs to stand up new alerting coverage for a resource, silence/own an in-progress alert, or confirm an alert has cleared. Assumes an authenticated API client for the target platform (API key/app key pair, OAuth bearer token, or username/passhash, per platform) is already available; this skill does not provision or authenticate that client. Does not cover notification-channel/integration setup (Slack, PagerDuty, email routing), dashboard creation, log-pipeline configuration, or synthetic/browser-check authoring beyond what a platform's own monitor-creation call requires.

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 3 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Not for configuring where notifications get delivered (notification channels, escalation targets outside the alert-rule fields a platform's …

    line 20
  2. Distinctive:

    Does not cover notification-channel or escalation-chain setup as a standalone task — only the fields a platform's own monitor/alert-rule …

    line 71
  3. Distinctive:

    Does not cover dashboards, log-pipeline/processing configuration, or synthetic/browser-check script authoring, even on platforms where …

    line 75

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Alert & Monitor Management

Creates a monitor/check that evaluates a condition against collected data, and acknowledges or resolves an alert raised by one, on one of the supported observability platforms, following that platform's own documented API and alert lifecycle.

When to use this skill

  • A monitor, check, alert condition, or alert rule needs to be created on Datadog, New Relic, LogicMonitor, Paessler PRTG, Centreon, OP5 Monitor, or Thruk so that a metric, log signal, sensor, or resource starts producing alerts when a threshold is breached.
  • An already-raised alert on one of these platforms needs to be acknowledged (claimed/silenced while work is ongoing) or resolved/closed.
  • Not for configuring where notifications get delivered (notification channels, escalation targets outside the alert-rule fields a platform's own create-monitor call requires), building dashboards, or authoring synthetic/browser test scripts — those are separate concerns even when the same platform's UI bundles them nearby.

Universal concepts (apply regardless of platform)

  • A monitor/check is a standing evaluation, not a one-time test. Every platform here re-runs its configured condition against fresh data on an ongoing basis; creating one does not itself raise an alert — an alert only appears once the condition's threshold is breached on a later evaluation.
  • Alerts trend toward auto-clearing, not manual closure, everywhere in scope. Where a platform documents it, an active alert closes on its own once the signal that caused it returns to a non-violating state for the condition's own configured recovery window (Datadog monitors return to OK; New Relic NRQL violations close after the signal holds below threshold for the recovery period; LogicMonitor alerts clear once the datapoint fails the threshold for its configured clear interval; PRTG sensors leave the Down/alarm state once the channel is back within limits). Some platforms document no call that forces a problem out of its non-OK state, so a later healthy result is the only path. Do not assume a platform requires — or even offers — a manual "resolve" call as the normal path; check that platform's own procedure.md for whether one exists and what it actually does.
  • Acknowledging is ownership/notification control, not resolution. On every platform in scope, acknowledging an alert marks it as being worked and typically suppresses further repeat notifications for that occurrence, but it does not change the underlying measured condition and does not by itself clear the alert — the alert still clears (or doesn't) based on the data, exactly as the point above describes.
  • Creating a monitor and routing its notifications are frequently two separate configuration objects. Where a platform splits "what condition triggers an alert" from "who gets notified and how" (most visibly LogicMonitor's DataSource-datapoint-threshold vs. Alert Rule split), that split is platform-specific, not something this skill flattens away — follow the platform's own procedure.md for which object owns which part.
  • This capability has no single wire-level standard grounding it (unlike, say, DNS with RFC 1035) — each platform below defines its own monitor/alert model from its own API documentation; treat any claim above as accurate only to the extent each vendor's own procedure.md confirms it holds for that platform.

Procedure

  1. Identify the target platform — from the caller's request or the platform argument. If it isn't one of the platforms in references/INDEX.md, stop: this skill doesn't cover it, and does not improvise a generic procedure for an unsupported platform.
  2. Open references/<platform>/procedure.md and follow it exactly — it contains that platform's full operational procedure for creating a monitor/check and for acknowledging/resolving an alert, including its own error handling, patterns, known limitations, tools, and verification checklist. This file is intentionally the router; the real steps live per platform because the platforms genuinely do not share one alerting model (see references/INDEX.md for the shape differences, e.g. LogicMonitor's DataSource/AlertRule split vs. the other platforms' single monitor/check object).
  3. Report the outcome using that platform procedure's own status vocabulary (e.g. CREATED, ACKNOWLEDGED, RESOLVED, REJECTED_*, FAILED) — don't normalize it away.

Known limitations

  • Does not cover notification-channel or escalation-chain setup as a standalone task — only the fields a platform's own monitor/alert-rule creation call requires inline (e.g. an escalation chain ID reference). Building a new notification channel or escalation policy from scratch is out of scope on every platform here.
  • Does not cover dashboards, log-pipeline/processing configuration, or synthetic/browser-check script authoring, even on platforms where those live in the same product area as monitors.
  • Platform coverage is exactly the seven vendors in references/INDEX.md — a platform not listed there needs a new references/<platform>/procedure.md before this skill can be used for it.
  • Where a platform's own documentation does not state a manual "resolve" action exists (or states explicitly that creating new monitor objects outright isn't supported via API), that platform's procedure.md says so plainly rather than this skill inventing one — see each platform file's own Known Limitations.

Contributing to this skill

SKILL.md itself should only ever change for two reasons: a bug in the universal concepts/router logic above, or adding a pointer to a newly-supported platform. Platform-specific procedure changes — a new error code, a corrected field name, a changed API path, a changed alert-lifecycle detail — belong in that platform's own references/<platform>/procedure.md and endpoints.md, never here. If you're about to add vendor-specific detail to this file, put it in the platform's reference file instead.

Tools

Each platform assumes its own authenticated client (REST API, GraphQL/NerdGraph API, PRTG's HTTP/REST v2 API, or an external-command API) — see references/INDEX.md for which, and that platform's procedure.md for the specific operations/mechanisms.

Verification checklist

  • Target platform confirmed against references/INDEX.md before proceeding
  • That platform's own procedure.md was followed, including its own confirmation gates and error handling — not a generic/improvised sequence
  • Outcome reported using the platform procedure's own status vocabulary

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.