Skip to content
cccskills
FreePublished on GitHub

monitor-iot-edge

Monitor reviewed Grafana telemetry and freshness evidence for IoT devices and edge fleets. Use for smart devices, home automation sensors, remote equipment, wind turbines, space hardware, device-fleet freshness, missing telemetry, and bounded edge-health comparisons.

The English summary is being prepared. Showing the original description.

Run score Not tested

See how to install

Files included (2)

  • SKILL.md3.5 KB
  • agents/openai.yaml289 B

SKILL.md (original)

Review the instructions the agent will receive before you install.

Monitor IoT and Edge

Run the fixed grafana.iot-edge pack for an enrolled device or fleet scope and preserve telemetry gaps explicitly.

Safety contract

  • Use only grafana_profile_status, grafana_list_packs, grafana_run_pack, grafana_get_evidence, and grafana_resolve_link.
  • Accept only profile and scope aliases; one declared LIVE, REPLAY, or SIMULATED origin; bounded current and baseline windows; and opaque references returned by these tools.
  • Never accept or construct provider queries, SQL, URLs, headers, datasource IDs, device identifiers, panel JSON, variables, or raw telemetry requests.
  • Do not send device commands, change sampling, restart equipment, edit Grafana, or mutate alerts or datasources. This skill is observational only.
  • Confirm LIVE access. Never combine LIVE, REPLAY, and SIMULATED telemetry into one fleet claim.

Workflow

  1. Establish profileAlias, an enrolled device or fleet scopeAlias, origin, and a current window no longer than seven days. Use an equal-duration, non-overlapping baseline only when change analysis is needed.
  2. Call grafana_profile_status({profileAlias}). Stop on ERROR or blocked readiness; disclose PARTIAL readiness.
  3. Call grafana_list_packs({profileAlias}) and require the exact active pack ID grafana.iot-edge.
  4. Create one non-secret idempotency key and reuse it only for an exact retry.
  5. Call grafana_run_pack with that key, the aliases, packId: "grafana.iot-edge", origin, current window, and optional baseline window.
  6. Review telemetry first and device freshness second. Preserve returned timestamps, units, severity, direction, nulls, and fleet/device distinctions.
  7. Call grafana_get_evidence({reportRef,evidenceRef}) only for report-owned references needed to support a material device or fleet claim.
  8. Call grafana_resolve_link({reportRef,linkRef}) only for a same-report link reference that helps a human inspect the enrolled Grafana view. Never accept a URL as input.

A stale device is not necessarily offline, unsafe, or failed. An empty telemetry result is an evidence gap, not a zero measurement. Treat a freshness issue as rule-backed only when it has classification: SIGNED_RULE; treat classification: DEVIATION only as a deviation.

Interpret report status exactly

  • COMPLETE: expected telemetry and freshness evidence completed; evaluate the returned device and fleet signals.
  • EMPTY: no usable telemetry was returned; do not substitute zero values or declare devices healthy.
  • STALE: the latest evidence exceeds the reviewed freshness bound; state the age and affected scope.
  • PARTIAL: some devices, signals, or expected evidence are missing, failed, or mixed; avoid a fleet-wide conclusion.
  • CONTRADICTORY: telemetry conflicts; preserve the evidence references and withhold a single state claim.
  • ERROR: collection failed or was rejected; report the sanitized error and stop.

Completion format

Return scope, origin, windows, overall status, telemetry summary and freshness by returned scope with returned severity and direction, issues with their returned classification, gaps and contradictions, evidence references for material claims, and only successfully resolved report-owned links.

Reviews

No reviews yet. Tell others what you thought.

Comments

No login needed. Share how it worked for you, ask a question, or note where it did not run.

No comments yet. Be the first to write one.

Comments are published right away. Do not include personal information such as email addresses or phone numbers. Up to 2 URLs.

0 / 1,000 characters

This form is protected by reCAPTCHA, and the Google Privacy Policy and Terms of Service apply.

More skills from this repository

Analyze reviewed Grafana log evidence for event patterns, error-rate changes, and enrolled security markers. Use when a user needs bounded log analytics for an incident window, error-log comparison, anomaly triage, operational debugging, or a cautious check of approved security signals.

The English summary is being prepared. Showing the original description.

openai/community-plugins17Updated Oct 20, 2026

Run a read-only App Store-readiness preflight for a declared iOS target without claiming approval or submitting anything.

The English summary is being prepared. Showing the original description.

openai/community-plugins17Updated Oct 20, 2026

Audit whether a declared local review evaluation can support fair quality claims. Use only when the user explicitly asks to audit review-eval validity or use the bundled synthetic fixture.

The English summary is being prepared. Showing the original description.

openai/community-plugins17Updated Oct 20, 2026

Deterministically score comparable imported review-run lanes with root-cause metrics and uncertainty. Use only when the user explicitly asks to benchmark normalized local review runs or use the bundled synthetic fixture.

The English summary is being prepared. Showing the original description.

openai/community-plugins17Updated Oct 20, 2026

Stage and verify Fusion exports, drawing PDFs, flat-pattern DXF, viewport images and local renders. Use for engineering deliverables, not automatic publication or release.

The English summary is being prepared. Showing the original description.

openai/community-plugins17Updated Oct 20, 2026

Prepare, execute and verify authorized Fusion CAD changes using resolved entities, explicit units and state-bound plans. Use for reviewed parameter, sketch, feature, assembly or configuration operations.

The English summary is being prepared. Showing the original description.

openai/community-plugins17Updated Oct 20, 2026

See all skills by openai

Report a problem with this skill