betterstack-uptime-triage

name	betterstack-uptime-triage
description	BetterStack Uptime incident and monitor triage workflow. Use when asked to inspect BetterStack Uptime status, summarize current incidents, identify likely impact, or propose immediate mitigation and follow-up actions based on Uptime API data.

BetterStack Uptime Triage

Overview

Use this skill to fetch BetterStack Uptime monitor and incident data, classify impact, and provide operator-ready triage output.

Execute Workflow

1. Verify required credentials:

• BETTERSTACK_API_TOKEN
• BETTERSTACK_API_BASE_URL (default https://uptime.betterstack.com/api/v2 if missing)

2. Capture current Uptime signals:

• Use scripts/uptime_snapshot.sh for a standard monitor+incident pull.
• Use scripts/uptime_get.sh for targeted endpoint pulls.
• Use scripts/uptime_incident_names.sh for incident-name summaries.

3. Summarize incident state:

• active incidents
• affected monitors/components
• customer/reliability impact
• start time and current status

4. Classify output explicitly:

• facts (API-confirmed)
• hypotheses (likely causes)
• unknowns (data still needed)

5. Recommend actions:

• immediate containment/mitigation
• verification steps
• durable follow-up actions
• owner + ETA proposal

Scripts

scripts/uptime_snapshot.sh

Run a standard snapshot and store JSON files for current incidents and monitors.

bash scripts/uptime_snapshot.sh

scripts/uptime_get.sh

Fetch any Uptime API endpoint path quickly.

bash scripts/uptime_get.sh /incidents
bash scripts/uptime_get.sh /monitors

scripts/uptime_incident_names.sh

Return incident names as a numbered list, without jq.

bash scripts/uptime_incident_names.sh 5

Critical Rules

• Do not use jq (it is unavailable in this runtime).
• Do not call /api/v1/* endpoints.
• Do not append /api/v2 manually if BETTERSTACK_API_BASE_URL already includes it.
• Prefer the provided scripts over ad-hoc curl commands.

Output Contract

Return triage output in this order:

1. Current status
2. Impact summary
3. Findings (fact/hypothesis/unknown)
4. Immediate actions
5. Follow-up plan

Guardrails

• Do not claim incident resolution without verification evidence.
• Do not execute production changes without explicit operator approval.
• Keep recommendations reversible first, invasive changes second.

References

• BetterStack Uptime API quick start and endpoint patterns: references/uptime-api.md