skills.rest
amplitude
amplitudeOFFICIAL
Explore all skills by this author →
WebsiteGitHub

monitor-reliability

Generates reliability health reports from Amplitude auto-captured network, error, and error-click events.

Updated Apr 3, 2026
Search Tags:#analytics#release validation#error tracking#amplitude#kpi report#reliability monitoring#page health
One-click install
npx skills add https://github.com/amplitude/amplitude-copilot-plugin --skill monitor-reliability-amplitude
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: monitor-reliability
Source: https://github.com/amplitude/amplitude-copilot-plugin/tree/main/skills/plugins/amplitude/monitor-reliability
Command: npx skills add https://github.com/amplitude/amplitude-copilot-plugin --skill monitor-reliability-amplitude

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Product teams often learn about broken releases and rising error rates only after users complain. This Skill proactively computes reliability KPIs from Amplitude's auto-captured network request, JavaScript error, and error click data, so you know whether the product is healthy, degrading, or broken before support tickets arrive. ## Core Features & Use Cases - Reliability KPI Report: Computes network failure rate, slow request rate, JS error rate, error-free session rate, and error click rate, comparing a 7-day baseline against the current 7-day window. - Page Health Scoring: Grades individual pages as Healthy, Degraded, Unhealthy, or Critical based on error signals and affected users, surfacing the worst offenders. - Release Validation: Compares pre-deploy and post-deploy metrics to classify a release as clean, minor regression, significant regression, or rollback candidate. - Use Case: After shipping v4.3, ask "did the release break anything?" and receive a verdict listing new errors, affected endpoints, and user impact with recommended fixes. ## Quick Start Ask your AI assistant to run a reliability check on your Amplitude project for the last 14 days.

Frequently Asked Questions about monitor-reliability

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I check if a release broke anything in Amplitude?▼

Provide the deploy date or let the skill find it via get_deployments, then it compares pre-deploy and post-deploy windows for network failures, JS errors, and error clicks. It classifies the release as clean, minor regression, significant regression, or rollback candidate.

What reliability metrics can I track with Amplitude autocapture data?▼

You can track network failure rate (4xx/5xx), slow request rate over 3000ms, JS error rate, error-free session rate, error click rate, and affected user counts. All come from the three auto-captured events: Network Request, Error Logged, and Error Click.

Does this work without Session Replay or autocapture enabled?▼

No. The skill depends on Amplitude's auto-captured network, error, and error click events. If they are missing, it reports that autocapture is not enabled and describes what data enabling it would provide.

Why does my reliability report show misleading error spikes?▼

Common causes include partial-day data, weekend versus weekday traffic differences, bot traffic hitting API endpoints, and expected errors like 401s on auth flows. The skill validates against these artifacts before flagging regressions.

When should I use diagnose-errors instead of a reliability check?▼

Use the reliability monitor for proactive health checks when you do not know whether anything is wrong. Use the diagnose-errors skill for reactive investigation of a specific known issue or error you have already identified.

Related Skills

Explore More Data & Analytics Skills→
amplitudeamplitude
28

integration-javascript_web

Integrate Amplitude analytics into client-side JavaScript web applications.

Official
Intermediate
PostHogPostHog
713

triaging-error-issues

Ranks PostHog error tracking issues by users affected and recency for daily triage reviews.

Official
Intermediate
delta-and-betadelta-and-beta

tester-experiments

Run Braze A/B and feature flag experiments using random_bucket_number segments.

Community
Advanced
bobmatnycbobmatnyc
68

datadog-observability

Integrate Datadog APM, logs, metrics, synthetics, and RUM for full-stack observability.

Community
Advanced