skills.rest
opensearch-project
opensearch-projectOFFICIAL
Explore all skills by this author →
WebsiteGitHub

apm-red

Compute RED metrics from Prometheus and OpenSearch trace spans.

34|32|Updated Jan 28, 2026
Search Tags:#monitoring#observability#traces#promql#apm#ppl#red
One-click install
npx skills add https://github.com/opensearch-project/observability-stack --skill apm-red
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: apm-red
Source: https://github.com/opensearch-project/observability-stack/tree/main/claude-code-observability-plugin/skills/apm-red
Command: npx skills add https://github.com/opensearch-project/observability-stack --skill apm-red

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Monitors service health by providing RED (Rate, Errors, Duration) metrics through PromQL queries against Prometheus and PPL queries against OpenSearch trace spans as an alternative.

Core Features & Use Cases

  • Per-service and per-endpoint rate calculations to gauge traffic and capacity needs.
  • Error rate and error counts derived from HTTP spans and logs to identify fragile components.
  • Latency distribution queries using PromQL and PPL-based span data to surface p50, p95, and p99 timings.

Quick Start

Run the included PromQL and PPL templates against your stack to surface per-service RED metrics.

Frequently Asked Questions about apm-red

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

FAQPage Schema
How do I compute RED metrics for microservices using Prometheus and OpenSearch?▼

Compute RED metrics by running included PromQL templates against Prometheus and PPL queries against OpenSearch trace spans to calculate per-service rate, error rate, and latency duration.

What is the best way to monitor service latency distribution with PromQL and PPL?▼

Surface p50, p95, and p99 latency timings by executing the provided PromQL and PPL-based span data templates to query latency distribution across your microservices and endpoints.

Can I calculate error rates from HTTP spans for continuous production monitoring?▼

Error rate and error counts can be derived from HTTP spans and logs to identify fragile components across your production stack for continuous monitoring.

How do I measure per-endpoint traffic and capacity needs with RED metrics?▼

Per-service and per-endpoint rate calculations gauge traffic and capacity needs by applying standardized querying templates to your Prometheus and OpenSearch environments.

Does this approach require specific credentials for querying OpenSearch trace spans?▼

Querying OpenSearch trace spans with PPL templates requires configurable endpoints and credentials set up in your environment to execute the standardized RED metrics queries.

When should I use PPL queries instead of PromQL for service health monitoring?▼

PPL queries against OpenSearch trace spans serve as an alternative to PromQL against Prometheus when you need to identify service health gaps in microservices and endpoints.

Related Skills

Explore More Software Engineering Skills→
opensearch-projectopensearch-project
34

traces

Query OpenSearch trace data with PPL to analyze agent invocations and errors.

Official
Intermediate
vecearvecear

backend-latency-profiler-helper

Profile REST or gRPC endpoints to detect latency hotspots and generate a fix roadmap.

Community
Intermediate
wfukatsuwfukatsu
12

design-observability

Design observability systems with SLI/SLOs, distributed tracing, and log aggregation.

Community
Advanced
hung-phanhung-phan

use-red-methods

Diagnose distributed system performance issues using USE and RED methodologies.

Community
Advanced