monitoring-observability

Set up monitoring, logging, and observability for applications and infrastructure. Use when implementing health checks, metrics collection, log aggregation, or alerting systems. Handles Prometheus, Grafana, ELK Stack, Datadog, and monitoring best practices.

Install with Tessl CLI

npx tessl i github:supercent-io/skills-template --skill monitoring-observability

What are skills?

1.14x

Quality

82%

Does it follow best practices?

Impact

94%

1.14x

Average score across 3 eval scenarios

SKILL.md

Review

Evals

Evaluation results

98%

-2%

Add Observability to an Order Processing Service

Prometheus metrics instrumentation

Criteria

Without context

With context

prom-client library

100%

collectDefaultMetrics called

100%

Histogram metric name

100%

Counter metric name

100%

Correct label set

100%

/metrics endpoint

100%

Request tracking middleware

100%

Prometheus scrape_interval

100%

metrics_path config

100%

node-exporter job

100%

Alertmanager config

100%

No high-cardinality labels

100%

80%

Without context: $0.3322 · 1m 17s · 20 turns · 27 in / 4,943 out tokens

With context: $0.5470 · 1m 41s · 25 turns · 31 in / 5,902 out tokens

100%

Improve Log Quality for a User Authentication Service

Structured logging with Winston

Criteria

Without context

With context

Winston library used

100%

JSON log format

100%

Timestamp in format

100%

Error stack trace

100%

defaultMeta service field

100%

defaultMeta environment field

100%

LOG_LEVEL env var

100%

Error log file path

62%

100%

Combined log file path

25%

100%

Console colorized format

100%

No sensitive data logged

100%

Structured log fields

100%

Without context: $0.2623 · 1m 17s · 17 turns · 19 in / 4,391 out tokens

With context: $0.4878 · 2m 26s · 25 turns · 30 in / 5,132 out tokens

86%

29%

Production Readiness: Health Checks and Alerting for a Payment API

Health checks and alert rules

Criteria

Without context

With context

Health endpoint path

100%

Three-state status

37%

100%

503 for unhealthy

70%

100%

Database check with latency

100%

Redis check with latency

100%

Timestamp and uptime

50%

100%

Alert group interval

100%

HighErrorRate alert threshold

100%

SlowResponseTime alert

50%

HighMemoryUsage alert

Alert severity labels

87%

75%

Runbook present

100%

Focused alert set

100%

Without context: $0.4147 · 1m 45s · 23 turns · 28 in / 6,132 out tokens

With context: $0.4808 · 1m 42s · 22 turns · 209 in / 6,023 out tokens

Evaluated: 3 days ago
Agent: Claude Code
Model: Claude Sonnet 4.6

Table of Contents

Add Observability to an Order Processing Service Improve Log Quality for a User Authentication Service Production Readiness: Health Checks and Alerting for a Payment API

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.