skills.rest
ruvnet
ruvnetCOMMUNITY
Explore all skills by this author →
Website𝕏@ruvGitHub

observe-metrics

Aggregates system metrics from memory namespaces and flags anomalies against baselines.

70.1k|8.4k|Updated Jun 2, 2025
Search Tags:#metrics#agent swarm#anomaly detection#health score#Monitoring#Observability#memory namespace
One-click install
npx skills add https://github.com/ruvnet/claude-flow --skill observe-metrics
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: observe-metrics
Source: https://github.com/ruvnet/claude-flow/tree/main/plugins/ruflo-observability/skills/observe-metrics
Command: npx skills add https://github.com/ruvnet/claude-flow --skill observe-metrics

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Monitoring an agent swarm's health requires manually querying scattered metric records and judging whether values are abnormal. This Skill aggregates counters, gauges, and histograms from the observability namespace and automatically flags deviations from established baselines.

Core Features & Use Cases

  • Metric Aggregation: Sums counters (tasks completed, errors, token usage), reads gauges (active agents, memory usage), and computes p50/p95/p99 percentiles for duration histograms.
  • Anomaly Detection: Compares current values against baselines retrieved via pattern search and flags metrics deviating more than two standard deviations, with direction and severity.
  • Health Reporting: Produces a per-metric report with trend and anomaly flags plus an overall green/yellow/red health score.
  • Use Case: After running a multi-agent swarm for an hour, invoke this Skill to check task completion rates, error rates, and token consumption, and immediately see whether any metric has degraded versus its baseline.

Quick Start

Ask the agent to observe system metrics for the last hour and report any anomalies in swarm performance.

Frequently Asked Questions about observe-metrics

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

FAQPage Schema
How do I monitor agent swarm performance metrics?▼

Invoke the observe-metrics skill with an optional period argument such as --period 1h. It retrieves metric records from the observability namespace, aggregates counters, gauges, and histograms, and reports an overall health score with anomaly flags.

How does anomaly detection work for system metrics?▼

The skill fetches baseline values for each metric using agentdb pattern search, then flags any metric deviating more than two standard deviations from its baseline. Each anomaly is marked with direction (above or below) and severity.

Which memory tools should I use for namespace-routed metric queries?▼

Use the memory_search and memory_list tools with the observability namespace, since the memory_* tool family routes by namespace. The agentdb pattern tools do not accept a namespace argument and should only be used for baseline pattern storage and retrieval.

Can I query observability metrics from the command line?▼

Yes, run npx @claude-flow/cli@latest memory search with a query string and the observability namespace flag. This provides a CLI alternative to the agent-driven workflow for fetching metric records.

What metrics does the observability skill track?▼

It tracks counters like tasks completed, errors, and token usage; gauges like active agents and memory usage bytes; and histograms like task and span duration in milliseconds, computing p50, p95, and p99 percentiles.

Related Skills

Explore More Software Engineering Skills→
CleanExpoCleanExpo
1

metrics-collector

Codify database-backed metrics instrumentation for NodeJS-Starter-V1 using Supabase/PostgreSQL.

Community
Advanced
TzeusyTzeusy

trend-interpreter

Analyze health measurement trends and flag anomalies using statistical guidelines.

Community
Advanced
jamesaphoenixjamesaphoenix
1

debug-from-telemetry

Correlate traces, logs, and metrics to diagnose distributed failures.

Community
Advanced
taiyousan15taiyousan15

phase2-monitoring

Set up Prometheus, Grafana, Loki, and Alertmanager for system monitoring and log aggregation.

Community
Advanced