PlugWiki

service-mesh-observability Skill for Claude: What It Does and How to Install It

Makes Claude set up monitoring for a service mesh: metrics, tracing, dashboards, Kiali, OpenTelemetry and alert rules.

Installs
10.5K
on skills.sh
Stars
42.1K
on GitHub
Files
1
besides SKILL.md
Updated
Oct 2026
last repo commit

This is an observability guide from wshobson's agents collection. SKILL.md covers the three pillars (metrics, traces, logs) and a table of four golden signals with alert thresholds.

The seven templates are in references/details.md: Prometheus and Grafana for Istio, key metric queries, Jaeger tracing, the Linkerd viz dashboard, a Grafana dashboard JSON, Kiali and OpenTelemetry. The same file ends with sample alerting rules.

Good for

  • SREs adding monitoring to an Istio or Linkerd mesh
  • Engineers chasing latency and error problems between services
  • Teams defining SLOs for service communication
  • Platform teams that need a service dependency map

Install service-mesh-observability

Claude Code

  1. 1

    Copy this prompt.

    Install the service-mesh-observability skill in Claude Code. Run this command:

/plugin marketplace add wshobson/agents
/plugin install cloud-infrastructure@claude-code-workflows

Then check that it loaded and tell me how to use it.
  2. 2

    Paste it into Claude Code. It shows the change and asks before running it.

Pick one route. The plugin routes install the whole cloud-infrastructure plugin: 7 agents, 0 commands and 8 skills (counted from the plugin folder in the repo tree, which has no hooks or MCP config files). The skills-only routes install just this skill, with no agents or commands.

Built from the package data we hold. Not yet tested by us. Source.

What's in the skill

When it loads

Implement comprehensive observability for service meshes including distributed tracing, metrics, and visualization. Use when setting up mesh monitoring, debugging latency issues, or implementing SLOs for service communication.

Sections in SKILL.md

  1. When to Use This Skill
  2. Core Concepts
  3. 1. Three Pillars of Observability
  4. 2. Golden Signals for Mesh
  5. Templates and detailed worked examples
  6. Best Practices
  7. Do's
  8. Don'ts

Key facts

Author
wshobson (repo owner)
License
MIT
Installs on skills.sh
10,510 (Nov 2, 2026)
GitHub stars
42,081 for the whole wshobson/agents repo (Nov 2, 2026)
Last repo commit
Oct 30, 2026
What it includes
2 files including SKILL.md: references/details.md (7 templates and alerting rules). No scripts
skills.sh audits
Gen Agent Trust Hub: pass; Socket: warn; Snyk: pass (Nov 2, 2026)
Part of
The cloud-infrastructure plugin (version 1.3.3) in wshobson/agents. Counted from the plugin folder: 7 agents, 0 commands and 8 skills
Works in
Claude Code, Codex CLI, Cursor, OpenCode, Antigravity CLI, GitHub Copilot and Pi, per the README. Skills-only installers ask which tool you use

What it includes

  • Metrics, traces and logs; latency, traffic, errors and saturation with alert levels

    SKILL.md lists P99 above 500ms, error rate above 1% and saturation above 80%.

    Three pillars and golden signals

  • 5 do's and 4 don'ts, such as sampling 100% in dev and 1-10% in prod

    Best practices

  • Prometheus scrape config for Istio and a ServiceMonitor

    Write · Template 1: Prometheus and Grafana

  • Request rate, 5xx rate, P99 latency, TCP connections and request size

    Read · Template 2: Key Istio metric queries

  • Turn on tracing in Istio and deploy Jaeger

    Write · Template 3: Jaeger tracing

  • Install the viz extension and use top, routes, tap and edges

    Uses linkerd viz install piped to kubectl apply.

    Runs code · Template 4: Linkerd viz

  • A Grafana dashboard written as JSON

    Write · Template 5: Grafana dashboard JSON

  • Service graph with Prometheus, tracing and Grafana links

    Write · Template 6: Kiali

  • Collector config and Istio telemetry settings

    Write · Template 7: OpenTelemetry

  • Prometheus rules for error rate, P99 latency and expiring mesh certificates

    Write · Alerting rules

From the official docs

Permissions

Low risk

What it can do

  • Read your data
  • Create and change things
  • Delete things
  • Read-only mode

Watch out for

  • The Kiali template sets anonymous login

    Template 6 has auth strategy: anonymous, with openid and token as the alternatives in a comment. Pick one of those before anyone outside your team can reach Kiali. github.com

  • Tracing at 100% sampling costs storage

    The Jaeger template samples 100% of requests, with a comment to use less in production. SKILL.md says 1-10% in prod. github.com

  • Linkerd viz install changes your cluster

    Template 4 pipes linkerd viz install into kubectl apply. Check your kubectl context first. github.com

Security advisories

No published security advisories as of Nov 2, 2026.

Technical details
Turning on read-onlyNo read-only mode. Use version control and review commands before they run.
Sign-in and keysSKILL.md and details.md do not mention any account or key.
Scopes

  • Claude can write Prometheus, Jaeger, Kiali, OpenTelemetry and alert rule files in your project

  • Template 4 runs linkerd viz install piped to kubectl apply

  • The Kiali and Jaeger templates connect to Prometheus, Grafana and Jaeger addresses inside istio-system

Questions

How do I install service-mesh-observability?

Use the install box at the top of this page. It offers the cloud-infrastructure plugin and two skills-only commands. Pick a skills-only command if you want just this skill without the plugin's other parts.

What does the service-mesh-observability skill do?

It gives Claude patterns for monitoring a service mesh. Claude uses them to write Prometheus, Jaeger, Grafana, Kiali and OpenTelemetry config, and to build alert rules for latency and errors.

Does it work with Linkerd as well as Istio?

Yes. SKILL.md describes itself as a guide for Istio, Linkerd and service mesh deployments, and the details file has a Linkerd viz template next to the Istio ones.

Which alert thresholds does it use?

The two files differ. SKILL.md's golden signals table says P99 above 500ms and errors above 1%. The sample alert rules in the details file fire at a 5% error rate and a P99 above 1000ms. Pick your own values.

Should I install the whole cloud-infrastructure plugin?

Only if you want the rest. The plugin folder holds 7 agents, 0 commands and 8 skills, counted from the repo tree, and the tree shows no hooks or MCP config files in it. The plugin's other skills include istio-traffic-management, linkerd-patterns and mtls-configuration.

Is service-mesh-observability safe to install?

The folder holds SKILL.md and one reference file, with no scripts. Claude writes monitoring YAML and may suggest the linkerd and kubectl commands in the file, so check your kubectl context before you approve them. The Kiali template sets anonymous authentication, so change that before you expose it.

Does service-mesh-observability work in Codex, Cursor or the Claude app?

The README documents Claude Code and Codex CLI plugin routes, and skills-only installs for the tools it lists: Cursor, OpenCode, Antigravity CLI, GitHub Copilot and Pi. We left the Claude app upload out of the install box because this skill's commands run in a terminal.

Information

updated Nov 2, 2026

Repository
wshobson/agents
Folder
plugins/cloud-infrastructure/skills/service-mesh-observability

wshobson/agentsskills.sh

Sources

Docs, issue trackers and discussions this page is built from.

Show all 3
  1. service-mesh-observability SKILL.md · github.com
  2. service-mesh-observability references/details.md · github.com
  3. wshobson/agents README · github.com

Updated Nov 2, 2026.